System

Abstract

A system is provided.SOLUTION: A system comprising: means for transmitting video data acquired from a drive recorder in real time; means for storing the received video data; means for preprocessing the stored video data; means for analyzing the preprocessed data and detecting a damaged portion of a road; means for classifying the detected damaged portion; means for generating a repair policy based on a classification result; and means for presenting the generated repair policy to a user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Modern road infrastructure is aging, requiring rapid and efficient maintenance. However, it is difficult to develop efficient road repair plans within the constraints of limited personnel and budgets. Conventional methods mainly involve periodic manual inspection of roads, identifying damaged areas, and formulating repair policies, which are time-consuming and cost-intensive. Furthermore, it is difficult to prioritize repairs and optimize costs, and it may not be possible to prevent the progression of deterioration. The present invention aims to solve these problems and provide a system that enables road managers to develop efficient and accurate road repair plans. [Means for solving the problem]

[0005] The present invention provides the following means: a system including means for transmitting video data acquired from a drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damage, means for generating a repair policy based on the classification results, and means for presenting the generated repair policy to a user. This allows for efficient analysis of data collected from a drive recorder and rapid development of a repair plan. Furthermore, by proposing an appropriate repair method based on the damage information obtained by the analysis and classification means, repair costs can be optimized. Furthermore, the accuracy of the analysis can be improved by using preprocessing means such as noise removal and resolution adjustment. A user interface allows users to easily access the information they need, supporting efficient road management.

[0006] A "dashcam" is a device installed in a vehicle that records video footage while driving.

[0007] "Video data" refers to video information of roads acquired by a drive recorder.

[0008] "Real time" means that processing and communication are carried out immediately without delay.

[0009] "Transmission" means sending data from one point to another.

[0010] "Receiving" means receiving transmitted data.

[0011] "Storage" means keeping data for a certain period of time.

[0012] "Preprocessing" refers to various filtering and adjustments that are performed on data before it is analyzed.

[0013] "Analysis" means examining data in detail and understanding its meaning and content.

[0014] "Broken area" refers to an area where the road surface is damaged.

[0015] "Detection" means finding an object based on certain conditions.

[0016] "Classification" refers to the division of detected objects into categories according to specific criteria.

[0017] A "repair policy" is a plan for what repairs and restorations will be made to damaged areas.

[0018] "User" refers to road administrators and related parties who use this system.

[0019] "Presentation" means showing information in a visible form.

[0020] A "system" is a set of mechanisms in which multiple devices and programs function in cooperation with each other. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7]FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0022] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0023] First, the terms used in the following description will be explained.

[0024] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0025] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0026] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0027] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0028] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0032] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0033] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0034] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0035] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0036] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0039] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0040] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0041] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0042] This invention provides a system that analyzes video data collected from a drive recorder and automatically identifies road damage and generates repair plans. The system's main components are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[0043] Program processing

[0044] 1. Image data collection

[0045] Device:

[0046] A drive recorder is installed in the vehicle and records video in real time as the vehicle travels along the road. The drive recorder acquires GPS information and adds location information and timestamps to the video. This video data is then sent to a server in real time via a 5G communication module.

[0047] 2. Receipt and storage of data

[0048] server:

[0049] The system receives video data sent from the device. The received data is verified for format and integrity, and if there are no problems, it is stored in a database. The stored data includes not only the video data itself, but also the corresponding location information and timestamp.

[0050] 3. Data Preprocessing

[0051] server:

[0052] A noise reduction filter is applied to the received video data. At this stage, the data is resized to a format that is optimal for the analysis model. Additionally, frames containing important information are extracted and unnecessary frames are removed. This preprocessing improves the accuracy and efficiency of the analysis.

[0053] 4. Analysis of image data

[0054] server:

[0055] The preprocessed data is then input into a generative AI model, which detects damage to the road surface and extracts attribute information such as its type, extent, and depth. The model is particularly effective at accurately detecting critical damage, such as large potholes and cracks.

[0056] 5. Damage Level Classification

[0057] server:

[0058] Based on the information of the detected damage, a classification algorithm is applied to assess the level of damage: the damage is classified as minor, caution, or severe, and stored in a database along with the corresponding location information.

[0059] 6. Repair Policy Generation

[0060] server:

[0061] Based on the classification results, the system automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs. This information allows road managers to quickly and accurately develop repair plans.

[0062] 7. Presentation of results

[0063] server:

[0064] An interface is generated to present the generated repair policy and information on damaged areas to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using a map display or list display.

[0065] User:

[0066] Through the provided interface, users (road administrators) can view detailed information about damaged areas and repair policies, which will enable them to prioritize repairs and develop specific repair plans.

[0067] Specific examples

[0068] For example, a vehicle traveling on a highway may capture a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10cm deep and 50cm wide. A classification algorithm then evaluates it as "severely damaged" and stores it in a database along with its location. Finally, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers via a user interface, allowing them to quickly develop specific repair plans.

[0069] In this way, the present invention provides a system that uses video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair policies, thereby reducing the burden on road administrators and realizing efficient road maintenance management.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] Device:

[0073] The dashcam is installed in vehicles traveling on the road and records video in real time. The built-in GPS sensor periodically acquires location information while recording, adding a timestamp and location information to the video data. The video data is then sent to a server in real time using a 5G communication module.

[0074] Step 2:

[0075] server:

[0076] Receives video data sent from the device. The received data is verified for format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, corresponding location information, and timestamp.

[0077] Step 3:

[0078] server:

[0079] The stored video data is preprocessed by applying a noise reduction filter, resizing the video resolution to a format suitable for analysis, and extracting important frames and removing unnecessary ones.

[0080] Step 4:

[0081] server:

[0082] The preprocessed data is input into a generative AI model, which detects road damage (potholes, cracks, etc.) from the collected video data. At this stage, attribute information such as the type, extent, depth, and location of the damage is extracted.

[0083] Step 5:

[0084] server:

[0085] The damage level is assessed based on the extracted information on the damaged areas. A unique classification algorithm is applied to categorize the damage into several levels, such as minor, caution, and serious. The classification results and location information are stored in a database.

[0086] Step 6:

[0087] server:

[0088] Based on the classification results, an algorithm is applied to automatically generate an optimal repair policy. The repair policy includes the urgency of the repair, the recommended repair method (e.g., partial patching, full repaving), and the estimated repair cost. The generated repair policy is then compiled in the form of a report.

[0089] Step 7:

[0090] server:

[0091] An interface is generated to provide the generated repair policy and related information to the user. The interface is designed for ease of use, including map and list views, allowing users to quickly check the information they need.

[0092] Step 8:

[0093] User:

[0094] Users (road administrators) access the server through a web interface or a dedicated application. On the dashboard, they can view detailed information, such as a map of damaged areas and a list of damage levels. Based on the repair policy provided, they can quickly create specific repair plans.

[0095] In this way, the system of the present invention helps road managers to plan road repairs efficiently and accurately. The automated process shortens the lead time for repairs and optimizes repair costs.

[0096] Example 1

[0097] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0098] In modern transportation infrastructure, it is important to quickly and accurately detect road damage and develop appropriate repair plans. However, conventional methods have the drawback of requiring a large amount of manpower, time, and cost. Furthermore, there is a risk of overlooking road damage and the difficulty of prioritizing repairs. The present invention aims to solve these problems and provide an efficient and accurate road management method.

[0099] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0100] In this invention, the server includes a device for transmitting video data acquired from the drive recorder in real time, a device for saving the received video data, a device for preprocessing the saved video data, a generative model for analyzing the preprocessed data and detecting road damage, an algorithm for classifying the detected damage, a method for generating a repair policy based on the classification results, and an interface for presenting the generated repair policy to the user. This enables quick and accurate detection of road damage and efficient planning of repairs.

[0101] A "drive recorder" is a device that is installed in a vehicle and records video data while the vehicle is driving.

[0102] "Video data" refers to data including images and video information recorded by a drive recorder.

[0103] A "real-time transmission device" is a communication device for instantly transferring acquired video data to a remote server.

[0104] The "receiving device" is a part of the server that receives video data transmitted from a remote location.

[0105] The "storing device" is a storage device that stores the received video data in a database or the like so that it can be used later.

[0106] A "preprocessing device" is a device that converts received video data into a format suitable for analysis. Specifically, it performs processes such as noise removal and resolution adjustment.

[0107] The "generative model for analysis" is an AI model that identifies damaged areas on roads based on preprocessed video data and extracts their detailed attribute information.

[0108] The "classification algorithm" is an algorithm for evaluating and classifying the level of damage based on information about the detected damaged portion.

[0109] The "method for generating a repair policy" is a process for determining the optimum repair method, urgency, and estimated repair cost for the classified damaged areas.

[0110] An "interface" is a system that visually displays the generated repair policy and information on damaged areas, allowing users to easily access it.

[0111] This invention relates to a system that analyzes video data collected from a drive recorder and automatically generates road damage information and repair plans. The main components of the system are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[0112] The drive recorder is installed in the vehicle and records video in real time while driving. The drive recorder acquires GPS information and adds location information and timestamps to the video. Using a 5G communication module, this video data is sent to a server with low latency.

[0113] The server receives the video data and verifies the format and integrity of the received data. If there are no problems, the data is stored in a database. The stored data includes not only the video but also the corresponding location information and timestamp. Image processing libraries such as OpenCV are used to preprocess the video data, such as noise reduction and resizing.

[0114] The pre-processed data is then analyzed by a generative AI model, which uses deep learning frameworks such as TensorFlow or PyTorch to detect damage to the road surface and extract its type, extent, and depth, for example using a convolutional neural network (CNN).

[0115] The server then applies a classification algorithm based on the detected damage information, which assesses the level of damage based on parameters such as the area, depth, and width of the damage. The data is then classified into levels such as minor, caution, and severe and stored in a database.

[0116] Furthermore, the server automatically generates an optimal repair policy based on the classification results, which includes the urgency of repair, recommended repair methods, and estimated repair costs, enabling road managers to quickly develop repair plans.

[0117] The generated repair plan and damage location information are presented to the user through an interface that can be used in a web browser or mobile application, and the information is presented in an easy-to-understand manner using map and list views. Front-end frameworks such as React and Angular are used, and Node.js and Django are used for the back-end.

[0118] As a concrete example, a vehicle traveling on a highway captures a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10 cm deep and 50 cm wide. A classification algorithm evaluates it as "severely damaged" and stores it in a database along with its location information. Ultimately, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers through a user interface, allowing them to quickly develop specific repair plans.

[0119] An example of a prompt to be input into the generative AI model is, "Analyze the video data acquired while the vehicle is traveling and detect any newly occurring potholes or cracks. Also, measure the depth and width of the damaged areas and propose a repair plan."

[0120] In this way, the present invention provides a system that utilizes video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair plans, thereby achieving more efficient road management and reducing costs.

[0121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0122] Step 1: Collecting image data

[0123] Device:

[0124] The dashcam is installed in a vehicle and records high-resolution video in real time as the vehicle travels along the road. The dashcam acquires GPS information and adds location information and timestamps to the video data. Using a 5G communication module, this video data is transmitted to a server with low latency.

[0125] Input: Real-time video and GPS information acquired while the vehicle is running

[0126] Data processing: Add location information and timestamp to each video frame

[0127] Output: Video data with additional information

[0128] Step 2: Receiving and saving video data

[0129] server:

[0130] The server receives the video data sent from the device. The format and integrity of the received data are verified, and if there are no problems, it is stored in a database. The stored data includes not only the video, but also the corresponding location information and timestamp.

[0131] Input: Video data with additional information sent from the device

[0132] Data processing: verifying data format and integrity (e.g., CRC check, format verification)

[0133] Output: Verified video data is stored in a database

[0134] Step 3: Preprocessing the data

[0135] server:

[0136] The server applies a noise reduction filter to the received video data. This process resizes the video to a format that is optimized for the analysis model. It also extracts frames that contain important information for analysis and removes unnecessary frames.

[0137] Input: Verified video data stored in a database

[0138] Data processing: noise removal (e.g., Gaussian filter application), resolution adjustment, extraction of important frames, deletion of unnecessary frames

[0139] Output: Pre-processed video data

[0140] Step 4: Analyzing the image data

[0141] server:

[0142] The server inputs the preprocessed data into a generative AI model to detect damage on the road surface, which uses deep learning frameworks such as TensorFlow and PyTorch to extract the type, extent, and depth of the damage.

[0143] Input: Preprocessed video data

[0144] Data calculation: Deep learning model for detecting damage and extracting attribute information

[0145] Output: Specific attribute information and location information of detected damage points

[0146] Step 5: Classify the level of damage

[0147] server:

[0148] The server then applies a classification algorithm to assess the level of damage detected: damage is classified as minor, minor, or severe, and these classifications are stored in a database.

[0149] Input: Attribute information and location information of the detected damage

[0150] Data processing: Assessing the level of corruption with classification algorithms

[0151] Output: Damage level assessment data and location information are saved in a database.

[0152] Step 6: Generate repair policies

[0153] server:

[0154] Based on the classification results, the server automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs.

[0155] Input: Damage level assessment data and location information

[0156] Data calculation: Urgency assessment, repair method proposal, repair cost estimate

[0157] Output: Integrated information including repair policy

[0158] Step 7: Presenting the results

[0159] server:

[0160] The server generates an interface to present the generated repair policy and damage location information to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using map and list views.

[0161] User:

[0162] The user (road administrator) can view detailed information about the damaged areas and repair policies through the provided interface.

[0163] Input: Integrated information including repair policy

[0164] Data calculation: Visualization of information (e.g., displaying potholes on a map, displaying repair policies in list format)

[0165] Output: A user-visible interface

[0166] (Application example 1)

[0167] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0168] Conventional road management systems have difficulty quickly and accurately detecting road damage and providing appropriate repair plans. In particular, autonomous vehicles require real-time road condition monitoring to ensure safety, making efficient road management essential.

[0169] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0170] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damaged areas, means for generating a repair policy based on the classification results, means for presenting the generated repair policy to a user, means for displaying information on the damaged areas on a map, and means for notifying the user of repairs to the damaged areas in real time, thereby enabling real-time road condition monitoring and rapid planning of repairs.

[0171] A "dashcam" is a device installed in a vehicle that records video footage while driving.

[0172] "Means for transmitting in real time" refers to a communication means for transferring video data to a receiving destination such as a server immediately without delay.

[0173] "Means for storing received video data" refers to a storage device or database for storing the received video data in an appropriate format.

[0174] The "means for pre-processing stored video data" is a function for performing initial processing such as noise removal and format adjustment on video data.

[0175] The "means for analyzing the pre-processed data and detecting road damage" is an algorithm or software for identifying damaged portions of the road from the pre-processed data using image analysis techniques.

[0176] The "means for classifying detected damage areas" refers to a classification algorithm for classifying and rating damage areas based on the type and severity of damage.

[0177] The "means for generating repair policies based on classification results" refers to algorithms or systems that automatically plan and propose optimal repair methods based on classified damage location information.

[0178] The "means for presenting the generated repair policy to the user" refers to a display interface or notification system for providing the user with the repair plan and information on the damaged area.

[0179] The "means for displaying information on damaged areas on a map" is a function that visually shows identified damaged areas on a map using a geographic information system (GIS).

[0180] "Means for providing real-time notification of damaged area repairs" refers to a communication and notification system that immediately notifies of the need for repairs when a damaged area is detected.

[0181] This invention is a system that uses video data acquired from a vehicle's onboard dashcam to detect road damage in real time and automatically generate and present repair plans. The system's main hardware components are a dashcam, GPS module, and communication module, while the software components include a server, database, generative AI model, and user interface.

[0182] System Overview

[0183] The server receives, stores, pre-processes, analyzes, and detects damages in real time, classifies them, and generates repair plans. This data is then provided to the user through a user interface, including map displays and notifications.

[0184] Hardware and software used

[0185] Drive recorder: Acquires video data and GPS information.

[0186] GPS module: Adds location information and timestamps to video data.

[0187] 5G communication module: Transmits video data to a server in real time.

[0188] Server: Receives, stores, preprocesses, analyzes, and generates repair policies for video data.

[0189] Database: Stores received video data and analysis results.

[0190] Generative AI model: Analyzes video data and detects damaged areas.

[0191] User interface: Displays broken areas and repair plans on a map and notifies users in real time.

[0192] Processing flow

[0193] The device (drive recorder) transmits video data captured while the vehicle is driving to a server in real time. The server stores the received data in a database, performs preprocessing, and uses a generative AI model to detect road damage. It then uses a classification algorithm to evaluate the importance of the damage and generates a repair policy. The generated repair policy and information on the damaged areas are provided to the user through a user interface.

[0194] Specific examples

[0195] For example, while an autonomous vehicle is traveling on a highway, its dashcam captures damaged areas. This video data is sent to a server in real time and stored. The server preprocesses the video data and uses a generative AI model to detect the damaged areas, then classifies and evaluates the information. For example, a pothole that is 10 cm deep and 50 cm wide is classified as "severe damage." The information is then stored in a database along with its location, and "urgent repairs" or "complete repaving" are recommended, with estimated repair costs calculated. This information is provided to the autonomous vehicle operator through a user interface, allowing a specific repair plan to be quickly developed.

[0196] Prompt Sentence Examples

[0197] An example of a prompt for a model for automatically detecting road breaks is:

[0198] "Please use the following data to generate a repair plan for a pothole discovered while driving on a highway. Please propose the optimal repair plan based on the dashcam video data, GPS information, and details of the damaged area (location, type of damage, and level of damage)."

[0199] In this way, the present invention provides a system that enables real-time road monitoring and rapid planning of repair plans, and supports safe operation of autonomous vehicles.

[0200] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0201] Step 1: Image data collection process The device (drive recorder) collects video data taken while the vehicle is driving. At the same time, it also obtains location information and timestamps from the GPS module and adds them to the video data. This creates spatiotemporal information for each video frame.

[0202] Input: Video data during driving, GPS location information, timestamp

[0203] Output: Video data with location information and timestamp added

[0204] Step 2: Sending video data The device sends the collected video data to the server in real time via the communication module (5G). Thanks to high-speed communication, the data is immediately transferred to the server.

[0205] Input: Video data with location information and timestamps

[0206] Output: Video data sent to the server

[0207] Step 3: Receiving and storing data The server receives the video data sent from the device, verifies the data integrity and format, and then stores it in the database.

[0208] Input: Video data sent from the device

[0209] Output: Video data stored in a database

[0210] Step 4: Data Preprocessing The server applies a noise reduction filter to the stored video data, then resizes it to a format suitable for analysis, extracts frames containing important information, and removes unnecessary frames.

[0211] Input: Video data stored in the database

[0212] Output: Denoised, resized and frame extracted data

[0213] Step 5: Image data analysis. The server inputs the preprocessed data into the generative AI model. The model performs image analysis and automatically detects road damage. Attribute information such as the type, extent, and depth of the detected damage is acquired.

[0214] Input: Preprocessed video data

[0215] Output: Attribute information of the damaged area (type, range, depth)

[0216] Step 6: Classification of Damage Level Based on the acquired attribute information of the damaged area, the server evaluates the damage level using a classification algorithm. Damage areas are classified into levels such as minor, caution, and serious.

[0217] Input: Attribute information of the damaged area

[0218] Output: Classified damage levels

[0219] Step 7: Generation of repair policy The server automatically generates a repair policy based on the classification results. The generated repair policy includes the urgency of repair, recommended repair methods, and estimated repair costs.

[0220] Input: Classified damage level

[0221] Output: Repair policy (urgency, repair method, estimated cost)

[0222] Step 8: Presenting the results The server presents the generated repair policy and information on the damaged areas to the user through a user interface. This information is displayed on a map so that the user can visually check it. Repair notifications are also sent in real time.

[0223] Input: Repair policy, information on damaged areas

[0224] Output: Information presentation and repair notification through user interface

[0225] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0226] This invention combines an emotion engine with a system that analyzes video data collected from a drive recorder and automatically generates road damage locations and repair plans, making it possible to present repair plans that are more suited to the user.The main components of this system are a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[0227] Program processing

[0228] 1. Image data collection

[0229] Device:

[0230] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[0231] 2. Receipt and storage of data

[0232] server:

[0233] The system receives video data sent from the device and verifies its format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, location information, and timestamp.

[0234] 3. Data Preprocessing

[0235] server:

[0236] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[0237] 4. Analysis of image data

[0238] server:

[0239] The preprocessed video data is input into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage.

[0240] 5. Damage Level Classification

[0241] server:

[0242] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or severe. This classification result and location information are stored in a database.

[0243] 6. Repair Policy Generation

[0244] server:

[0245] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[0246] 7. Leveraging Emotional Engines

[0247] server:

[0248] When an interaction with a user (road administrator) occurs through the user interface, the emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.). Based on the recognized emotions, the content of the repair policy to be presented and the way it is displayed are adjusted.

[0249] 8. Presentation of results

[0250] server:

[0251] The emotional data obtained from the emotion engine is analyzed to optimize the way repair plans are presented. For example, if the user is feeling anxious, detailed explanations and additional data can be provided to reassure them. The generated repair plans are then displayed in the user interface in an easy-to-understand format, such as a map or list view.

[0252] User:

[0253] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[0254] Specific examples

[0255] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10 cm deep and 50 cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[0256] In this way, the present invention realizes a system that supports road administrators in formulating road repair plans efficiently and accurately, and provides appropriate information while taking into consideration the feelings of users, thereby shortening the lead time until repairs and optimizing repair costs.

[0257] The processing flow will be explained below.

[0258] Step 1:

[0259] Device:

[0260] The drive recorder is installed in the vehicle and records video in real time while the vehicle is driving on the road. The built-in GPS sensor adds location information and a timestamp to the video data. The video data is then sent to a server in real time via a 5G communication module.

[0261] Step 2:

[0262] server:

[0263] Receives video data sent from the device. Automatically verifies the format and integrity of the received data. If there are no problems, stores the received data in storage. This includes the video data, location information, and timestamp.

[0264] Step 3:

[0265] server:

[0266] The stored video data is subjected to a noise reduction filter, the video resolution is resized to a format suitable for analysis, and pre-processing is performed to extract important frames and remove unnecessary frames.

[0267] Step 4:

[0268] server:

[0269] The preprocessed data is input into a generative AI model, which analyzes the video data to detect road damage (potholes, cracks, etc.) and extracts information on the type, extent, depth, and location of the damage.

[0270] Step 5:

[0271] server:

[0272] Based on the information on the detected damaged areas, a unique classification algorithm is applied to evaluate the level of damage, classifying the damaged areas into levels such as minor, caution, or serious, and saving the results in a database.

[0273] Step 6:

[0274] server:

[0275] Based on the classification results, an optimal repair plan is automatically generated. This repair plan includes the urgency of repair, recommended repair methods, estimated repair costs, etc. The repair plan is then created in report format.

[0276] Step 7:

[0277] server:

[0278] The system prepares to present a repair policy to the user through the user interface. When presenting the policy, the emotion engine analyzes the user's facial expressions and voice to recognize emotions in real time. Based on the recognized emotions (e.g., relief, anxiety, surprise, etc.), the way the repair policy is presented is adjusted.

[0279] Step 8:

[0280] server:

[0281] Based on feedback from the emotion engine, the repair plan presented to the user is optimized. For example, if the user is feeling anxious, the plan will be adjusted to provide detailed explanations and additional materials to reassure the user. The repair plan is displayed in the user interface in an easy-to-use format such as a map or list.

[0282] Step 9:

[0283] User:

[0284] The user (road administrator) can view detailed information about the damaged areas and repair plans through a web interface or a dedicated application. The information presented is adjusted by the emotion engine, allowing the user to quickly and appropriately obtain the information they need. Based on this, a specific repair plan can be quickly formulated.

[0285] Through the above processing flow, the system of the present invention helps road managers to create repair plans efficiently and accurately. In addition, the emotion engine allows the system to present information that takes into account the user's emotions, thereby improving user satisfaction.

[0286] Example 2

[0287] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0288] Although technology already exists to analyze road footage acquired from dashcams and detect damaged areas, there has not previously been any technology that provides a method for presenting repair plans that take the user's emotions into consideration. As a result, repair plans are not presented at the optimal time and in the optimal way based on the user's emotions, which leads to user anxiety and a lack of understanding.

[0289] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0290] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damage, means for generating a repair plan based on the classification results, means for presenting the generated repair plan to the user, and means for recognizing the user's emotions and adjusting the content and method of presenting the repair plan based on the emotions. This makes it possible to present a repair plan that takes the user's emotions into consideration, and to provide information with the most appropriate timing and content for the user.

[0291] A "dashcam" is a device installed in a vehicle that records video in real time while driving.

[0292] "Video data" refers to data that includes video frames recorded by a drive recorder and additional information such as associated location information and timestamps.

[0293] "Means for transmitting in real time" refers to the communication module and protocol for transferring video data obtained from the drive recorder to the server without delay.

[0294] "Means for storage" refers to a database system or storage device for stably and persistently storing received video data.

[0295] "Preprocessing means" refers to tools and algorithms that process stored video data, such as removing noise, adjusting resolution, and extracting important frames.

[0296] "Means of analysis" refers to models or software that identify road damage areas and extract attribute information based on preprocessed data.

[0297] "Means for classifying damage points" refers to an algorithm that analyzes the severity and extent of the detected damage points and groups them according to different classification criteria.

[0298] "Means for generating repair policies" refers to a system that automatically calculates the repair method, urgency, estimated cost, etc. for damaged areas based on the classification results, and creates an appropriate repair plan.

[0299] "Presentation means" refers to a system or software for visually displaying the generated repair policy through a user interface in an easy-to-understand manner.

[0300] "Means for recognizing emotions" refers to engines or algorithms that analyze a user's facial expressions and voice on the user interface and identify emotions in real time.

[0301] "Adjustment means" refers to a mechanism that dynamically changes the content and method of presenting repair policies based on the recognized user emotions.

[0302] This invention is a system that analyzes video data collected from a drive recorder, detects damaged areas on roads, and automatically generates repair plans. By combining this system with an emotion engine, it is possible to present appropriate repair plans according to the user's emotions.

[0303] The system mainly consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[0304] Drive recorder and communication module

[0305] Device:

[0306] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[0307] Servers and Databases

[0308] server:

[0309] The server stores the received video data, which undergoes pre-processing to apply noise reduction filters and resize the resolution to a format optimized for analysis, improving the quality of the video data and making it easier to analyze.

[0310] Analysis using generative AI models

[0311] server:

[0312] The preprocessed data is input into a generative AI model. This model detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage. It then applies a classification algorithm based on the damage information to classify the damage level as minor, caution, or severe. The classification results and location information are stored in a database.

[0313] Generation and presentation of repair policies

[0314] server:

[0315] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[0316] Use of emotion engine

[0317] server:

[0318] The emotion engine recognizes the user's emotions in real time. It analyzes the user's facial expressions and voice to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and presentation of the repair plan presented.

[0319] User Interface

[0320] User:

[0321] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[0322] Specific examples

[0323] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10cm deep and 50cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[0324] Prompt Sentence Examples

[0325] Below are some example prompts to input to a generative AI model:

[0326] "Analyze the following video data to detect road damage. Damage data should include attribute information such as type, extent, depth, and location."

[0327] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0328] Step 1:

[0329] Image data collection

[0330] Device:

[0331] A drive recorder is installed in the vehicle. While the vehicle is driving on the road, the drive recorder records video in real time. The built-in GPS sensor acquires the current location information and adds a timestamp and location information to the video data.

[0332] Specific behavior:

[0333] The dashcam starts up and the camera starts recording video, and this video data is the input.

[0334] The GPS sensor obtains the current latitude and longitude information.

[0335] The acquired location information and timestamp are added to the video frame.

[0336] This becomes the video data as output.

[0337] Step 2:

[0338] Sending and Receiving Data

[0339] Devices and Servers:

[0340] Using a 5G communication module, the video data collected in real time is transmitted to a server, which receives and stores the data.

[0341] Specific behavior:

[0342] The video data is transmitted through the 5G communication module, and this transmitted data is the input.

[0343] The server listens to receive data.

[0344] Check the format and integrity of the data received.

[0345] The validated data is saved in a database. The saved data is the output.

[0346] Step 3:

[0347] Data Preprocessing

[0348] server:

[0349] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[0350] Specific behavior:

[0351] The saved video data is read out. This data is the input.

[0352] Improve the quality of your footage with noise reduction filters.

[0353] Resize the video resolution to the optimal format for AI analysis.

[0354] Frame analysis is performed to select and extract important frames.

[0355] Optimize data volume by removing unnecessary frames.

[0356] After these operations, the preprocessed data is the output.

[0357] Step 4:

[0358] Image data analysis

[0359] server:

[0360] The preprocessed data is input into a generative AI model, which detects road damage from the video data and extracts attribute information such as its type, extent, depth, and location.

[0361] Specific behavior:

[0362] The preprocessed data is fed into the generative AI model. This data is the input.

[0363] The AI ​​model detects the damaged areas.

[0364] Attribute information such as the type, extent, and depth of the damage is extracted.

[0365] The extracted information is the output.

[0366] Step 5:

[0367] Damage Level Classification

[0368] server:

[0369] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or serious. This classification result and location information are stored in a database.

[0370] Specific behavior:

[0371] The extracted damage information is fed into a classification algorithm, and this data is the input.

[0372] An algorithm classifies the level of damage as either "minor," "cautionary," or "severe."

[0373] The classification results and location information are saved in a database. The saved data is the output.

[0374] Step 6:

[0375] Generating repair policies

[0376] server:

[0377] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[0378] Specific behavior:

[0379] Based on the classification results, a repair policy generation algorithm is applied. This data is the input.

[0380] The algorithm calculates the urgency of repairs, recommended repair methods, and estimated repair costs.

[0381] The repair policy is compiled in a report format. This is the output report.

[0382] Step 7:

[0383] Use of emotion engine

[0384] server:

[0385] The system uses an emotion engine to recognize user emotions in real time. It analyzes the user's facial expressions and voice on the user interface to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and display method of the repair policy presented.

[0386] Specific behavior:

[0387] The user's facial expressions and voice data are input to the emotion engine. This data is the input.

[0388] The emotion engine analyzes and identifies the user's emotions.

[0389] Based on the recognized emotion data, the presentation method and content of the repair policy are adjusted, and the adjusted presentation method is the output.

[0390] Step 8:

[0391] Presentation of results

[0392] Servers and Users:

[0393] The generated repair policy is visually presented to the user through a user interface, and the user can then create a specific repair plan based on this information.

[0394] Specific behavior:

[0395] The repair policy adjusted by the emotion engine is displayed on the user interface, and the repair policy report is the input.

[0396] The user views the repair policy through a web interface or a dedicated application.

[0397] The user quickly creates a repair plan based on the information, and the repair plan with the corrective measures is the output.

[0398] (Application example 2)

[0399] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0400] Conventional systems for detecting road damage and generating repair plans have the problem of being unable to respond quickly because it takes time to analyze video data and generate repair plans.In addition, they do not provide information that takes into account the situation and emotions of each user, so they are unable to provide users with a sufficient sense of security.

[0401] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0402] In this invention, the server includes a means for transmitting video data acquired from the drive recorder in real time, a means for saving the received video data, and a means for preprocessing the saved video data. This enables real-time collection and saving of video data. The server also includes a means for analyzing the preprocessed data to detect road damage, a means for classifying the detected damage, a means for generating a repair plan based on the classification results, and a means for optimizing and presenting the generated repair plan in response to the user's emotions. This enables information to be provided quickly and in consideration of the user's emotions.

[0403] A "drive recorder" is a device that is installed in a vehicle and records video and audio data in real time while the vehicle is driving.

[0404] "Means for receiving" refers to a device or method for receiving signals or data transmitted from an external source and storing or processing them.

[0405] The term "storing means" refers to a device or method that stores received or generated data in a storage device and keeps it in a reusable state as needed.

[0406] "Preprocessing means" refers to a device or method that performs a series of initial processes to improve the efficiency of data analysis, including noise removal and resolution adjustment.

[0407] The "means for analyzing and detecting road damage" refers to a device or method that uses a specific algorithm to find damaged portions of the road from the pre-processed data.

[0408] The "classifying means" refers to a device or method for grouping detected damaged areas based on specific criteria, including classification by the degree or type of damage.

[0409] The "means for generating a repair policy" is a device or method that automatically determines the optimal repair method and priorities based on the classification results.

[0410] "Means for optimizing and presenting according to emotions" refers to a device or method that analyzes the user's emotional state in real time, adjusts the display content and display method based on that information, and reduces stress and anxiety.

[0411] An "emotion engine" is software or hardware that analyzes emotions from a user's facial expressions, voice, etc., and detects specific emotions.

[0412] This invention is a system for autonomous vehicles that detects road damage in real time, generates appropriate repair policies, and presents optimized information according to the user's emotions. The system consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[0413] 1. System configuration and functions

[0414] 1.1 Drive recorder

[0415] A dashcam is a device that records video data in real time while a vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. Furthermore, using a 5G communication module, this video data is sent to a server without delay.

[0416] 1.2 Server

[0417] The server stores the received video data and verifies its format and completeness. The stored data includes video data, location information, and timestamps. During the data preprocessing stage, the server applies a noise reduction filter and adjusts the resolution to create an optimal format for analysis. Next, a generative AI model is used to detect road damage from the preprocessed video data. The detection results are extracted as attribute information, such as the type and extent of the damage, and location information, and the damage level is classified based on this information.

[0418] 1.3 Generating Repair Policies

[0419] The server automatically generates an optimal repair policy based on the classification results. This repair policy includes the urgency of repairs, recommended repair methods, and estimated repair costs, and is generated in the form of a report.

[0420] 1.4 Emotion Engine

[0421] When an interaction with the user (road administrator) occurs through the user interface, an emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions such as "relief," "anxiety," and "surprise." Based on this emotion, the content of the repair policy presented and the way it is displayed can be adjusted. For example, if the user feels anxious, detailed explanations and additional data can be provided to reassure them.

[0422] 1.5 Presentation of results

[0423] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair plans are presented. For example, if the user is feeling anxious, it adds more detailed explanations and provides information that reassures the user. The generated repair plans are displayed in an easy-to-understand format on the user interface, such as a map or list view. Through this interface, the user can quickly check the repair plans and detailed information about the damaged areas, and then quickly create a specific repair plan.

[0424] 2. Specific examples

[0425] For example, consider the case where a large pothole has newly appeared on a road. The dashcam installed in an autonomous vehicle captures the incident, and the video data is sent to a server in real time. The server preprocesses the data and detects the pothole using a generative AI model. The resulting pothole is evaluated as 10 cm deep and 50 cm wide and classified as severely damaged. Repair options of "urgent repair" and "complete repaving" are recommended, and an estimated repair cost is calculated. When this information is presented to the user through the user interface, an emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, the system adjusts its response by providing additional explanations and data, presenting more detailed information to reassure the user. In this way, the system helps road managers plan repairs efficiently and accurately.

[0426] 3. Examples of prompts

[0427] Hardware used: Dashcam, GPS module, 5G communication module, server, user interface device

[0428] Software used: OpenCV, TensorFlow, EmotionRecognizer, and GPS module libraries

[0429] Examples of specific prompts include the following:

[0430] Prompt: "Real-time transmission of video data from dashcams, detection of road damage, and information display using emotion recognition."

[0431] As a result, the present invention has a function in which the entire system operates in cooperation with one another and can provide information in a form that is easy for the user to understand and that takes into consideration the user's emotions.

[0432] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0433] Step 1: Collecting video data from the dashcam

[0434] The device's drive recorder records video data in real time while the vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. The collected video data is then sent to a server via a 5G communication module.

[0435] Input: Video data captured while the vehicle is moving, GPS location information, and timestamp

[0436] Output: Real-time video data, location information, and time-stamped data sent to the server

[0437] Step 2: Receiving and storing data

[0438] The server receives the video data sent from the device in real time and verifies its format and integrity. If there are no problems, the data is stored in a database. The data includes the video data, location information, and timestamp.

[0439] Input: Real-time video data, location information, timestamp

[0440] Output: Validated data stored in a database

[0441] Step 3: Preprocessing the data

[0442] The server applies a noise reduction filter to the video data stored in the database, then adjusts the resolution to make it optimal for analysis, and preprocesses the data by extracting important frames and removing unnecessary ones.

[0443] Input: Video data stored in the database, location information, timestamp

[0444] Output: Preprocessed data with denoising and resolution adjustment

[0445] Step 4: Analyze the image data and detect any damage

[0446] The server inputs the preprocessed video data into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type and extent of the damage, as well as location information.

[0447] Input: Preprocessed video data

[0448] Output: Damage detection results (damage type, range, location information)

[0449] Step 5: Classify the level of damage

[0450] The server applies a classification algorithm based on the location of the damage and classifies the damage as minor, minor, or severe. This classification and location information are then stored in a database.

[0451] Input: Damage detection results (damage type, extent, location information)

[0452] Output: Classification result (damage level), location information

[0453] Step 6: Generate repair policies

[0454] The server automatically generates an optimal repair policy based on the classification results. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is saved in a database in report format.

[0455] Input: Classification result (damage level), location information

[0456] Output: Repair policy (urgency of repair, recommended repair method, estimated repair cost)

[0457] Step 7: Leverage your emotional engine

[0458] The server uses an emotion engine to recognize the user's emotions in real time when an interaction occurs with the user through the user interface. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.).

[0459] Input: User facial expressions and voice data obtained via the user interface

[0460] Output: User's emotional data (relief, anxiety, surprise, etc.)

[0461] Step 8: Presenting the results

[0462] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair policies are presented. The generated repair policies are displayed on the user interface in an easy-to-understand format, such as a map or list. Based on this information, users can quickly create specific repair plans.

[0463] Input: Repair policy, user emotion data

[0464] Output: Optimized information presentation and support for creating specific repair plans

[0465] This allows the entire system to work in harmony, providing users with fast, accurate information and support that takes their emotions into consideration.

[0466] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0467] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0468] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0469] [Second embodiment]

[0470] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0471] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0472] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0473] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0474] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0475] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0476] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0477] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0478] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0479] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0480] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0481] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0482] This invention provides a system that analyzes video data collected from a drive recorder and automatically identifies road damage and generates repair plans. The system's main components are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[0483] Program processing

[0484] 1. Image data collection

[0485] Device:

[0486] A drive recorder is installed in the vehicle and records video in real time as the vehicle travels along the road. The drive recorder acquires GPS information and adds location information and timestamps to the video. This video data is then sent to a server in real time via a 5G communication module.

[0487] 2. Receipt and storage of data

[0488] server:

[0489] The system receives video data sent from the device. The received data is verified for format and integrity, and if there are no problems, it is stored in a database. The stored data includes not only the video data itself, but also the corresponding location information and timestamp.

[0490] 3. Data Preprocessing

[0491] server:

[0492] A noise reduction filter is applied to the received video data. At this stage, the data is resized to a format that is optimal for the analysis model. Additionally, frames containing important information are extracted and unnecessary frames are removed. This preprocessing improves the accuracy and efficiency of the analysis.

[0493] 4. Analysis of image data

[0494] server:

[0495] The preprocessed data is then input into a generative AI model, which detects damage to the road surface and extracts attribute information such as its type, extent, and depth. The model is particularly effective at accurately detecting critical damage, such as large potholes and cracks.

[0496] 5. Damage Level Classification

[0497] server:

[0498] Based on the information of the detected damage, a classification algorithm is applied to assess the level of damage: the damage is classified as minor, caution, or severe, and stored in a database along with the corresponding location information.

[0499] 6. Repair Policy Generation

[0500] server:

[0501] Based on the classification results, the system automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs. This information allows road managers to create quick and accurate repair plans.

[0502] 7. Presentation of results

[0503] server:

[0504] An interface is generated to present the generated repair policy and information on damaged areas to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using a map display or list display.

[0505] User:

[0506] Through the provided interface, users (road administrators) can view detailed information about damaged areas and repair policies, which will enable them to prioritize repairs and develop specific repair plans.

[0507] Specific examples

[0508] For example, a vehicle traveling on a highway may capture a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10cm deep and 50cm wide. A classification algorithm then evaluates it as "severely damaged" and stores it in a database along with its location. Finally, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers via a user interface, allowing them to quickly develop specific repair plans.

[0509] In this way, the present invention provides a system that uses video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair policies, thereby reducing the burden on road administrators and realizing efficient road maintenance management.

[0510] The processing flow will be explained below.

[0511] Step 1:

[0512] Device:

[0513] The dashcam is installed in vehicles traveling on the road and records video in real time. The built-in GPS sensor periodically acquires location information while recording, adding a timestamp and location information to the video data. The video data is then sent to a server in real time using a 5G communication module.

[0514] Step 2:

[0515] server:

[0516] Receives video data sent from the device. The received data is verified for format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, corresponding location information, and timestamp.

[0517] Step 3:

[0518] server:

[0519] The stored video data is preprocessed by applying a noise reduction filter, resizing the video resolution to a format suitable for analysis, and extracting important frames and removing unnecessary ones.

[0520] Step 4:

[0521] server:

[0522] The preprocessed data is input into a generative AI model, which detects road damage (potholes, cracks, etc.) from the collected video data. At this stage, attribute information such as the type, extent, depth, and location of the damage is extracted.

[0523] Step 5:

[0524] server:

[0525] The damage level is assessed based on the extracted information on the damaged areas. A unique classification algorithm is applied to categorize the damage into several levels, such as minor, caution, and serious. The classification results and location information are stored in a database.

[0526] Step 6:

[0527] server:

[0528] Based on the classification results, an algorithm is applied to automatically generate an optimal repair policy. The repair policy includes the urgency of the repair, the recommended repair method (e.g., partial patching, full repaving), and the estimated repair cost. The generated repair policy is then compiled in the form of a report.

[0529] Step 7:

[0530] server:

[0531] An interface is generated to provide the generated repair policy and related information to the user. The interface is designed for ease of use, including map and list views, allowing users to quickly check the information they need.

[0532] Step 8:

[0533] User:

[0534] Users (road administrators) access the server through a web interface or a dedicated application. On the dashboard, they can view detailed information, such as a map of damaged areas and a list of damage levels. Based on the repair policy provided, they can quickly create specific repair plans.

[0535] In this way, the system of the present invention helps road managers to plan road repairs efficiently and accurately. The automated process shortens the lead time for repairs and optimizes repair costs.

[0536] Example 1

[0537] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0538] In modern transportation infrastructure, it is important to quickly and accurately detect road damage and develop appropriate repair plans. However, conventional methods have the drawback of requiring a large amount of manpower, time, and cost. Furthermore, there is a risk of overlooking road damage and the difficulty of prioritizing repairs. The present invention aims to solve these problems and provide an efficient and accurate road management method.

[0539] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0540] In this invention, the server includes a device for transmitting video data acquired from the drive recorder in real time, a device for saving the received video data, a device for preprocessing the saved video data, a generative model for analyzing the preprocessed data and detecting road damage, an algorithm for classifying the detected damage, a method for generating a repair policy based on the classification results, and an interface for presenting the generated repair policy to the user. This enables quick and accurate detection of road damage and efficient planning of repairs.

[0541] A "drive recorder" is a device that is installed in a vehicle and records video data while the vehicle is driving.

[0542] "Video data" refers to data including images and video information recorded by a drive recorder.

[0543] A "real-time transmission device" is a communication device for instantly transferring acquired video data to a remote server.

[0544] The "receiving device" is a part of the server that receives video data transmitted from a remote location.

[0545] The "storing device" is a storage device that stores the received video data in a database or the like so that it can be used later.

[0546] A "preprocessing device" is a device that converts received video data into a format suitable for analysis. Specifically, it performs processes such as noise removal and resolution adjustment.

[0547] The "generative model for analysis" is an AI model that identifies damaged areas on roads based on preprocessed video data and extracts their detailed attribute information.

[0548] The "classification algorithm" is an algorithm for evaluating and classifying the level of damage based on information about the detected damaged portion.

[0549] The "method for generating a repair policy" is a process for determining the optimum repair method, urgency, and estimated repair cost for the classified damaged areas.

[0550] An "interface" is a system that visually displays the generated repair policy and information on damaged areas, allowing users to easily access them.

[0551] This invention relates to a system that analyzes video data collected from a drive recorder and automatically generates road damage information and repair plans. The main components of the system are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[0552] The drive recorder is installed in the vehicle and records video in real time while driving. The drive recorder acquires GPS information and adds location information and timestamps to the video. Using a 5G communication module, this video data is sent to a server with low latency.

[0553] The server receives the video data and verifies the format and integrity of the received data. If there are no problems, the data is stored in a database. The stored data includes not only the video but also the corresponding location information and timestamp. Image processing libraries such as OpenCV are used to preprocess the video data, such as noise reduction and resizing.

[0554] The pre-processed data is then analyzed by a generative AI model, which uses deep learning frameworks such as TensorFlow or PyTorch to detect damage to the road surface and extract its type, extent, and depth, for example using a convolutional neural network (CNN).

[0555] The server then applies a classification algorithm based on the detected damage information, which assesses the level of damage based on parameters such as the area, depth, and width of the damage. The data is then classified into levels such as minor, caution, and severe and stored in a database.

[0556] Furthermore, the server automatically generates an optimal repair policy based on the classification results, which includes the urgency of repair, recommended repair methods, and estimated repair costs, enabling road managers to quickly develop repair plans.

[0557] The generated repair plan and damage location information are presented to the user through an interface that can be used in a web browser or mobile application, and the information is presented in an easy-to-understand manner using map and list views. Front-end frameworks such as React and Angular are used, and Node.js and Django are used for the back-end.

[0558] As a concrete example, a vehicle traveling on a highway captures a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10 cm deep and 50 cm wide. A classification algorithm evaluates it as "severely damaged" and stores it in a database along with its location information. Ultimately, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers through a user interface, allowing them to quickly develop specific repair plans.

[0559] An example of a prompt to be input into the generative AI model is, "Analyze the video data acquired while the vehicle is traveling and detect any newly occurring potholes or cracks. Also, measure the depth and width of the damaged areas and propose a repair plan."

[0560] In this way, the present invention provides a system that utilizes video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair plans, thereby achieving more efficient road management and reducing costs.

[0561] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0562] Step 1: Collecting image data

[0563] Device:

[0564] The dashcam is installed in a vehicle and records high-resolution video in real time as the vehicle travels along the road. The dashcam acquires GPS information and adds location information and timestamps to the video data. Using a 5G communication module, this video data is transmitted to a server with low latency.

[0565] Input: Real-time video and GPS information acquired while the vehicle is running

[0566] Data processing: Add location information and timestamp to each video frame

[0567] Output: Video data with additional information

[0568] Step 2: Receiving and saving video data

[0569] server:

[0570] The server receives the video data sent from the device. The format and integrity of the received data are verified, and if there are no problems, it is stored in a database. The stored data includes not only the video, but also the corresponding location information and timestamp.

[0571] Input: Video data with additional information sent from the device

[0572] Data processing: verifying data format and integrity (e.g., CRC check, format verification)

[0573] Output: Verified video data is stored in a database

[0574] Step 3: Preprocessing the data

[0575] server:

[0576] The server applies a noise reduction filter to the received video data. This process resizes the video to a format that is optimized for the analysis model. It also extracts frames that contain important information for analysis and removes unnecessary frames.

[0577] Input: Verified video data stored in a database

[0578] Data processing: noise removal (e.g., Gaussian filter application), resolution adjustment, extraction of important frames, deletion of unnecessary frames

[0579] Output: Pre-processed video data

[0580] Step 4: Analyzing the image data

[0581] server:

[0582] The server inputs the preprocessed data into a generative AI model to detect damage on the road surface, which uses deep learning frameworks such as TensorFlow and PyTorch to extract the type, extent, and depth of the damage.

[0583] Input: Preprocessed video data

[0584] Data calculation: Deep learning model for detecting damage and extracting attribute information

[0585] Output: Specific attribute information and location information of detected damage points

[0586] Step 5: Classify the level of damage

[0587] server:

[0588] The server then applies a classification algorithm to assess the level of damage detected: damage is classified as minor, minor, or severe, and these classifications are stored in a database.

[0589] Input: Attribute information and location information of the detected damage

[0590] Data processing: Assessing the level of corruption with classification algorithms

[0591] Output: Damage level assessment data and location information are saved in a database.

[0592] Step 6: Generate repair policies

[0593] server:

[0594] Based on the classification results, the server automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs.

[0595] Input: Damage level assessment data and location information

[0596] Data calculation: Urgency assessment, repair method proposal, repair cost estimate

[0597] Output: Integrated information including repair policy

[0598] Step 7: Presenting the results

[0599] server:

[0600] The server generates an interface to present the generated repair policy and damage location information to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using map and list views.

[0601] User:

[0602] The user (road administrator) can view detailed information about the damaged areas and repair policies through the provided interface.

[0603] Input: Integrated information including repair policy

[0604] Data calculation: Visualization of information (e.g., displaying potholes on a map, displaying repair policies in list format)

[0605] Output: A user-visible interface

[0606] (Application example 1)

[0607] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0608] Conventional road management systems have difficulty quickly and accurately detecting road damage and providing appropriate repair plans. In particular, autonomous vehicles require real-time road condition monitoring to ensure safety, making efficient road management essential.

[0609] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0610] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damaged areas, means for generating a repair policy based on the classification results, means for presenting the generated repair policy to a user, means for displaying information on the damaged areas on a map, and means for notifying the user of repairs to the damaged areas in real time, thereby enabling real-time road condition monitoring and rapid planning of repairs.

[0611] A "dashcam" is a device installed in a vehicle that records video footage while driving.

[0612] "Means for transmitting in real time" refers to a communication means for transferring video data to a receiving destination such as a server immediately without delay.

[0613] "Means for storing received video data" refers to a storage device or database for storing the received video data in an appropriate format.

[0614] The "means for pre-processing stored video data" is a function for performing initial processing such as noise removal and format adjustment on video data.

[0615] The "means for analyzing the pre-processed data and detecting road damage" is an algorithm or software for identifying damaged portions of the road from the pre-processed data using image analysis techniques.

[0616] The "means for classifying detected damage areas" refers to a classification algorithm for classifying and rating damage areas based on the type and severity of damage.

[0617] The "means for generating repair policies based on classification results" refers to algorithms or systems that automatically plan and propose optimal repair methods based on classified damage location information.

[0618] The "means for presenting the generated repair policy to the user" refers to a display interface or notification system for providing the user with the repair plan and information on the damaged area.

[0619] The "means for displaying information on damaged areas on a map" is a function that visually shows identified damaged areas on a map using a geographic information system (GIS).

[0620] "Means for providing real-time notification of damaged area repairs" refers to a communication and notification system that immediately notifies of the need for repairs when a damaged area is detected.

[0621] This invention is a system that uses video data acquired from a vehicle's onboard dashcam to detect road damage in real time and automatically generate and present repair plans. The system's main hardware components are a dashcam, GPS module, and communication module, while the software components include a server, database, generative AI model, and user interface.

[0622] System Overview

[0623] The server receives, stores, pre-processes, analyzes, and detects damages in real time, classifies them, and generates repair plans. This data is then provided to the user through a user interface, including map displays and notifications.

[0624] Hardware and software used

[0625] Drive recorder: Acquires video data and GPS information.

[0626] GPS module: Adds location information and timestamps to video data.

[0627] 5G communication module: Transmits video data to a server in real time.

[0628] Server: Receives, stores, preprocesses, analyzes, and generates repair policies for video data.

[0629] Database: Stores received video data and analysis results.

[0630] Generative AI model: Analyzes video data and detects damaged areas.

[0631] User interface: Displays broken areas and repair plans on a map and notifies users in real time.

[0632] Processing flow

[0633] The device (drive recorder) transmits video data captured while the vehicle is driving to a server in real time. The server stores the received data in a database, performs preprocessing, and uses a generative AI model to detect road damage. It then uses a classification algorithm to evaluate the importance of the damage and generates a repair policy. The generated repair policy and information on the damaged areas are provided to the user through a user interface.

[0634] Specific examples

[0635] For example, while an autonomous vehicle is traveling on a highway, its dashcam captures damaged areas. This video data is sent to a server in real time and stored. The server preprocesses the video data and uses a generative AI model to detect the damaged areas, then classifies and evaluates the information. For example, a pothole that is 10 cm deep and 50 cm wide is classified as "severe damage." The information is then stored in a database along with its location, and "urgent repairs" or "complete repaving" are recommended, with estimated repair costs calculated. This information is provided to the autonomous vehicle operator through a user interface, allowing a specific repair plan to be quickly developed.

[0636] Prompt Sentence Examples

[0637] An example of a prompt for a model for automatically detecting road breaks is:

[0638] "Please use the following data to generate a repair plan for a pothole discovered while driving on a highway. Please propose the optimal repair plan based on the dashcam video data, GPS information, and details of the damaged area (location, type of damage, and level of damage)."

[0639] In this way, the present invention provides a system that enables real-time road monitoring and rapid planning of repair plans, and supports safe operation of autonomous vehicles.

[0640] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0641] Step 1: Image data collection process The device (drive recorder) collects video data taken while the vehicle is driving. At the same time, it also obtains location information and timestamps from the GPS module and adds them to the video data. This creates spatiotemporal information for each video frame.

[0642] Input: Video data during driving, GPS location information, timestamp

[0643] Output: Video data with location information and timestamp added

[0644] Step 2: Sending video data The device sends the collected video data to the server in real time via the communication module (5G). Thanks to high-speed communication, the data is immediately transferred to the server.

[0645] Input: Video data with location information and timestamps

[0646] Output: Video data sent to the server

[0647] Step 3: Receiving and storing data The server receives the video data sent from the device, verifies the data integrity and format, and then stores it in the database.

[0648] Input: Video data sent from the device

[0649] Output: Video data stored in a database

[0650] Step 4: Data Preprocessing The server applies a noise reduction filter to the stored video data, then resizes it to a format suitable for analysis, extracts frames containing important information, and removes unnecessary frames.

[0651] Input: Video data stored in the database

[0652] Output: Denoised, resized and frame extracted data

[0653] Step 5: Image data analysis. The server inputs the preprocessed data into the generative AI model. The model performs image analysis and automatically detects road damage. Attribute information such as the type, extent, and depth of the detected damage is acquired.

[0654] Input: Preprocessed video data

[0655] Output: Attribute information of the damaged area (type, range, depth)

[0656] Step 6: Classification of Damage Level Based on the acquired attribute information of the damaged area, the server evaluates the damage level using a classification algorithm. Damage areas are classified into levels such as minor, caution, and serious.

[0657] Input: Attribute information of the damaged area

[0658] Output: Classified damage levels

[0659] Step 7: Generation of repair policy The server automatically generates a repair policy based on the classification results. The generated repair policy includes the urgency of repair, recommended repair methods, and estimated repair costs.

[0660] Input: Classified damage level

[0661] Output: Repair policy (urgency, repair method, estimated cost)

[0662] Step 8: Presenting the results The server presents the generated repair policy and information on the damaged areas to the user through a user interface. This information is displayed on a map so that the user can visually check it. Repair notifications are also sent in real time.

[0663] Input: Repair policy, information on damaged area

[0664] Output: Information presentation and repair notification through user interface

[0665] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0666] This invention combines an emotion engine with a system that analyzes video data collected from a drive recorder and automatically generates road damage locations and repair plans, making it possible to present repair plans that are more suited to the user.The main components of this system are a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[0667] Program processing

[0668] 1. Image data collection

[0669] Device:

[0670] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[0671] 2. Receipt and storage of data

[0672] server:

[0673] The system receives video data sent from the device and verifies its format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, location information, and timestamp.

[0674] 3. Data Preprocessing

[0675] server:

[0676] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[0677] 4. Analysis of image data

[0678] server:

[0679] The preprocessed video data is input into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage.

[0680] 5. Damage Level Classification

[0681] server:

[0682] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or severe. This classification result and location information are stored in a database.

[0683] 6. Repair Policy Generation

[0684] server:

[0685] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[0686] 7. Leveraging Emotional Engines

[0687] server:

[0688] When an interaction with a user (road administrator) occurs through the user interface, the emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.). Based on the recognized emotions, the content of the repair policy to be presented and the way it is displayed are adjusted.

[0689] 8. Presentation of results

[0690] server:

[0691] The emotional data obtained from the emotion engine is analyzed to optimize the way repair plans are presented. For example, if the user is feeling anxious, detailed explanations and additional data can be provided to reassure them. The generated repair plans are then displayed in the user interface in an easy-to-understand format, such as a map or list view.

[0692] User:

[0693] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[0694] Specific examples

[0695] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10 cm deep and 50 cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[0696] In this way, the present invention realizes a system that supports road administrators in formulating road repair plans efficiently and accurately, and provides appropriate information while taking into consideration the feelings of users, thereby shortening the lead time until repairs and optimizing repair costs.

[0697] The processing flow will be explained below.

[0698] Step 1:

[0699] Device:

[0700] The drive recorder is installed in the vehicle and records video in real time while the vehicle is driving on the road. The built-in GPS sensor adds location information and a timestamp to the video data. The video data is then sent to a server in real time via a 5G communication module.

[0701] Step 2:

[0702] server:

[0703] Receives video data sent from the device. Automatically verifies the format and integrity of the received data. If there are no problems, stores the received data in storage. This includes the video data, location information, and timestamp.

[0704] Step 3:

[0705] server:

[0706] The stored video data is subjected to a noise reduction filter, the video resolution is resized to a format suitable for analysis, and pre-processing is performed to extract important frames and remove unnecessary frames.

[0707] Step 4:

[0708] server:

[0709] The preprocessed data is input into a generative AI model, which analyzes the video data to detect road damage (potholes, cracks, etc.) and extracts information on the type, extent, depth, and location of the damage.

[0710] Step 5:

[0711] server:

[0712] Based on the information on the detected damaged areas, a unique classification algorithm is applied to evaluate the level of damage, classifying the damaged areas into levels such as minor, caution, or serious, and saving the results in a database.

[0713] Step 6:

[0714] server:

[0715] Based on the classification results, an optimal repair plan is automatically generated. This repair plan includes the urgency of repair, recommended repair methods, estimated repair costs, etc. The repair plan is then created in report format.

[0716] Step 7:

[0717] server:

[0718] The system prepares to present a repair policy to the user through the user interface. When presenting the policy, the emotion engine analyzes the user's facial expressions and voice to recognize emotions in real time. Based on the recognized emotions (e.g., relief, anxiety, surprise, etc.), the way the repair policy is presented is adjusted.

[0719] Step 8:

[0720] server:

[0721] Based on feedback from the emotion engine, the repair plan presented to the user is optimized. For example, if the user is feeling anxious, the plan will be adjusted to provide detailed explanations and additional materials to reassure the user. The repair plan is displayed in the user interface in an easy-to-use format such as a map or list.

[0722] Step 9:

[0723] User:

[0724] The user (road administrator) can view detailed information about the damaged areas and repair plans through a web interface or a dedicated application. The information presented is adjusted by the emotion engine, allowing the user to quickly and appropriately obtain the information they need. Based on this, a specific repair plan can be quickly formulated.

[0725] Through the above processing flow, the system of the present invention helps road managers to create repair plans efficiently and accurately. In addition, the emotion engine allows the system to present information that takes into account the user's emotions, thereby improving user satisfaction.

[0726] Example 2

[0727] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0728] Although technology already exists to analyze road footage acquired from dashcams and detect damaged areas, there has not previously been any technology that provides a method for presenting repair plans that take the user's emotions into consideration. As a result, repair plans are not presented at the optimal time and in the optimal way based on the user's emotions, which leads to user anxiety and a lack of understanding.

[0729] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0730] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damage, means for generating a repair plan based on the classification results, means for presenting the generated repair plan to the user, and means for recognizing the user's emotions and adjusting the content and method of presenting the repair plan based on the emotions. This makes it possible to present a repair plan that takes the user's emotions into consideration, and to provide information with the most appropriate timing and content for the user.

[0731] A "dashcam" is a device installed in a vehicle that records video in real time while driving.

[0732] "Video data" refers to data that includes video frames recorded by a drive recorder and additional information such as associated location information and timestamps.

[0733] "Means for transmitting in real time" refers to the communication module and protocol for transferring video data obtained from the drive recorder to the server without delay.

[0734] "Means for storage" refers to a database system or storage device for stably and persistently storing received video data.

[0735] "Preprocessing means" refers to tools and algorithms that process stored video data, such as removing noise, adjusting resolution, and extracting important frames.

[0736] "Means of analysis" refers to models or software that identify road damage areas and extract attribute information based on preprocessed data.

[0737] "Means for classifying damage points" refers to an algorithm that analyzes the severity and extent of the detected damage points and groups them according to different classification criteria.

[0738] "Means for generating repair policies" refers to a system that automatically calculates the repair method, urgency, estimated cost, etc. for damaged areas based on the classification results, and creates an appropriate repair plan.

[0739] "Presentation means" refers to a system or software for visually displaying the generated repair policy through a user interface in an easy-to-understand manner.

[0740] "Means for recognizing emotions" refers to engines or algorithms that analyze a user's facial expressions and voice on the user interface and identify emotions in real time.

[0741] "Adjustment means" refers to a mechanism that dynamically changes the content and method of presenting repair policies based on the recognized user emotions.

[0742] This invention is a system that analyzes video data collected from a drive recorder, detects damaged areas on roads, and automatically generates repair plans. By combining this system with an emotion engine, it is possible to present appropriate repair plans according to the user's emotions.

[0743] The system mainly consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[0744] Drive recorder and communication module

[0745] Device:

[0746] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[0747] Servers and Databases

[0748] server:

[0749] The server stores the received video data, which undergoes pre-processing to apply noise reduction filters and resize the resolution to a format optimized for analysis, improving the quality of the video data and making it easier to analyze.

[0750] Analysis using generative AI models

[0751] server:

[0752] The preprocessed data is input into a generative AI model. This model detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage. It then applies a classification algorithm based on the damage information to classify the damage level as minor, caution, or severe. The classification results and location information are stored in a database.

[0753] Generation and presentation of repair policies

[0754] server:

[0755] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[0756] Use of emotion engine

[0757] server:

[0758] The emotion engine recognizes the user's emotions in real time. It analyzes the user's facial expressions and voice to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and presentation of the repair plan presented.

[0759] User Interface

[0760] User:

[0761] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[0762] Specific examples

[0763] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10cm deep and 50cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[0764] Prompt Sentence Examples

[0765] Below are some example prompts to input to a generative AI model:

[0766] "Analyze the following video data to detect road damage. Damage data should include attribute information such as type, extent, depth, and location."

[0767] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0768] Step 1:

[0769] Image data collection

[0770] Device:

[0771] A drive recorder is installed in the vehicle. While the vehicle is driving on the road, the drive recorder records video in real time. The built-in GPS sensor acquires the current location information and adds a timestamp and location information to the video data.

[0772] Specific behavior:

[0773] The dashcam starts up and the camera starts recording video, and this video data is the input.

[0774] The GPS sensor obtains the current latitude and longitude information.

[0775] The acquired location information and timestamp are added to the video frame.

[0776] This becomes the video data as output.

[0777] Step 2:

[0778] Sending and Receiving Data

[0779] Devices and Servers:

[0780] Using a 5G communication module, the video data collected in real time is transmitted to a server, which receives and stores the data.

[0781] Specific behavior:

[0782] The video data is transmitted through the 5G communication module, and this transmitted data is the input.

[0783] The server listens to receive data.

[0784] Check the format and integrity of the data received.

[0785] The validated data is saved in a database. The saved data is the output.

[0786] Step 3:

[0787] Data Preprocessing

[0788] server:

[0789] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[0790] Specific behavior:

[0791] The saved video data is read out. This data is the input.

[0792] Improve the quality of your footage with noise reduction filters.

[0793] Resize the video resolution to the optimal format for AI analysis.

[0794] Frame analysis is performed to select and extract important frames.

[0795] Optimize data volume by removing unnecessary frames.

[0796] After these operations, the preprocessed data is the output.

[0797] Step 4:

[0798] Image data analysis

[0799] server:

[0800] The preprocessed data is input into a generative AI model, which detects road damage from the video data and extracts attribute information such as its type, extent, depth, and location.

[0801] Specific behavior:

[0802] The preprocessed data is fed into the generative AI model. This data is the input.

[0803] The AI ​​model detects the damaged areas.

[0804] Attribute information such as the type, extent, and depth of the damage is extracted.

[0805] The extracted information is the output.

[0806] Step 5:

[0807] Damage Level Classification

[0808] server:

[0809] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or serious. This classification result and location information are stored in a database.

[0810] Specific behavior:

[0811] The extracted damage information is fed into a classification algorithm, and this data is the input.

[0812] An algorithm classifies the level of damage as either "minor," "cautionary," or "severe."

[0813] The classification results and location information are saved in a database. The saved data is the output.

[0814] Step 6:

[0815] Generating repair policies

[0816] server:

[0817] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[0818] Specific behavior:

[0819] Based on the classification results, a repair policy generation algorithm is applied. This data is the input.

[0820] The algorithm calculates the urgency of repairs, recommended repair methods, and estimated repair costs.

[0821] The repair policy is compiled in a report format. This is the output report.

[0822] Step 7:

[0823] Use of emotion engine

[0824] server:

[0825] The system uses an emotion engine to recognize user emotions in real time. It analyzes the user's facial expressions and voice on the user interface to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and display method of the repair policy presented.

[0826] Specific behavior:

[0827] The user's facial expressions and voice data are input to the emotion engine. This data is the input.

[0828] The emotion engine analyzes and identifies the user's emotions.

[0829] Based on the recognized emotion data, the presentation method and content of the repair policy are adjusted, and the adjusted presentation method is the output.

[0830] Step 8:

[0831] Presentation of results

[0832] Servers and Users:

[0833] The generated repair policy is visually presented to the user through a user interface, and the user can then create a specific repair plan based on this information.

[0834] Specific behavior:

[0835] The repair policy adjusted by the emotion engine is displayed on the user interface, and the repair policy report is the input.

[0836] The user views the repair policy through a web interface or a dedicated application.

[0837] The user quickly creates a repair plan based on the information, and the repair plan with the corrective measures is the output.

[0838] (Application example 2)

[0839] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0840] Conventional systems for detecting road damage and generating repair plans have the problem of being unable to respond quickly because it takes time to analyze video data and generate repair plans.In addition, they do not provide information that takes into account the situation and emotions of each user, so they are unable to provide users with a sufficient sense of security.

[0841] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0842] In this invention, the server includes a means for transmitting video data acquired from the drive recorder in real time, a means for saving the received video data, and a means for preprocessing the saved video data. This enables real-time collection and saving of video data. The server also includes a means for analyzing the preprocessed data to detect road damage, a means for classifying the detected damage, a means for generating a repair plan based on the classification results, and a means for optimizing and presenting the generated repair plan in response to the user's emotions. This enables information to be provided quickly and in consideration of the user's emotions.

[0843] A "drive recorder" is a device that is installed in a vehicle and records video and audio data in real time while the vehicle is driving.

[0844] "Means for receiving" refers to a device or method for receiving signals or data transmitted from an external source and storing or processing them.

[0845] The term "storing means" refers to a device or method that stores received or generated data in a storage device and keeps it in a reusable state as needed.

[0846] "Preprocessing means" refers to a device or method that performs a series of initial processes to improve the efficiency of data analysis, including noise removal and resolution adjustment.

[0847] The "means for analyzing and detecting road damage" refers to a device or method that uses a specific algorithm to find damaged portions of the road from the pre-processed data.

[0848] The "classifying means" refers to a device or method for grouping detected damaged areas based on specific criteria, including classification by the degree or type of damage.

[0849] The "means for generating a repair policy" is a device or method that automatically determines the optimal repair method and priorities based on the classification results.

[0850] "Means for optimizing and presenting according to emotions" refers to a device or method that analyzes the user's emotional state in real time, adjusts the display content and display method based on that information, and reduces stress and anxiety.

[0851] An "emotion engine" is software or hardware that analyzes emotions from a user's facial expressions, voice, etc., and detects specific emotions.

[0852] This invention is a system for autonomous vehicles that detects road damage in real time, generates appropriate repair policies, and presents optimized information according to the user's emotions. The system consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[0853] 1. System configuration and functions

[0854] 1.1 Drive recorder

[0855] A dashcam is a device that records video data in real time while a vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. Furthermore, using a 5G communication module, this video data is sent to a server without delay.

[0856] 1.2 Server

[0857] The server stores the received video data and verifies its format and completeness. The stored data includes video data, location information, and timestamps. During the data preprocessing stage, the server applies a noise reduction filter and adjusts the resolution to create an optimal format for analysis. Next, a generative AI model is used to detect road damage from the preprocessed video data. The detection results are extracted as attribute information, such as the type and extent of the damage, and location information, and the damage level is classified based on this information.

[0858] 1.3 Generating Repair Policies

[0859] The server automatically generates an optimal repair policy based on the classification results. This repair policy includes the urgency of repairs, recommended repair methods, and estimated repair costs, and is generated in the form of a report.

[0860] 1.4 Emotion Engine

[0861] When an interaction with the user (road administrator) occurs through the user interface, an emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions such as "relief," "anxiety," and "surprise." Based on this emotion, the content of the repair policy presented and the way it is displayed can be adjusted. For example, if the user feels anxious, detailed explanations and additional data can be provided to reassure them.

[0862] 1.5 Presentation of results

[0863] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair plans are presented. For example, if the user is feeling anxious, it adds more detailed explanations and provides information that reassures the user. The generated repair plans are displayed in an easy-to-understand format on the user interface, such as a map or list view. Through this interface, the user can quickly check the repair plans and detailed information about the damaged areas, and then quickly create a specific repair plan.

[0864] 2. Specific examples

[0865] For example, consider the case where a large pothole has newly appeared on a road. The dashcam installed in an autonomous vehicle captures the incident, and the video data is sent to a server in real time. The server preprocesses the data and detects the pothole using a generative AI model. The resulting pothole is evaluated as 10 cm deep and 50 cm wide and classified as severely damaged. Repair options of "urgent repair" and "complete repaving" are recommended, and an estimated repair cost is calculated. When this information is presented to the user through the user interface, an emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, the system adjusts its response by providing additional explanations and data, presenting more detailed information to reassure the user. In this way, the system helps road managers plan repairs efficiently and accurately.

[0866] 3. Examples of prompts

[0867] Hardware used: Dashcam, GPS module, 5G communication module, server, user interface device

[0868] Software used: OpenCV, TensorFlow, EmotionRecognizer, and GPS module libraries

[0869] Examples of specific prompts include the following:

[0870] Prompt: "Real-time transmission of video data from dashcams, detection of road damage, and information display using emotion recognition."

[0871] As a result, the present invention has a function in which the entire system operates in cooperation with one another and can provide information in a form that is easy for the user to understand and that takes into consideration the user's emotions.

[0872] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0873] Step 1: Collecting video data from the dashcam

[0874] The device's drive recorder records video data in real time while the vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. The collected video data is then sent to a server via a 5G communication module.

[0875] Input: Video data captured while the vehicle is moving, GPS location information, and timestamp

[0876] Output: Real-time video data, location information, and time-stamped data sent to the server

[0877] Step 2: Receiving and storing data

[0878] The server receives the video data sent from the device in real time and verifies its format and integrity. If there are no problems, the data is stored in a database. The data includes the video data, location information, and timestamp.

[0879] Input: Real-time video data, location information, timestamp

[0880] Output: Validated data stored in a database

[0881] Step 3: Preprocessing the data

[0882] The server applies a noise reduction filter to the video data stored in the database, then adjusts the resolution to make it optimal for analysis, and preprocesses the data by extracting important frames and removing unnecessary ones.

[0883] Input: Video data stored in the database, location information, timestamp

[0884] Output: Preprocessed data with denoising and resolution adjustment

[0885] Step 4: Analyze the image data and detect any damage

[0886] The server inputs the preprocessed video data into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type and extent of the damage, as well as location information.

[0887] Input: Preprocessed video data

[0888] Output: Damage detection results (damage type, range, location information)

[0889] Step 5: Classify the level of damage

[0890] The server applies a classification algorithm based on the location of the damage and classifies the damage as minor, minor, or severe. This classification and location information are then stored in a database.

[0891] Input: Damage detection results (damage type, extent, location information)

[0892] Output: Classification result (damage level), location information

[0893] Step 6: Generate repair policies

[0894] The server automatically generates an optimal repair policy based on the classification results. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is saved in a database in report format.

[0895] Input: Classification result (damage level), location information

[0896] Output: Repair policy (urgency of repair, recommended repair method, estimated repair cost)

[0897] Step 7: Leverage your emotional engine

[0898] The server uses an emotion engine to recognize the user's emotions in real time when an interaction occurs with the user through the user interface. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.).

[0899] Input: User facial expressions and voice data obtained via the user interface

[0900] Output: User's emotional data (relief, anxiety, surprise, etc.)

[0901] Step 8: Presenting the results

[0902] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair policies are presented. The generated repair policies are displayed on the user interface in an easy-to-understand format, such as a map or list. Based on this information, users can quickly create specific repair plans.

[0903] Input: Repair policy, user emotion data

[0904] Output: Optimized information presentation and support for creating specific repair plans

[0905] This allows the entire system to work in harmony, providing users with fast, accurate information and support that takes their emotions into consideration.

[0906] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0907] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0908] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0909] [Third embodiment]

[0910] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0911] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0912] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0913] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0914] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0915] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0916] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0917] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0918] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0919] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0920] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0921] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0922] This invention provides a system that analyzes video data collected from a drive recorder and automatically identifies road damage and generates repair plans. The system's main components are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[0923] Program processing

[0924] 1. Image data collection

[0925] Device:

[0926] A drive recorder is installed in the vehicle and records video in real time as the vehicle travels along the road. The drive recorder acquires GPS information and adds location information and timestamps to the video. This video data is then sent to a server in real time via a 5G communication module.

[0927] 2. Receipt and storage of data

[0928] server:

[0929] The system receives video data sent from the device. The received data is verified for format and integrity, and if there are no problems, it is stored in a database. The stored data includes not only the video data itself, but also the corresponding location information and timestamp.

[0930] 3. Data Preprocessing

[0931] server:

[0932] A noise reduction filter is applied to the received video data. At this stage, the data is resized to a format that is optimal for the analysis model. Additionally, frames containing important information are extracted and unnecessary frames are removed. This preprocessing improves the accuracy and efficiency of the analysis.

[0933] 4. Analysis of image data

[0934] server:

[0935] The preprocessed data is then input into a generative AI model, which detects damage to the road surface and extracts attribute information such as its type, extent, and depth. The model is particularly effective at accurately detecting critical damage, such as large potholes and cracks.

[0936] 5. Damage Level Classification

[0937] server:

[0938] Based on the information of the detected damage, a classification algorithm is applied to assess the level of damage: the damage is classified as minor, caution, or severe, and stored in a database along with the corresponding location information.

[0939] 6. Repair Policy Generation

[0940] server:

[0941] Based on the classification results, the system automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs. This information allows road managers to create quick and accurate repair plans.

[0942] 7. Presentation of results

[0943] server:

[0944] An interface is generated to present the generated repair policy and information on damaged areas to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using a map display or list display.

[0945] User:

[0946] Through the provided interface, users (road administrators) can view detailed information about damaged areas and repair policies, which will enable them to prioritize repairs and develop specific repair plans.

[0947] Specific examples

[0948] For example, a vehicle traveling on a highway may capture a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10cm deep and 50cm wide. A classification algorithm then evaluates it as "severely damaged" and stores it in a database along with its location. Finally, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers via a user interface, allowing them to quickly develop specific repair plans.

[0949] In this way, the present invention provides a system that uses video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair policies, thereby reducing the burden on road administrators and realizing efficient road maintenance management.

[0950] The processing flow will be explained below.

[0951] Step 1:

[0952] Device:

[0953] The dashcam is installed in vehicles traveling on the road and records video in real time. The built-in GPS sensor periodically acquires location information while recording, adding a timestamp and location information to the video data. The video data is then sent to a server in real time using a 5G communication module.

[0954] Step 2:

[0955] server:

[0956] Receives video data sent from the device. The received data is verified for format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, corresponding location information, and timestamp.

[0957] Step 3:

[0958] server:

[0959] The stored video data is preprocessed by applying a noise reduction filter, resizing the video resolution to a format suitable for analysis, and extracting important frames and removing unnecessary ones.

[0960] Step 4:

[0961] server:

[0962] The preprocessed data is input into a generative AI model, which detects road damage (potholes, cracks, etc.) from the collected video data. At this stage, attribute information such as the type, extent, depth, and location of the damage is extracted.

[0963] Step 5:

[0964] server:

[0965] The damage level is assessed based on the extracted information on the damaged areas. A unique classification algorithm is applied to categorize the damage into several levels, such as minor, caution, and serious. The classification results and location information are stored in a database.

[0966] Step 6:

[0967] server:

[0968] Based on the classification results, an algorithm is applied to automatically generate an optimal repair policy. The repair policy includes the urgency of the repair, the recommended repair method (e.g., partial patching, full repaving), and the estimated repair cost. The generated repair policy is then compiled in the form of a report.

[0969] Step 7:

[0970] server:

[0971] An interface is generated to provide the generated repair policy and related information to the user. The interface is designed for ease of use, including map and list views, allowing users to quickly check the information they need.

[0972] Step 8:

[0973] User:

[0974] Users (road administrators) access the server through a web interface or a dedicated application. On the dashboard, they can view detailed information, such as a map of damaged areas and a list of damage levels. Based on the repair policy provided, they can quickly create specific repair plans.

[0975] In this way, the system of the present invention helps road managers to plan road repairs efficiently and accurately. The automated process shortens the lead time for repairs and optimizes repair costs.

[0976] Example 1

[0977] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0978] In modern transportation infrastructure, it is important to quickly and accurately detect road damage and develop appropriate repair plans. However, conventional methods have the drawback of requiring a large amount of manpower, time, and cost. Furthermore, there is a risk of overlooking road damage and the difficulty of prioritizing repairs. The present invention aims to solve these problems and provide an efficient and accurate road management method.

[0979] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0980] In this invention, the server includes a device for transmitting video data acquired from the drive recorder in real time, a device for saving the received video data, a device for preprocessing the saved video data, a generative model for analyzing the preprocessed data and detecting road damage, an algorithm for classifying the detected damage, a method for generating a repair policy based on the classification results, and an interface for presenting the generated repair policy to the user. This enables quick and accurate detection of road damage and efficient planning of repairs.

[0981] A "drive recorder" is a device that is installed in a vehicle and records video data while the vehicle is driving.

[0982] "Video data" refers to data including images and video information recorded by a drive recorder.

[0983] A "real-time transmission device" is a communication device for instantly transferring acquired video data to a remote server.

[0984] The "receiving device" is a part of the server that receives video data transmitted from a remote location.

[0985] The "storing device" is a storage device that stores the received video data in a database or the like so that it can be used later.

[0986] A "preprocessing device" is a device that converts received video data into a format suitable for analysis. Specifically, it performs processes such as noise removal and resolution adjustment.

[0987] The "generative model for analysis" is an AI model that identifies damaged areas on roads based on preprocessed video data and extracts their detailed attribute information.

[0988] The "classification algorithm" is an algorithm for evaluating and classifying the level of damage based on information about the detected damaged portion.

[0989] The "method for generating a repair policy" is a process for determining the optimum repair method, urgency, and estimated repair cost for the classified damaged areas.

[0990] An "interface" is a system that visually displays the generated repair policy and information on damaged areas, allowing users to easily access them.

[0991] This invention relates to a system that analyzes video data collected from a drive recorder and automatically generates road damage information and repair plans. The main components of the system are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[0992] The drive recorder is installed in the vehicle and records video in real time while driving. The drive recorder acquires GPS information and adds location information and timestamps to the video. Using a 5G communication module, this video data is sent to a server with low latency.

[0993] The server receives the video data and verifies the format and integrity of the received data. If there are no problems, the data is stored in a database. The stored data includes not only the video but also the corresponding location information and timestamp. Image processing libraries such as OpenCV are used to preprocess the video data, such as noise reduction and resizing.

[0994] The pre-processed data is then analyzed by a generative AI model, which uses deep learning frameworks such as TensorFlow or PyTorch to detect damage to the road surface and extract its type, extent, and depth, for example using a convolutional neural network (CNN).

[0995] The server then applies a classification algorithm based on the detected damage information, which assesses the level of damage based on parameters such as the area, depth, and width of the damage. The data is then classified into levels such as minor, caution, and severe and stored in a database.

[0996] Furthermore, the server automatically generates an optimal repair policy based on the classification results, which includes the urgency of repair, recommended repair methods, and estimated repair costs, enabling road managers to quickly develop repair plans.

[0997] The generated repair plan and damage location information are presented to the user through an interface that can be used in a web browser or mobile application, and the information is presented in an easy-to-understand manner using map and list views. Front-end frameworks such as React and Angular are used, and Node.js and Django are used for the back-end.

[0998] As a concrete example, a vehicle traveling on a highway captures a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10 cm deep and 50 cm wide. A classification algorithm evaluates it as "severely damaged" and stores it in a database along with its location information. Ultimately, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers through a user interface, allowing them to quickly develop specific repair plans.

[0999] An example of a prompt to be input into the generative AI model is, "Analyze the video data acquired while the vehicle is traveling and detect any newly occurring potholes or cracks. Also, measure the depth and width of the damaged areas and propose a repair plan."

[1000] In this way, the present invention provides a system that utilizes video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair plans, thereby achieving more efficient road management and reducing costs.

[1001] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1002] Step 1: Collecting image data

[1003] Device:

[1004] The dashcam is installed in a vehicle and records high-resolution video in real time as the vehicle travels along the road. The dashcam acquires GPS information and adds location information and timestamps to the video data. Using a 5G communication module, this video data is transmitted to a server with low latency.

[1005] Input: Real-time video and GPS information acquired while the vehicle is running

[1006] Data processing: Add location information and timestamp to each video frame

[1007] Output: Video data with additional information

[1008] Step 2: Receiving and saving video data

[1009] server:

[1010] The server receives the video data sent from the device. The format and integrity of the received data are verified, and if there are no problems, it is stored in a database. The stored data includes not only the video, but also the corresponding location information and timestamp.

[1011] Input: Video data with additional information sent from the device

[1012] Data processing: verifying data format and integrity (e.g., CRC check, format verification)

[1013] Output: Verified video data is stored in a database

[1014] Step 3: Preprocessing the data

[1015] server:

[1016] The server applies a noise reduction filter to the received video data. This process resizes the video to a format that is optimized for the analysis model. It also extracts frames that contain important information for analysis and removes unnecessary frames.

[1017] Input: Verified video data stored in a database

[1018] Data processing: noise removal (e.g., Gaussian filter application), resolution adjustment, extraction of important frames, deletion of unnecessary frames

[1019] Output: Pre-processed video data

[1020] Step 4: Analyzing the image data

[1021] server:

[1022] The server inputs the preprocessed data into a generative AI model to detect damage on the road surface, which uses deep learning frameworks such as TensorFlow and PyTorch to extract the type, extent, and depth of the damage.

[1023] Input: Preprocessed video data

[1024] Data calculation: Deep learning model for detecting damage and extracting attribute information

[1025] Output: Specific attribute information and location information of detected damage points

[1026] Step 5: Classify the level of damage

[1027] server:

[1028] The server then applies a classification algorithm to assess the level of damage detected: damage is classified as minor, minor, or severe, and these classifications are stored in a database.

[1029] Input: Attribute information and location information of the detected damage

[1030] Data processing: Assessing the level of corruption with classification algorithms

[1031] Output: Damage level assessment data and location information are saved in a database.

[1032] Step 6: Generate repair policies

[1033] server:

[1034] Based on the classification results, the server automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs.

[1035] Input: Damage level assessment data and location information

[1036] Data calculation: Urgency assessment, repair method proposal, repair cost estimate

[1037] Output: Integrated information including repair policy

[1038] Step 7: Presenting the results

[1039] server:

[1040] The server generates an interface to present the generated repair policy and damage location information to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using map and list views.

[1041] User:

[1042] The user (road administrator) can view detailed information about the damaged areas and repair policies through the provided interface.

[1043] Input: Integrated information including repair policy

[1044] Data calculation: Visualization of information (e.g., displaying potholes on a map, displaying repair policies in list format)

[1045] Output: A user-visible interface

[1046] (Application example 1)

[1047] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1048] Conventional road management systems have difficulty quickly and accurately detecting road damage and providing appropriate repair plans. In particular, autonomous vehicles require real-time road condition monitoring to ensure safety, making efficient road management essential.

[1049] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1050] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damaged areas, means for generating a repair policy based on the classification results, means for presenting the generated repair policy to a user, means for displaying information on the damaged areas on a map, and means for notifying the user of repairs to the damaged areas in real time, thereby enabling real-time road condition monitoring and rapid planning of repairs.

[1051] A "dashcam" is a device installed in a vehicle that records video footage while driving.

[1052] "Means for transmitting in real time" refers to a communication means for transferring video data to a receiving destination such as a server immediately without delay.

[1053] "Means for storing received video data" refers to a storage device or database for storing the received video data in an appropriate format.

[1054] The "means for pre-processing stored video data" is a function for performing initial processing such as noise removal and format adjustment on video data.

[1055] The "means for analyzing the pre-processed data and detecting road damage" is an algorithm or software for identifying damaged portions of the road from the pre-processed data using image analysis techniques.

[1056] The "means for classifying detected damage areas" refers to a classification algorithm for classifying and rating damage areas based on the type and severity of damage.

[1057] The "means for generating repair policies based on classification results" refers to algorithms or systems that automatically plan and propose optimal repair methods based on classified damage location information.

[1058] The "means for presenting the generated repair policy to the user" refers to a display interface or notification system for providing the user with the repair plan and information on the damaged area.

[1059] The "means for displaying information on damaged areas on a map" is a function that visually shows identified damaged areas on a map using a geographic information system (GIS).

[1060] "Means for providing real-time notification of damaged area repairs" refers to a communication and notification system that immediately notifies of the need for repairs when a damaged area is detected.

[1061] This invention is a system that uses video data acquired from a vehicle's onboard dashcam to detect road damage in real time and automatically generate and present repair plans. The system's main hardware components are a dashcam, GPS module, and communication module, while the software components include a server, database, generative AI model, and user interface.

[1062] System Overview

[1063] The server receives, stores, pre-processes, analyzes, and detects damages in real time, classifies them, and generates repair plans. This data is then provided to the user through a user interface, including map displays and notifications.

[1064] Hardware and software used

[1065] Drive recorder: Acquires video data and GPS information.

[1066] GPS module: Adds location information and timestamps to video data.

[1067] 5G communication module: Transmits video data to a server in real time.

[1068] Server: Receives, stores, preprocesses, analyzes, and generates repair policies for video data.

[1069] Database: Stores received video data and analysis results.

[1070] Generative AI model: Analyzes video data and detects damaged areas.

[1071] User interface: Displays broken areas and repair plans on a map and notifies users in real time.

[1072] Processing flow

[1073] The device (drive recorder) transmits video data captured while the vehicle is driving to a server in real time. The server stores the received data in a database, performs preprocessing, and uses a generative AI model to detect road damage. It then uses a classification algorithm to evaluate the importance of the damage and generates a repair policy. The generated repair policy and information on the damaged areas are provided to the user through a user interface.

[1074] Specific examples

[1075] For example, while an autonomous vehicle is traveling on a highway, its dashcam captures damaged areas. This video data is sent to a server in real time and stored. The server preprocesses the video data and uses a generative AI model to detect the damaged areas, then classifies and evaluates the information. For example, a pothole that is 10 cm deep and 50 cm wide is classified as "severe damage." The information is then stored in a database along with its location, and "urgent repairs" or "complete repaving" are recommended, with estimated repair costs calculated. This information is provided to the autonomous vehicle operator through a user interface, allowing a specific repair plan to be quickly developed.

[1076] Prompt Sentence Examples

[1077] An example of a prompt for a model for automatically detecting road breaks is:

[1078] "Please use the following data to generate a repair plan for a pothole discovered while driving on a highway. Please propose the optimal repair plan based on the dashcam video data, GPS information, and details of the damaged area (location, type of damage, and level of damage)."

[1079] In this way, the present invention provides a system that enables real-time road monitoring and rapid planning of repair plans, and supports safe operation of autonomous vehicles.

[1080] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1081] Step 1: Image data collection process The device (drive recorder) collects video data taken while the vehicle is driving. At the same time, it also obtains location information and timestamps from the GPS module and adds them to the video data. This creates spatiotemporal information for each video frame.

[1082] Input: Video data during driving, GPS location information, timestamp

[1083] Output: Video data with location information and timestamp added

[1084] Step 2: Sending video data The device sends the collected video data to the server in real time via the communication module (5G). Thanks to high-speed communication, the data is immediately transferred to the server.

[1085] Input: Video data with location information and timestamps

[1086] Output: Video data sent to the server

[1087] Step 3: Receiving and storing data The server receives the video data sent from the device, verifies the data integrity and format, and then stores it in the database.

[1088] Input: Video data sent from the device

[1089] Output: Video data stored in a database

[1090] Step 4: Data Preprocessing The server applies a noise reduction filter to the stored video data, then resizes it to a format suitable for analysis, extracts frames containing important information, and removes unnecessary frames.

[1091] Input: Video data stored in the database

[1092] Output: Denoised, resized and frame extracted data

[1093] Step 5: Image data analysis. The server inputs the preprocessed data into the generative AI model. The model performs image analysis and automatically detects road damage. Attribute information such as the type, extent, and depth of the detected damage is acquired.

[1094] Input: Preprocessed video data

[1095] Output: Attribute information of the damaged area (type, range, depth)

[1096] Step 6: Classification of Damage Level Based on the acquired attribute information of the damaged area, the server evaluates the damage level using a classification algorithm. Damage areas are classified into levels such as minor, caution, and serious.

[1097] Input: Attribute information of the damaged area

[1098] Output: Classified damage levels

[1099] Step 7: Generation of repair policy The server automatically generates a repair policy based on the classification results. The generated repair policy includes the urgency of repair, recommended repair methods, and estimated repair costs.

[1100] Input: Classified damage level

[1101] Output: Repair policy (urgency, repair method, estimated cost)

[1102] Step 8: Presenting the results The server presents the generated repair policy and information on the damaged areas to the user through a user interface. This information is displayed on a map so that the user can visually check it. Repair notifications are also sent in real time.

[1103] Input: Repair policy, information on damaged areas

[1104] Output: Information presentation and repair notification through user interface

[1105] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1106] This invention combines an emotion engine with a system that analyzes video data collected from a drive recorder and automatically generates road damage locations and repair plans, making it possible to present repair plans that are more suited to the user.The main components of this system are a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[1107] Program processing

[1108] 1. Image data collection

[1109] Device:

[1110] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[1111] 2. Receipt and storage of data

[1112] server:

[1113] The system receives video data sent from the device and verifies its format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, location information, and timestamp.

[1114] 3. Data Preprocessing

[1115] server:

[1116] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[1117] 4. Analysis of image data

[1118] server:

[1119] The preprocessed video data is input into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage.

[1120] 5. Damage Level Classification

[1121] server:

[1122] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or severe. This classification result and location information are stored in a database.

[1123] 6. Repair Policy Generation

[1124] server:

[1125] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[1126] 7. Leveraging Emotional Engines

[1127] server:

[1128] When an interaction with a user (road administrator) occurs through the user interface, the emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.). Based on the recognized emotions, the content of the repair policy to be presented and the way it is displayed are adjusted.

[1129] 8. Presentation of results

[1130] server:

[1131] The emotional data obtained from the emotion engine is analyzed to optimize the way repair plans are presented. For example, if the user is feeling anxious, detailed explanations and additional data can be provided to reassure them. The generated repair plans are then displayed in the user interface in an easy-to-understand format, such as a map or list view.

[1132] User:

[1133] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[1134] Specific examples

[1135] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10 cm deep and 50 cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[1136] In this way, the present invention realizes a system that supports road administrators in formulating road repair plans efficiently and accurately, and provides appropriate information while taking into consideration the feelings of users, thereby shortening the lead time until repairs and optimizing repair costs.

[1137] The processing flow will be explained below.

[1138] Step 1:

[1139] Device:

[1140] The drive recorder is installed in the vehicle and records video in real time while the vehicle is driving on the road. The built-in GPS sensor adds location information and a timestamp to the video data. The video data is then sent to a server in real time via a 5G communication module.

[1141] Step 2:

[1142] server:

[1143] Receives video data sent from the device. Automatically verifies the format and integrity of the received data. If there are no problems, stores the received data in storage. This includes the video data, location information, and timestamp.

[1144] Step 3:

[1145] server:

[1146] The stored video data is subjected to a noise reduction filter, the video resolution is resized to a format suitable for analysis, and pre-processing is performed to extract important frames and remove unnecessary frames.

[1147] Step 4:

[1148] server:

[1149] The preprocessed data is input into a generative AI model, which analyzes the video data to detect road damage (potholes, cracks, etc.) and extracts information on the type, extent, depth, and location of the damage.

[1150] Step 5:

[1151] server:

[1152] Based on the information on the detected damaged areas, a unique classification algorithm is applied to evaluate the level of damage, classifying the damaged areas into levels such as minor, caution, or serious, and saving the results in a database.

[1153] Step 6:

[1154] server:

[1155] Based on the classification results, an optimal repair plan is automatically generated. This repair plan includes the urgency of repair, recommended repair methods, estimated repair costs, etc. The repair plan is then created in report format.

[1156] Step 7:

[1157] server:

[1158] The system prepares to present a repair policy to the user through the user interface. When presenting the policy, the emotion engine analyzes the user's facial expressions and voice to recognize emotions in real time. Based on the recognized emotions (e.g., relief, anxiety, surprise, etc.), the way the repair policy is presented is adjusted.

[1159] Step 8:

[1160] server:

[1161] Based on feedback from the emotion engine, the repair plan presented to the user is optimized. For example, if the user is feeling anxious, the plan will be adjusted to provide detailed explanations and additional materials to reassure the user. The repair plan is displayed in the user interface in an easy-to-use format such as a map or list.

[1162] Step 9:

[1163] User:

[1164] The user (road administrator) can view detailed information about the damaged areas and repair plans through a web interface or a dedicated application. The information presented is adjusted by the emotion engine, allowing the user to quickly and appropriately obtain the information they need. Based on this, a specific repair plan can be quickly formulated.

[1165] Through the above processing flow, the system of the present invention helps road managers to create repair plans efficiently and accurately. In addition, the emotion engine allows the system to present information that takes into account the user's emotions, thereby improving user satisfaction.

[1166] Example 2

[1167] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1168] Although technology already exists to analyze road footage acquired from dashcams and detect damaged areas, there has not previously been any technology that provides a method for presenting repair plans that take the user's emotions into consideration. As a result, repair plans are not presented at the optimal time and in the optimal way based on the user's emotions, which leads to user anxiety and a lack of understanding.

[1169] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1170] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damage, means for generating a repair plan based on the classification results, means for presenting the generated repair plan to the user, and means for recognizing the user's emotions and adjusting the content and method of presenting the repair plan based on the emotions. This makes it possible to present a repair plan that takes the user's emotions into consideration, and to provide information with the most appropriate timing and content for the user.

[1171] A "dashcam" is a device installed in a vehicle that records video in real time while driving.

[1172] "Video data" refers to data that includes video frames recorded by a drive recorder and additional information such as associated location information and timestamps.

[1173] "Means for transmitting in real time" refers to the communication module and protocol for transferring video data obtained from the drive recorder to the server without delay.

[1174] "Means for storage" refers to a database system or storage device for stably and persistently storing received video data.

[1175] "Preprocessing means" refers to tools and algorithms that process stored video data, such as removing noise, adjusting resolution, and extracting important frames.

[1176] "Means of analysis" refers to models or software that identify road damage areas and extract attribute information based on preprocessed data.

[1177] "Means for classifying damage points" refers to an algorithm that analyzes the severity and extent of the detected damage points and groups them according to different classification criteria.

[1178] "Means for generating repair policies" refers to a system that automatically calculates the repair method, urgency, estimated cost, etc. for damaged areas based on the classification results, and creates an appropriate repair plan.

[1179] "Presentation means" refers to a system or software for visually displaying the generated repair policy through a user interface in an easy-to-understand manner.

[1180] "Means for recognizing emotions" refers to engines or algorithms that analyze a user's facial expressions and voice on the user interface and identify emotions in real time.

[1181] "Adjustment means" refers to a mechanism that dynamically changes the content and method of presenting repair policies based on the recognized user emotions.

[1182] This invention is a system that analyzes video data collected from a drive recorder, detects damaged areas on roads, and automatically generates repair plans. By combining this system with an emotion engine, it is possible to present appropriate repair plans according to the user's emotions.

[1183] The system mainly consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[1184] Drive recorder and communication module

[1185] Device:

[1186] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[1187] Servers and Databases

[1188] server:

[1189] The server stores the received video data, which undergoes pre-processing to apply noise reduction filters and resize the resolution to a format optimized for analysis, improving the quality of the video data and making it easier to analyze.

[1190] Analysis using generative AI models

[1191] server:

[1192] The preprocessed data is input into a generative AI model. This model detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage. It then applies a classification algorithm based on the damage information to classify the damage level as minor, caution, or severe. The classification results and location information are stored in a database.

[1193] Generation and presentation of repair policies

[1194] server:

[1195] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[1196] Use of emotion engine

[1197] server:

[1198] The emotion engine recognizes the user's emotions in real time. It analyzes the user's facial expressions and voice to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and presentation of the repair plan presented.

[1199] User Interface

[1200] User:

[1201] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[1202] Specific examples

[1203] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10cm deep and 50cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[1204] Prompt Sentence Examples

[1205] Below are some example prompts to input to a generative AI model:

[1206] "Analyze the following video data to detect road damage. Damage data should include attribute information such as type, extent, depth, and location."

[1207] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1208] Step 1:

[1209] Image data collection

[1210] Device:

[1211] A drive recorder is installed in the vehicle. While the vehicle is driving on the road, the drive recorder records video in real time. The built-in GPS sensor acquires the current location information and adds a timestamp and location information to the video data.

[1212] Specific behavior:

[1213] The dashcam starts up and the camera starts recording video, and this video data is the input.

[1214] The GPS sensor obtains the current latitude and longitude information.

[1215] The acquired location information and timestamp are added to the video frame.

[1216] This becomes the video data as output.

[1217] Step 2:

[1218] Sending and Receiving Data

[1219] Devices and Servers:

[1220] Using a 5G communication module, the video data collected in real time is transmitted to a server, which receives and stores the data.

[1221] Specific behavior:

[1222] The video data is transmitted through the 5G communication module, and this transmitted data is the input.

[1223] The server listens to receive data.

[1224] Check the format and integrity of the data received.

[1225] The validated data is saved in a database. The saved data is the output.

[1226] Step 3:

[1227] Data Preprocessing

[1228] server:

[1229] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[1230] Specific behavior:

[1231] The saved video data is read out. This data is the input.

[1232] Improve the quality of your footage with noise reduction filters.

[1233] Resize the video resolution to the optimal format for AI analysis.

[1234] Frame analysis is performed to select and extract important frames.

[1235] Optimize data volume by removing unnecessary frames.

[1236] After these operations, the preprocessed data is the output.

[1237] Step 4:

[1238] Image data analysis

[1239] server:

[1240] The preprocessed data is input into a generative AI model, which detects road damage from the video data and extracts attribute information such as its type, extent, depth, and location.

[1241] Specific behavior:

[1242] The preprocessed data is fed into the generative AI model. This data is the input.

[1243] The AI ​​model detects the damaged areas.

[1244] Attribute information such as the type, extent, and depth of the damage is extracted.

[1245] The extracted information is the output.

[1246] Step 5:

[1247] Damage Level Classification

[1248] server:

[1249] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or serious. This classification result and location information are stored in a database.

[1250] Specific behavior:

[1251] The extracted damage information is fed into a classification algorithm, and this data is the input.

[1252] An algorithm classifies the level of damage as either "minor," "cautionary," or "severe."

[1253] The classification results and location information are saved in a database. The saved data is the output.

[1254] Step 6:

[1255] Generating repair policies

[1256] server:

[1257] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[1258] Specific behavior:

[1259] Based on the classification results, a repair policy generation algorithm is applied. This data is the input.

[1260] The algorithm calculates the urgency of repairs, recommended repair methods, and estimated repair costs.

[1261] The repair policy is compiled in a report format. This is the output report.

[1262] Step 7:

[1263] Use of emotion engine

[1264] server:

[1265] The system uses an emotion engine to recognize user emotions in real time. It analyzes the user's facial expressions and voice on the user interface to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and display method of the repair policy presented.

[1266] Specific behavior:

[1267] The user's facial expressions and voice data are input to the emotion engine. This data is the input.

[1268] The emotion engine analyzes and identifies the user's emotions.

[1269] Based on the recognized emotion data, the presentation method and content of the repair policy are adjusted, and the adjusted presentation method is the output.

[1270] Step 8:

[1271] Presentation of results

[1272] Servers and Users:

[1273] The generated repair policy is visually presented to the user through a user interface, and the user can then create a specific repair plan based on this information.

[1274] Specific behavior:

[1275] The repair policy adjusted by the emotion engine is displayed on the user interface, and the repair policy report is the input.

[1276] The user views the repair policy through a web interface or a dedicated application.

[1277] The user quickly creates a repair plan based on the information, and the repair plan with the corrective measures is the output.

[1278] (Application example 2)

[1279] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1280] Conventional systems for detecting road damage and generating repair plans have the problem of being unable to respond quickly because it takes time to analyze video data and generate repair plans.In addition, they do not provide information that takes into account the situation and emotions of each user, so they are unable to provide users with a sufficient sense of security.

[1281] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1282] In this invention, the server includes a means for transmitting video data acquired from the drive recorder in real time, a means for saving the received video data, and a means for preprocessing the saved video data. This enables real-time collection and saving of video data. The server also includes a means for analyzing the preprocessed data to detect road damage, a means for classifying the detected damage, a means for generating a repair plan based on the classification results, and a means for optimizing and presenting the generated repair plan in response to the user's emotions. This enables information to be provided quickly and in consideration of the user's emotions.

[1283] A "drive recorder" is a device that is installed in a vehicle and records video and audio data in real time while the vehicle is driving.

[1284] "Means for receiving" refers to a device or method for receiving signals or data transmitted from an external source and storing or processing them.

[1285] The term "storing means" refers to a device or method that stores received or generated data in a storage device and keeps it in a reusable state as needed.

[1286] "Preprocessing means" refers to a device or method that performs a series of initial processes to improve the efficiency of data analysis, including noise removal and resolution adjustment.

[1287] The "means for analyzing and detecting road damage" refers to a device or method that uses a specific algorithm to find damaged portions of the road from the pre-processed data.

[1288] The "classifying means" refers to a device or method for grouping detected damaged areas based on specific criteria, including classification by the degree or type of damage.

[1289] The "means for generating a repair policy" is a device or method that automatically determines the optimal repair method and priorities based on the classification results.

[1290] "Means for optimizing and presenting according to emotions" refers to a device or method that analyzes the user's emotional state in real time, adjusts the display content and display method based on that information, and reduces stress and anxiety.

[1291] An "emotion engine" is software or hardware that analyzes emotions from a user's facial expressions, voice, etc., and detects specific emotions.

[1292] This invention is a system for autonomous vehicles that detects road damage in real time, generates appropriate repair policies, and presents optimized information according to the user's emotions. The system consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[1293] 1. System configuration and functions

[1294] 1.1 Drive recorder

[1295] A dashcam is a device that records video data in real time while a vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. Furthermore, using a 5G communication module, this video data is sent to a server without delay.

[1296] 1.2 Server

[1297] The server stores the received video data and verifies its format and completeness. The stored data includes video data, location information, and timestamps. During the data preprocessing stage, the server applies a noise reduction filter and adjusts the resolution to create an optimal format for analysis. Next, a generative AI model is used to detect road damage from the preprocessed video data. The detection results are extracted as attribute information, such as the type and extent of the damage, and location information, and the damage level is classified based on this information.

[1298] 1.3 Generating Repair Policies

[1299] The server automatically generates an optimal repair policy based on the classification results. This repair policy includes the urgency of repairs, recommended repair methods, and estimated repair costs, and is generated in the form of a report.

[1300] 1.4 Emotion Engine

[1301] When an interaction with the user (road administrator) occurs through the user interface, an emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions such as "relief," "anxiety," and "surprise." Based on this emotion, the content of the repair policy presented and the way it is displayed can be adjusted. For example, if the user feels anxious, detailed explanations and additional data can be provided to reassure them.

[1302] 1.5 Presentation of results

[1303] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair plans are presented. For example, if the user is feeling anxious, it adds more detailed explanations and provides information that reassures the user. The generated repair plans are displayed in an easy-to-understand format on the user interface, such as a map or list view. Through this interface, the user can quickly check the repair plans and detailed information about the damaged areas, and then quickly create a specific repair plan.

[1304] 2. Specific examples

[1305] For example, consider the case where a large pothole has newly appeared on a road. The dashcam installed in an autonomous vehicle captures the incident, and the video data is sent to a server in real time. The server preprocesses the data and detects the pothole using a generative AI model. The resulting pothole is evaluated as 10 cm deep and 50 cm wide and classified as severely damaged. Repair options of "urgent repair" and "complete repaving" are recommended, and an estimated repair cost is calculated. When this information is presented to the user through the user interface, an emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, the system adjusts its response by providing additional explanations and data, presenting more detailed information to reassure the user. In this way, the system helps road managers plan repairs efficiently and accurately.

[1306] 3. Examples of prompts

[1307] Hardware used: Dashcam, GPS module, 5G communication module, server, user interface device

[1308] Software used: OpenCV, TensorFlow, EmotionRecognizer, and GPS module libraries

[1309] Examples of specific prompts include the following:

[1310] Prompt: "Real-time transmission of video data from dashcams, detection of road damage, and information display using emotion recognition."

[1311] As a result, the present invention has a function in which the entire system operates in cooperation with one another and can provide information in a form that is easy for the user to understand and that takes into consideration the user's emotions.

[1312] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1313] Step 1: Collecting video data from the dashcam

[1314] The device's drive recorder records video data in real time while the vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. The collected video data is then sent to a server via a 5G communication module.

[1315] Input: Video data captured while the vehicle is moving, GPS location information, and timestamp

[1316] Output: Real-time video data, location information, and time-stamped data sent to the server

[1317] Step 2: Receiving and storing data

[1318] The server receives the video data sent from the device in real time and verifies its format and integrity. If there are no problems, the data is stored in a database. The data includes the video data, location information, and timestamp.

[1319] Input: Real-time video data, location information, timestamp

[1320] Output: Validated data stored in a database

[1321] Step 3: Preprocessing the data

[1322] The server applies a noise reduction filter to the video data stored in the database, then adjusts the resolution to make it optimal for analysis, and preprocesses the data by extracting important frames and removing unnecessary ones.

[1323] Input: Video data stored in the database, location information, timestamp

[1324] Output: Preprocessed data with denoising and resolution adjustment

[1325] Step 4: Analyze the image data and detect any damage

[1326] The server inputs the preprocessed video data into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type and extent of the damage, as well as location information.

[1327] Input: Preprocessed video data

[1328] Output: Damage detection results (damage type, range, location information)

[1329] Step 5: Classify the level of damage

[1330] The server applies a classification algorithm based on the location of the damage and classifies the damage as minor, minor, or severe. This classification and location information are then stored in a database.

[1331] Input: Damage detection results (damage type, extent, location information)

[1332] Output: Classification result (damage level), location information

[1333] Step 6: Generate repair policies

[1334] The server automatically generates an optimal repair policy based on the classification results. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is saved in a database in report format.

[1335] Input: Classification result (damage level), location information

[1336] Output: Repair policy (urgency of repair, recommended repair method, estimated repair cost)

[1337] Step 7: Leverage your emotional engine

[1338] The server uses an emotion engine to recognize the user's emotions in real time when an interaction occurs with the user through the user interface. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.).

[1339] Input: User facial expressions and voice data obtained via the user interface

[1340] Output: User's emotional data (relief, anxiety, surprise, etc.)

[1341] Step 8: Presenting the results

[1342] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair policies are presented. The generated repair policies are displayed on the user interface in an easy-to-understand format, such as a map or list. Based on this information, users can quickly create specific repair plans.

[1343] Input: Repair policy, user emotion data

[1344] Output: Optimized information presentation and support for creating specific repair plans

[1345] This allows the entire system to work in harmony, providing users with fast, accurate information and support that takes their emotions into consideration.

[1346] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1347] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1348] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1349] [Fourth embodiment]

[1350] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1351] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1352] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1353] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1354] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1355] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1356] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1357] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1358] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1359] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1360] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1361] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1362] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1363] This invention provides a system that analyzes video data collected from a drive recorder and automatically identifies road damage and generates repair plans. The system's main components are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[1364] Program processing

[1365] 1. Image data collection

[1366] Device:

[1367] A drive recorder is installed in the vehicle and records video in real time as the vehicle travels along the road. The drive recorder acquires GPS information and adds location information and timestamps to the video. This video data is then sent to a server in real time via a 5G communication module.

[1368] 2. Receipt and storage of data

[1369] server:

[1370] The system receives video data sent from the device. The received data is verified for format and integrity, and if there are no problems, it is stored in a database. The stored data includes not only the video data itself, but also the corresponding location information and timestamp.

[1371] 3. Data Preprocessing

[1372] server:

[1373] A noise reduction filter is applied to the received video data. At this stage, the data is resized to a format that is optimal for the analysis model. Additionally, frames containing important information are extracted and unnecessary frames are removed. This preprocessing improves the accuracy and efficiency of the analysis.

[1374] 4. Analysis of image data

[1375] server:

[1376] The preprocessed data is then input into a generative AI model, which detects damage to the road surface and extracts attribute information such as its type, extent, and depth. The model is particularly effective at accurately detecting critical damage, such as large potholes and cracks.

[1377] 5. Damage Level Classification

[1378] server:

[1379] Based on the information of the detected damage, a classification algorithm is applied to assess the level of damage: the damage is classified as minor, caution, or severe, and stored in a database along with the corresponding location information.

[1380] 6. Repair Policy Generation

[1381] server:

[1382] Based on the classification results, the system automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs. This information allows road managers to quickly and accurately develop repair plans.

[1383] 7. Presentation of results

[1384] server:

[1385] An interface is generated to present the generated repair policy and information on damaged areas to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using a map display or list display.

[1386] User:

[1387] Through the provided interface, users (road administrators) can view detailed information about damaged areas and repair policies, which will enable them to prioritize repairs and develop specific repair plans.

[1388] Specific examples

[1389] For example, a vehicle traveling on a highway may capture a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10cm deep and 50cm wide. A classification algorithm then evaluates it as "severely damaged" and stores it in a database along with its location. Finally, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers via a user interface, allowing them to quickly develop specific repair plans.

[1390] In this way, the present invention provides a system that uses video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair policies, thereby reducing the burden on road administrators and realizing efficient road maintenance management.

[1391] The processing flow will be explained below.

[1392] Step 1:

[1393] Device:

[1394] The dashcam is installed in vehicles traveling on the road and records video in real time. The built-in GPS sensor periodically acquires location information while recording, adding a timestamp and location information to the video data. The video data is then sent to a server in real time using a 5G communication module.

[1395] Step 2:

[1396] server:

[1397] Receives video data sent from the device. The received data is verified for format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, corresponding location information, and timestamp.

[1398] Step 3:

[1399] server:

[1400] The stored video data is preprocessed by applying a noise reduction filter, resizing the video resolution to a format suitable for analysis, and extracting important frames and removing unnecessary ones.

[1401] Step 4:

[1402] server:

[1403] The preprocessed data is input into a generative AI model, which detects road damage (potholes, cracks, etc.) from the collected video data. At this stage, attribute information such as the type, extent, depth, and location of the damage is extracted.

[1404] Step 5:

[1405] server:

[1406] The damage level is assessed based on the extracted information on the damaged areas. A unique classification algorithm is applied to categorize the damage into several levels, such as minor, caution, and serious. The classification results and location information are stored in a database.

[1407] Step 6:

[1408] server:

[1409] Based on the classification results, an algorithm is applied to automatically generate an optimal repair policy. The repair policy includes the urgency of the repair, the recommended repair method (e.g., partial patching, full repaving), and the estimated repair cost. The generated repair policy is then compiled in the form of a report.

[1410] Step 7:

[1411] server:

[1412] An interface is generated to provide the generated repair policy and related information to the user. The interface is designed for ease of use, including map and list views, allowing users to quickly check the information they need.

[1413] Step 8:

[1414] User:

[1415] Users (road administrators) access the server through a web interface or a dedicated application. On the dashboard, they can view detailed information, such as a map of damaged areas and a list of damage levels. Based on the repair policy provided, they can quickly create specific repair plans.

[1416] In this way, the system of the present invention helps road managers to plan road repairs efficiently and accurately. The automated process shortens the lead time for repairs and optimizes repair costs.

[1417] Example 1

[1418] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1419] In modern transportation infrastructure, it is important to quickly and accurately detect road damage and develop appropriate repair plans. However, conventional methods have the drawback of requiring a large amount of manpower, time, and cost. Furthermore, there is a risk of overlooking road damage and the difficulty of prioritizing repairs. The present invention aims to solve these problems and provide an efficient and accurate road management method.

[1420] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1421] In this invention, the server includes a device for transmitting video data acquired from the drive recorder in real time, a device for saving the received video data, a device for preprocessing the saved video data, a generative model for analyzing the preprocessed data and detecting road damage, an algorithm for classifying the detected damage, a method for generating a repair policy based on the classification results, and an interface for presenting the generated repair policy to the user. This enables quick and accurate detection of road damage and efficient planning of repairs.

[1422] A "drive recorder" is a device that is installed in a vehicle and records video data while the vehicle is driving.

[1423] "Video data" refers to data including images and video information recorded by a drive recorder.

[1424] A "real-time transmission device" is a communication device for instantly transferring acquired video data to a remote server.

[1425] The "receiving device" is a part of the server that receives video data transmitted from a remote location.

[1426] The "storing device" is a storage device that stores the received video data in a database or the like so that it can be used later.

[1427] A "preprocessing device" is a device that converts received video data into a format suitable for analysis. Specifically, it performs processes such as noise removal and resolution adjustment.

[1428] The "generative model for analysis" is an AI model that identifies damaged areas on roads based on preprocessed video data and extracts their detailed attribute information.

[1429] The "classification algorithm" is an algorithm for evaluating and classifying the level of damage based on information about the detected damaged portion.

[1430] The "method for generating a repair policy" is a process for determining the optimum repair method, urgency, and estimated repair cost for the classified damaged areas.

[1431] An "interface" is a system that visually displays the generated repair policy and information on damaged areas, allowing users to easily access them.

[1432] This invention relates to a system that analyzes video data collected from a drive recorder and automatically generates road damage information and repair plans. The main components of the system are a drive recorder, a communication module, a server, a database, a generative AI model, and a user interface.

[1433] The drive recorder is installed in the vehicle and records video in real time while driving. The drive recorder acquires GPS information and adds location information and timestamps to the video. Using a 5G communication module, this video data is sent to a server with low latency.

[1434] The server receives the video data and verifies the format and integrity of the received data. If there are no problems, the data is stored in a database. The stored data includes not only the video but also the corresponding location information and timestamp. Image processing libraries such as OpenCV are used to preprocess the video data, such as noise reduction and resizing.

[1435] The pre-processed data is then analyzed by a generative AI model, which uses deep learning frameworks such as TensorFlow or PyTorch to detect damage to the road surface and extract its type, extent, and depth, for example using a convolutional neural network (CNN).

[1436] The server then applies a classification algorithm based on the detected damage information, which assesses the level of damage based on parameters such as the area, depth, and width of the damage. The data is then classified into levels such as minor, caution, and severe and stored in a database.

[1437] Furthermore, the server automatically generates an optimal repair policy based on the classification results, which includes the urgency of repair, recommended repair methods, and estimated repair costs, enabling road managers to quickly develop repair plans.

[1438] The generated repair plan and damage location information are presented to the user through an interface that can be used in a web browser or mobile application, and the information is presented in an easy-to-understand manner using map and list views. Front-end frameworks such as React and Angular are used, and Node.js and Django are used for the back-end.

[1439] As a concrete example, a vehicle traveling on a highway captures a newly formed large pothole captured by its dashcam. This video data is sent to a server in real time and stored. After preprocessing, the server uses a generative AI model to detect the pothole and extract attribute information, such as a pothole 10 cm deep and 50 cm wide. A classification algorithm evaluates it as "severely damaged" and stores it in a database along with its location information. Ultimately, "urgent repairs" and "complete repaving" are recommended, and the estimated repair costs are calculated. This information is provided to road managers through a user interface, allowing them to quickly develop specific repair plans.

[1440] An example of a prompt to be input into the generative AI model is, "Analyze the video data acquired while the vehicle is traveling and detect any newly occurring potholes or cracks. Also, measure the depth and width of the damaged areas and propose a repair plan."

[1441] In this way, the present invention provides a system that utilizes video data collected from drive recorders to quickly and accurately detect road damage and automatically generate optimal repair policies, thereby improving the efficiency of road management and reducing costs.

[1442] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1443] Step 1: Collecting image data

[1444] Device:

[1445] The dashcam is installed in a vehicle and records high-resolution video in real time as the vehicle travels along the road. The dashcam acquires GPS information and adds location information and timestamps to the video data. Using a 5G communication module, this video data is transmitted to a server with low latency.

[1446] Input: Real-time video and GPS information acquired while the vehicle is running

[1447] Data processing: Add location information and timestamp to each video frame

[1448] Output: Video data with additional information

[1449] Step 2: Receiving and saving video data

[1450] server:

[1451] The server receives the video data sent from the device. The format and integrity of the received data are verified, and if there are no problems, it is stored in a database. The stored data includes not only the video, but also the corresponding location information and timestamp.

[1452] Input: Video data with additional information sent from the device

[1453] Data processing: verifying data format and integrity (e.g., CRC check, format verification)

[1454] Output: Verified video data is stored in a database

[1455] Step 3: Preprocessing the data

[1456] server:

[1457] The server applies a noise reduction filter to the received video data. This process resizes the video to a format that is optimized for the analysis model. It also extracts frames that contain important information for analysis and removes unnecessary frames.

[1458] Input: Verified video data stored in a database

[1459] Data processing: noise removal (e.g., Gaussian filter application), resolution adjustment, extraction of important frames, deletion of unnecessary frames

[1460] Output: Pre-processed video data

[1461] Step 4: Analyzing the image data

[1462] server:

[1463] The server inputs the preprocessed data into a generative AI model to detect damage on the road surface, which uses deep learning frameworks such as TensorFlow and PyTorch to extract the type, extent, and depth of the damage.

[1464] Input: Preprocessed video data

[1465] Data calculation: Deep learning model for detecting damage and extracting attribute information

[1466] Output: Specific attribute information and location information of detected damage points

[1467] Step 5: Classify the level of damage

[1468] server:

[1469] The server then applies a classification algorithm to assess the level of damage detected: damage is classified as minor, minor, or severe, and these classifications are stored in a database.

[1470] Input: Attribute information and location information of the detected damage

[1471] Data processing: Assessing the level of corruption with classification algorithms

[1472] Output: Damage level assessment data and location information are saved in a database.

[1473] Step 6: Generate repair policies

[1474] server:

[1475] Based on the classification results, the server automatically generates an optimal repair policy, which includes the urgency of repair, recommended repair methods, and estimated repair costs.

[1476] Input: Damage level assessment data and location information

[1477] Data calculation: Urgency assessment, repair method proposal, repair cost estimate

[1478] Output: Integrated information including repair policy

[1479] Step 7: Presenting the results

[1480] server:

[1481] The server generates an interface to present the generated repair policy and damage location information to the user. The interface can be used through a web browser or a mobile application, and presents the information in an easy-to-understand manner using map and list views.

[1482] User:

[1483] The user (road administrator) can view detailed information about the damaged areas and repair policies through the provided interface.

[1484] Input: Integrated information including repair policy

[1485] Data calculation: Visualization of information (e.g., displaying potholes on a map, displaying repair policies in list format)

[1486] Output: A user-visible interface

[1487] (Application example 1)

[1488] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1489] Conventional road management systems have difficulty quickly and accurately detecting road damage and providing appropriate repair plans. In particular, autonomous vehicles require real-time road condition monitoring to ensure safety, making efficient road management essential.

[1490] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1491] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damaged areas, means for generating a repair policy based on the classification results, means for presenting the generated repair policy to a user, means for displaying information on the damaged areas on a map, and means for notifying the user of repairs to the damaged areas in real time, thereby enabling real-time road condition monitoring and rapid planning of repairs.

[1492] A "dashcam" is a device installed in a vehicle that records video footage while driving.

[1493] "Means for transmitting in real time" refers to a communication means for transferring video data to a receiving destination such as a server immediately without delay.

[1494] "Means for storing received video data" refers to a storage device or database for storing the received video data in an appropriate format.

[1495] The "means for pre-processing stored video data" is a function for performing initial processing such as noise removal and format adjustment on video data.

[1496] The "means for analyzing the pre-processed data and detecting road damage" is an algorithm or software for identifying damaged portions of the road from the pre-processed data using image analysis techniques.

[1497] The "means for classifying detected damage areas" refers to a classification algorithm for classifying and rating damage areas based on the type and severity of damage.

[1498] The "means for generating repair policies based on classification results" refers to algorithms or systems that automatically plan and propose optimal repair methods based on classified damage location information.

[1499] The "means for presenting the generated repair policy to the user" refers to a display interface or notification system for providing the user with the repair plan and information on the damaged area.

[1500] The "means for displaying information on damaged areas on a map" is a function that visually shows identified damaged areas on a map using a geographic information system (GIS).

[1501] "Means for providing real-time notification of damaged area repairs" refers to a communication and notification system that immediately notifies of the need for repairs when a damaged area is detected.

[1502] This invention is a system that uses video data acquired from a vehicle's onboard dashcam to detect road damage in real time and automatically generate and present repair plans. The system's main hardware components are a dashcam, GPS module, and communication module, while the software components include a server, database, generative AI model, and user interface.

[1503] System Overview

[1504] The server receives, stores, pre-processes, analyzes, and detects damages in real time, classifies them, and generates repair plans. This data is then provided to the user through a user interface, including map displays and notifications.

[1505] Hardware and software used

[1506] Drive recorder: Acquires video data and GPS information.

[1507] GPS module: Adds location information and timestamps to video data.

[1508] 5G communication module: Transmits video data to a server in real time.

[1509] Server: Receives, stores, preprocesses, analyzes, and generates repair policies for video data.

[1510] Database: Stores received video data and analysis results.

[1511] Generative AI model: Analyzes video data and detects damaged areas.

[1512] User interface: Displays broken areas and repair plans on a map and notifies users in real time.

[1513] Processing flow

[1514] The device (drive recorder) transmits video data captured while the vehicle is driving to a server in real time. The server stores the received data in a database, performs preprocessing, and uses a generative AI model to detect road damage. It then uses a classification algorithm to evaluate the importance of the damage and generates a repair policy. The generated repair policy and information on the damaged areas are provided to the user through a user interface.

[1515] Specific examples

[1516] For example, while an autonomous vehicle is traveling on a highway, its dashcam captures damaged areas. This video data is sent to a server in real time and stored. The server preprocesses the video data and uses a generative AI model to detect the damaged areas, then classifies and evaluates the information. For example, a pothole that is 10 cm deep and 50 cm wide is classified as "severe damage." The information is then stored in a database along with its location, and "urgent repairs" or "complete repaving" are recommended, with estimated repair costs calculated. This information is provided to the autonomous vehicle operator through a user interface, allowing a specific repair plan to be quickly developed.

[1517] Prompt Sentence Examples

[1518] An example of a prompt for a model for automatically detecting road breaks is:

[1519] "Please use the following data to generate a repair plan for a pothole discovered while driving on a highway. Please propose the optimal repair plan based on the dashcam video data, GPS information, and details of the damaged area (location, type of damage, and level of damage)."

[1520] In this way, the present invention provides a system that enables real-time road monitoring and rapid planning of repair plans, and supports safe operation of autonomous vehicles.

[1521] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1522] Step 1: Image data collection process The device (drive recorder) collects video data taken while the vehicle is driving. At the same time, it also obtains location information and timestamps from the GPS module and adds them to the video data. This creates spatiotemporal information for each video frame.

[1523] Input: Video data during driving, GPS location information, timestamp

[1524] Output: Video data with location information and timestamp added

[1525] Step 2: Sending video data The device sends the collected video data to the server in real time via the communication module (5G). Thanks to high-speed communication, the data is immediately transferred to the server.

[1526] Input: Video data with location information and timestamps

[1527] Output: Video data sent to the server

[1528] Step 3: Receiving and storing data The server receives the video data sent from the device, verifies the data integrity and format, and then stores it in the database.

[1529] Input: Video data sent from the device

[1530] Output: Video data stored in a database

[1531] Step 4: Data Preprocessing The server applies a noise reduction filter to the stored video data, then resizes it to a format suitable for analysis, extracts frames containing important information, and removes unnecessary frames.

[1532] Input: Video data stored in the database

[1533] Output: Denoised, resized and frame extracted data

[1534] Step 5: Image data analysis. The server inputs the preprocessed data into the generative AI model. The model performs image analysis and automatically detects road damage. Attribute information such as the type, extent, and depth of the detected damage is acquired.

[1535] Input: Preprocessed video data

[1536] Output: Attribute information of the damaged area (type, range, depth)

[1537] Step 6: Classification of Damage Level Based on the acquired attribute information of the damaged area, the server evaluates the damage level using a classification algorithm. Damage areas are classified into levels such as minor, caution, and serious.

[1538] Input: Attribute information of the damaged area

[1539] Output: Classified damage levels

[1540] Step 7: Generation of repair policy The server automatically generates a repair policy based on the classification results. The generated repair policy includes the urgency of repair, recommended repair methods, and estimated repair costs.

[1541] Input: Classified damage level

[1542] Output: Repair policy (urgency, repair method, estimated cost)

[1543] Step 8: Presenting the results The server presents the generated repair policy and information on the damaged areas to the user through a user interface. This information is displayed on a map so that the user can visually check it. Repair notifications are also sent in real time.

[1544] Input: Repair policy, information on damaged areas

[1545] Output: Information presentation and repair notification through user interface

[1546] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1547] This invention combines an emotion engine with a system that analyzes video data collected from a drive recorder and automatically generates road damage locations and repair plans, making it possible to present repair plans that are more suited to the user.The main components of this system are a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[1548] Program processing

[1549] 1. Image data collection

[1550] Device:

[1551] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[1552] 2. Receipt and storage of data

[1553] server:

[1554] The system receives video data sent from the device and verifies its format and integrity. If there are no problems, the data is stored in a database. The stored data includes the video data, location information, and timestamp.

[1555] 3. Data Preprocessing

[1556] server:

[1557] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[1558] 4. Analysis of image data

[1559] server:

[1560] The preprocessed video data is input into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage.

[1561] 5. Damage Level Classification

[1562] server:

[1563] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or severe. This classification result and location information are stored in a database.

[1564] 6. Repair Policy Generation

[1565] server:

[1566] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[1567] 7. Leveraging Emotional Engines

[1568] server:

[1569] When an interaction with a user (road administrator) occurs through the user interface, the emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.). Based on the recognized emotions, the content of the repair policy to be presented and the way it is displayed are adjusted.

[1570] 8. Presentation of results

[1571] server:

[1572] The emotional data obtained from the emotion engine is analyzed to optimize the way repair plans are presented. For example, if the user is feeling anxious, detailed explanations and additional data can be provided to reassure them. The generated repair plans are then displayed in the user interface in an easy-to-understand format, such as a map or list view.

[1573] User:

[1574] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[1575] Specific examples

[1576] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10 cm deep and 50 cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[1577] In this way, the present invention realizes a system that supports road administrators in formulating road repair plans efficiently and accurately, and provides appropriate information while taking into consideration the feelings of users, thereby shortening the lead time until repairs and optimizing repair costs.

[1578] The processing flow will be explained below.

[1579] Step 1:

[1580] Device:

[1581] The drive recorder is installed in the vehicle and records video in real time while the vehicle is driving on the road. The built-in GPS sensor adds location information and a timestamp to the video data. The video data is then sent to a server in real time via a 5G communication module.

[1582] Step 2:

[1583] server:

[1584] Receives video data sent from the device. Automatically verifies the format and integrity of the received data. If there are no problems, stores the received data in storage. This includes the video data, location information, and timestamp.

[1585] Step 3:

[1586] server:

[1587] The stored video data is subjected to a noise reduction filter, the video resolution is resized to a format suitable for analysis, and pre-processing is performed to extract important frames and remove unnecessary frames.

[1588] Step 4:

[1589] server:

[1590] The preprocessed data is input into a generative AI model, which analyzes the video data to detect road damage (potholes, cracks, etc.) and extracts information on the type, extent, depth, and location of the damage.

[1591] Step 5:

[1592] server:

[1593] Based on the information on the detected damaged areas, a unique classification algorithm is applied to evaluate the level of damage, classifying the damaged areas into levels such as minor, caution, or serious, and saving the results in a database.

[1594] Step 6:

[1595] server:

[1596] Based on the classification results, an optimal repair plan is automatically generated. This repair plan includes the urgency of repair, recommended repair methods, estimated repair costs, etc. The repair plan is then created in report format.

[1597] Step 7:

[1598] server:

[1599] The system prepares to present a repair policy to the user through the user interface. When presenting the policy, the emotion engine analyzes the user's facial expressions and voice to recognize emotions in real time. Based on the recognized emotions (e.g., relief, anxiety, surprise, etc.), the way the repair policy is presented is adjusted.

[1600] Step 8:

[1601] server:

[1602] Based on feedback from the emotion engine, the repair plan presented to the user is optimized. For example, if the user is feeling anxious, the plan will be adjusted to provide detailed explanations and additional materials to reassure the user. The repair plan is displayed in the user interface in an easy-to-use format such as a map or list.

[1603] Step 9:

[1604] User:

[1605] The user (road administrator) can view detailed information about the damaged areas and repair plans through a web interface or a dedicated application. The information presented is adjusted by the emotion engine, allowing the user to quickly and appropriately obtain the information they need. Based on this, a specific repair plan can be quickly formulated.

[1606] Through the above processing flow, the system of the present invention helps road managers to create repair plans efficiently and accurately. In addition, the emotion engine allows the system to present information that takes into account the user's emotions, thereby improving user satisfaction.

[1607] Example 2

[1608] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1609] Although technology already exists to analyze road footage acquired from dashcams and detect damaged areas, there has not previously been any technology that provides a method for presenting repair plans that take the user's emotions into consideration. As a result, repair plans are not presented at the optimal time and in the optimal way based on the user's emotions, which leads to user anxiety and a lack of understanding.

[1610] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1611] In this invention, the server includes means for transmitting video data acquired from the drive recorder in real time, means for saving the received video data, means for preprocessing the saved video data, means for analyzing the preprocessed data and detecting road damage, means for classifying the detected damage, means for generating a repair plan based on the classification results, means for presenting the generated repair plan to the user, and means for recognizing the user's emotions and adjusting the content and method of presenting the repair plan based on the emotions. This makes it possible to present a repair plan that takes the user's emotions into consideration, and to provide information with the most appropriate timing and content for the user.

[1612] A "dashcam" is a device installed in a vehicle that records video in real time while driving.

[1613] "Video data" refers to data that includes video frames recorded by a drive recorder and additional information such as associated location information and timestamps.

[1614] "Means for transmitting in real time" refers to the communication module and protocol for transferring video data obtained from the drive recorder to the server without delay.

[1615] "Means for storage" refers to a database system or storage device for stably and persistently storing received video data.

[1616] "Preprocessing means" refers to tools and algorithms that process stored video data, such as removing noise, adjusting resolution, and extracting important frames.

[1617] "Means of analysis" refers to models or software that identify road damage areas and extract attribute information based on preprocessed data.

[1618] "Means for classifying damage points" refers to an algorithm that analyzes the severity and extent of the detected damage points and groups them according to different classification criteria.

[1619] "Means for generating repair policies" refers to a system that automatically calculates the repair method, urgency, estimated cost, etc. for damaged areas based on the classification results, and creates an appropriate repair plan.

[1620] "Presentation means" refers to a system or software for visually displaying the generated repair policy through a user interface in an easy-to-understand manner.

[1621] "Means for recognizing emotions" refers to engines or algorithms that analyze a user's facial expressions and voice on the user interface and identify emotions in real time.

[1622] "Adjustment means" refers to a mechanism that dynamically changes the content and method of presenting repair policies based on the recognized user emotions.

[1623] This invention is a system that analyzes video data collected from a drive recorder, detects damaged areas on roads, and automatically generates repair plans. By combining this system with an emotion engine, it is possible to present appropriate repair plans according to the user's emotions.

[1624] The system mainly consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[1625] Drive recorder and communication module

[1626] Device:

[1627] The dashcam is installed in the vehicle and records video in real time as the vehicle travels along the road. The built-in GPS sensor acquires location information and adds a timestamp and location information to the video data. This video data is then sent to a server in real time using a 5G communication module.

[1628] Servers and Databases

[1629] server:

[1630] The server stores the received video data, which undergoes pre-processing to apply noise reduction filters and resize the resolution to a format optimized for analysis, improving the quality of the video data and making it easier to analyze.

[1631] Analysis using generative AI models

[1632] server:

[1633] The preprocessed data is input into a generative AI model. This model detects road damage from the video data and extracts attribute information such as the type, extent, depth, and location of the damage. It then applies a classification algorithm based on the damage information to classify the damage level as minor, caution, or severe. The classification results and location information are stored in a database.

[1634] Generation and presentation of repair policies

[1635] server:

[1636] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[1637] Use of emotion engine

[1638] server:

[1639] The emotion engine recognizes the user's emotions in real time. It analyzes the user's facial expressions and voice to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and presentation of the repair plan presented.

[1640] User Interface

[1641] User:

[1642] Users (road administrators) can view detailed information about damaged areas and repair plans through a web interface or dedicated application. Information presentation tailored by the emotion engine allows users to obtain the necessary information quickly and with a clear understanding. Based on this information, specific repair plans can be quickly formulated.

[1643] Specific examples

[1644] For example, a dashcam captures a scene where a large new pothole has appeared. This video data is sent to a server in real time and stored. The server preprocesses the data and uses a generative AI model to detect the pothole. As a result, the pothole is evaluated as being 10cm deep and 50cm wide, and classified as "severely damaged." Repair options of "urgent repair" and "complete repaving" are recommended, and the estimated repair costs are calculated. When this information is presented to the user through the user interface, the emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, additional explanations and data are provided to reassure them.

[1645] Prompt Sentence Examples

[1646] Below are some example prompts to input to a generative AI model:

[1647] "Analyze the following video data to detect road damage. Damage data should include attribute information such as type, extent, depth, and location."

[1648] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1649] Step 1:

[1650] Image data collection

[1651] Device:

[1652] A drive recorder is installed in the vehicle. While the vehicle is driving on the road, the drive recorder records video in real time. The built-in GPS sensor acquires the current location information and adds a timestamp and location information to the video data.

[1653] Specific behavior:

[1654] The dashcam starts up and the camera starts recording video, and this video data is the input.

[1655] The GPS sensor obtains the current latitude and longitude information.

[1656] The acquired location information and timestamp are added to the video frame.

[1657] This becomes the video data as output.

[1658] Step 2:

[1659] Sending and Receiving Data

[1660] Devices and Servers:

[1661] Using a 5G communication module, the video data collected in real time is transmitted to a server, which receives and stores the data.

[1662] Specific behavior:

[1663] The video data is transmitted through the 5G communication module, and this transmitted data is the input.

[1664] The server listens to receive data.

[1665] Check the format and integrity of the data received.

[1666] The validated data is saved in a database. The saved data is the output.

[1667] Step 3:

[1668] Data Preprocessing

[1669] server:

[1670] The received data is filtered to remove noise, resized to a resolution suitable for analysis, and the important frames are extracted and unnecessary frames are removed.

[1671] Specific behavior:

[1672] The saved video data is read out. This data is the input.

[1673] Improve the quality of your footage with noise reduction filters.

[1674] Resize the video resolution to the optimal format for AI analysis.

[1675] Frame analysis is performed to select and extract important frames.

[1676] Optimize data volume by removing unnecessary frames.

[1677] After these operations, the preprocessed data is the output.

[1678] Step 4:

[1679] Image data analysis

[1680] server:

[1681] The preprocessed data is input into a generative AI model, which detects road damage from the video data and extracts attribute information such as its type, extent, depth, and location.

[1682] Specific behavior:

[1683] The preprocessed data is fed into the generative AI model. This data is the input.

[1684] The AI ​​model detects the damaged areas.

[1685] Attribute information such as the type, extent, and depth of the damage is extracted.

[1686] The extracted information is the output.

[1687] Step 5:

[1688] Damage Level Classification

[1689] server:

[1690] Based on the information on the damaged area, a classification algorithm is applied to classify the damage level as minor, caution, or serious. This classification result and location information are stored in a database.

[1691] Specific behavior:

[1692] The extracted damage information is fed into a classification algorithm, and this data is the input.

[1693] An algorithm classifies the level of damage as either "minor," "cautionary," or "severe."

[1694] The classification results and location information are saved in a database. The saved data is the output.

[1695] Step 6:

[1696] Generating repair policies

[1697] server:

[1698] Based on the classification results, an optimal repair policy is automatically generated. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is then presented in report format.

[1699] Specific behavior:

[1700] Based on the classification results, a repair policy generation algorithm is applied. This data is the input.

[1701] The algorithm calculates the urgency of repairs, recommended repair methods, and estimated repair costs.

[1702] The repair policy is compiled in a report format. This is the output report.

[1703] Step 7:

[1704] Use of emotion engine

[1705] server:

[1706] The system uses an emotion engine to recognize user emotions in real time. It analyzes the user's facial expressions and voice on the user interface to identify emotions such as relief, anxiety, and surprise. Based on the recognized emotion, it adjusts the content and display method of the repair policy presented.

[1707] Specific behavior:

[1708] The user's facial expressions and voice data are input to the emotion engine. This data is the input.

[1709] The emotion engine analyzes and identifies the user's emotions.

[1710] Based on the recognized emotion data, the presentation method and content of the repair policy are adjusted, and the adjusted presentation method is the output.

[1711] Step 8:

[1712] Presentation of results

[1713] Servers and Users:

[1714] The generated repair policy is visually presented to the user through a user interface, and the user can then create a specific repair plan based on this information.

[1715] Specific behavior:

[1716] The repair policy adjusted by the emotion engine is displayed on the user interface, and the repair policy report is the input.

[1717] The user views the repair policy through a web interface or a dedicated application.

[1718] The user quickly creates a repair plan based on the information, and the repair plan with the corrective measures is the output.

[1719] (Application example 2)

[1720] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1721] Conventional systems for detecting road damage and generating repair plans have the problem of being unable to respond quickly because it takes time to analyze video data and generate repair plans.In addition, they do not provide information that takes into account the situation and emotions of each user, so they are unable to provide users with a sufficient sense of security.

[1722] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1723] In this invention, the server includes a means for transmitting video data acquired from the drive recorder in real time, a means for saving the received video data, and a means for preprocessing the saved video data. This enables real-time collection and saving of video data. The server also includes a means for analyzing the preprocessed data to detect road damage, a means for classifying the detected damage, a means for generating a repair plan based on the classification results, and a means for optimizing and presenting the generated repair plan in response to the user's emotions. This enables information to be provided quickly and in consideration of the user's emotions.

[1724] A "drive recorder" is a device that is installed in a vehicle and records video and audio data in real time while the vehicle is driving.

[1725] "Means for receiving" refers to a device or method for receiving signals or data transmitted from an external source and storing or processing them.

[1726] The term "storing means" refers to a device or method that stores received or generated data in a storage device and keeps it in a reusable state as needed.

[1727] "Preprocessing means" refers to a device or method that performs a series of initial processes to improve the efficiency of data analysis, including noise removal and resolution adjustment.

[1728] The "means for analyzing and detecting road damage" refers to a device or method that uses a specific algorithm to find damaged portions of the road from the pre-processed data.

[1729] The "classifying means" refers to a device or method for grouping detected damaged areas based on specific criteria, including classification by the degree or type of damage.

[1730] The "means for generating a repair policy" is a device or method that automatically determines the optimal repair method and priorities based on the classification results.

[1731] "Means for optimizing and presenting according to emotions" refers to a device or method that analyzes the user's emotional state in real time, adjusts the display content and display method based on that information, and reduces stress and anxiety.

[1732] An "emotion engine" is software or hardware that analyzes emotions from a user's facial expressions, voice, etc., and detects specific emotions.

[1733] This invention is a system for autonomous vehicles that detects road damage in real time, generates appropriate repair policies, and presents optimized information according to the user's emotions. The system consists of a drive recorder, a communication module, a server, a database, a generative AI model, an emotion engine, and a user interface.

[1734] 1. System configuration and functions

[1735] 1.1 Drive recorder

[1736] A dashcam is a device that records video data in real time while a vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. Furthermore, using a 5G communication module, this video data is sent to a server without delay.

[1737] 1.2 Server

[1738] The server stores the received video data and verifies its format and completeness. The stored data includes video data, location information, and timestamps. During the data preprocessing stage, the server applies a noise reduction filter and adjusts the resolution to create an optimal format for analysis. Next, a generative AI model is used to detect road damage from the preprocessed video data. The detection results are extracted as attribute information, such as the type and extent of the damage, and location information, and the damage level is classified based on this information.

[1739] 1.3 Generating Repair Policies

[1740] The server automatically generates an optimal repair policy based on the classification results. This repair policy includes the urgency of repairs, recommended repair methods, and estimated repair costs, and is generated in the form of a report.

[1741] 1.4 Emotion Engine

[1742] When an interaction with the user (road administrator) occurs through the user interface, an emotion engine is used to recognize the user's emotions in real time. The emotion engine analyzes the user's facial expressions and voice to identify emotions such as "relief," "anxiety," and "surprise." Based on this emotion, the content of the repair policy presented and the way it is displayed can be adjusted. For example, if the user feels anxious, detailed explanations and additional data can be provided to reassure them.

[1743] 1.5 Presentation of results

[1744] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair plans are presented. For example, if the user is feeling anxious, it adds more detailed explanations and provides information that reassures the user. The generated repair plans are displayed in an easy-to-understand format on the user interface, such as a map or list view. Through this interface, the user can quickly check the repair plans and detailed information about the damaged areas, and then quickly create a specific repair plan.

[1745] 2. Specific examples

[1746] For example, consider the case where a large pothole has newly appeared on a road. The dashcam installed in an autonomous vehicle captures the incident, and the video data is sent to a server in real time. The server preprocesses the data and detects the pothole using a generative AI model. The resulting pothole is evaluated as 10 cm deep and 50 cm wide and classified as severely damaged. Repair options of "urgent repair" and "complete repaving" are recommended, and an estimated repair cost is calculated. When this information is presented to the user through the user interface, an emotion engine recognizes the user's emotions. For example, if the user is feeling anxious, the system adjusts its response by providing additional explanations and data, presenting more detailed information to reassure the user. In this way, the system helps road managers plan repairs efficiently and accurately.

[1747] 3. Examples of prompts

[1748] Hardware used: Dashcam, GPS module, 5G communication module, server, user interface device

[1749] Software used: OpenCV, TensorFlow, EmotionRecognizer, and GPS module libraries

[1750] Examples of specific prompts include the following:

[1751] Prompt: "Real-time transmission of video data from dashcams, detection of road damage, and information display using emotion recognition."

[1752] As a result, the present invention has a function in which the entire system operates in cooperation with one another and can provide information in a form that is easy for the user to understand and that takes into consideration the user's emotions.

[1753] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1754] Step 1: Collecting video data from the dashcam

[1755] The device's drive recorder records video data in real time while the vehicle is driving. The built-in GPS sensor acquires location information and adds it to the video data along with a timestamp. The collected video data is then sent to a server via a 5G communication module.

[1756] Input: Video data captured while the vehicle is moving, GPS location information, and timestamp

[1757] Output: Real-time video data, location information, and time-stamped data sent to the server

[1758] Step 2: Receiving and storing data

[1759] The server receives the video data sent from the device in real time and verifies its format and integrity. If there are no problems, the data is stored in a database. The data includes the video data, location information, and timestamp.

[1760] Input: Real-time video data, location information, timestamp

[1761] Output: Validated data stored in a database

[1762] Step 3: Preprocessing the data

[1763] The server applies a noise reduction filter to the video data stored in the database, then adjusts the resolution to make it optimal for analysis, and preprocesses the data by extracting important frames and removing unnecessary ones.

[1764] Input: Video data stored in the database, location information, timestamp

[1765] Output: Preprocessed data with denoising and resolution adjustment

[1766] Step 4: Analyze the image data and detect any damage

[1767] The server inputs the preprocessed video data into a generative AI model, which then detects road damage from the video data and extracts attribute information such as the type and extent of the damage, as well as location information.

[1768] Input: Preprocessed video data

[1769] Output: Damage detection results (damage type, range, location information)

[1770] Step 5: Classify the level of damage

[1771] The server applies a classification algorithm based on the location of the damage and classifies the damage as minor, minor, or severe. The classification results and location information are stored in a database.

[1772] Input: Damage detection results (damage type, extent, location information)

[1773] Output: Classification result (damage level), location information

[1774] Step 6: Generate repair policies

[1775] The server automatically generates an optimal repair policy based on the classification results. This policy includes the urgency of repair, recommended repair methods, and estimated repair costs. The generated repair policy is saved in a database in report format.

[1776] Input: Classification result (damage level), location information

[1777] Output: Repair policy (urgency of repair, recommended repair method, estimated repair cost)

[1778] Step 7: Leverage your emotional engine

[1779] The server uses an emotion engine to recognize the user's emotions in real time when an interaction occurs with the user through the user interface. The emotion engine analyzes the user's facial expressions and voice to identify emotions (e.g., relief, anxiety, surprise, etc.).

[1780] Input: User facial expressions and voice data obtained via the user interface

[1781] Output: User's emotional data (relief, anxiety, surprise, etc.)

[1782] Step 8: Presenting the results

[1783] The server analyzes the emotion data obtained from the emotion engine and optimizes the way repair policies are presented. The generated repair policies are displayed on the user interface in an easy-to-understand format, such as a map or list. Based on this information, users can quickly create specific repair plans.

[1784] Input: Repair policy, user emotion data

[1785] Output: Optimized information presentation and support for creating specific repair plans

[1786] This allows the entire system to work in harmony, providing users with fast, accurate information and support that takes their emotions into consideration.

[1787] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1788] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1789] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1790] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1791] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1792] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1793] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1794] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1795] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1796] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1797] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1798] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1799] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1800] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1801] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1802] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1803] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1804] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1805] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1806] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1807] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1808] The following is further disclosed regarding the above embodiment.

[1809] (Claim 1)

[1810] A means for transmitting video data acquired from the drive recorder in real time;

[1811] means for storing the received video data;

[1812] means for preprocessing the stored video data;

[1813] means for analyzing the preprocessed data to detect road damage;

[1814] a means for classifying the detected damage;

[1815] means for generating a repair policy based on the classification results;

[1816] A means for presenting the generated repair policy to a user;

[1817] A system including:

[1818] (Claim 2)

[1819] 10. The system of claim 1, further comprising means for denoising the preprocessed data.

[1820] (Claim 3)

[1821] 10. The system of claim 1, further comprising means for adjusting the resolution of the preprocessed data.

[1822] (Claim 4)

[1823] 2. The system according to claim 1, further comprising means for storing location information and damage level of the analyzed damaged portion in a database.

[1824] (Claim 5)

[1825] 10. The system of claim 1, further comprising: means for providing an interface for a user to view the repair policy.

[1826] "Example 1"

[1827] (Claim 1)

[1828] A device that transmits video data acquired from a drive recorder in real time;

[1829] a device for storing the received video data;

[1830] a device for preprocessing the stored video data;

[1831] A generative model that analyzes preprocessed data and detects road damage;

[1832] An algorithm to classify the detected damages;

[1833] A method for generating a repair policy based on the classification results;

[1834] The system includes an interface that presents the generated repair policy to a user.

[1835] (Claim 2)

[1836] 10. The system of claim 1, further comprising a pre-processed data denoiser.

[1837] (Claim 3)

[1838] 10. The system of claim 1, further comprising a device for adjusting the resolution of the preprocessed data.

[1839] "Application Example 1"

[1840] (Claim 1)

[1841] A means for transmitting video data acquired from the drive recorder in real time;

[1842] means for storing the received video data;

[1843] means for preprocessing the stored video data;

[1844] means for analyzing the preprocessed data to detect road damage;

[1845] a means for classifying the detected damage;

[1846] means for generating a repair policy based on the classification results;

[1847] A means for presenting the generated repair policy to a user;

[1848] A means for displaying information about the damaged area on a map;

[1849] A means of notifying repairs of damaged areas in real time,

[1850] A system including:

[1851] (Claim 2)

[1852] 10. The system of claim 1, further comprising means for denoising the preprocessed data.

[1853] (Claim 3)

[1854] 10. The system of claim 1, further comprising means for adjusting the resolution of the preprocessed data.

[1855] "Example 2: Combining Emotion Engines"

[1856] (Claim 1)

[1857] A means for transmitting video data acquired from the drive recorder in real time;

[1858] means for storing the received video data;

[1859] means for preprocessing the stored video data;

[1860] means for analyzing the preprocessed data to detect road damage;

[1861] a means for classifying the detected damage;

[1862] means for generating a repair policy based on the classification results;

[1863] A means for presenting the generated repair policy to a user;

[1864] A means for recognizing a user's emotion and adjusting the content and method of presenting a repair policy based on the emotion;

[1865] A system including:

[1866] (Claim 2)

[1867] 10. The system of claim 1, further comprising means for denoising the preprocessed data.

[1868] (Claim 3)

[1869] 10. The system of claim 1, further comprising means for adjusting the resolution of the preprocessed data.

[1870] "Application example 2 when combining emotion engines"

[1871] (Claim 1)

[1872] A means for transmitting video data acquired from the drive recorder in real time;

[1873] means for storing the received video data;

[1874] means for preprocessing the stored video data;

[1875] means for analyzing the preprocessed data to detect road damage;

[1876] a means for classifying the detected damage;

[1877] means for generating a repair policy based on the classification results;

[1878] A means for optimizing the generated repair policy according to the user's emotions and presenting the same;

[1879] A system including:

[1880] (Claim 2)

[1881] 10. The system of claim 1, further comprising means for denoising the preprocessed data.

[1882] (Claim 3)

[1883] 10. The system of claim 1, further comprising means for adjusting the resolution of the preprocessed data. [Explanation of symbols]

[1884] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for transmitting video data acquired from the drive recorder in real time; means for storing the received video data; means for preprocessing the stored video data; means for analyzing the preprocessed data to detect road damage; a means for classifying the detected damage; means for generating a repair policy based on the classification results; A means for presenting the generated repair policy to a user; A system including:

2. 10. The system of claim 1, further comprising means for denoising the preprocessed data.

3. The system of claim 1 further comprising means for adjusting the resolution of the preprocessed data.

4. 2. The system according to claim 1, further comprising means for storing location information and damage level of the analyzed damaged portion in a database.

5. 10. The system of claim 1, further comprising means for providing an interface for a user to view the repair policy.

Images