system

Abstract

We provide the system. [Solution] A means for receiving video data acquired from a monitoring device, Means for processing the aforementioned video data into an analyzable format, A means of identifying the location and hazardous state of an object using artificial intelligence image analysis, A means of evaluating areas for improvement based on identified hazardous conditions, A means for generating and sending a notification to the user based on the aforementioned evaluation, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of the chatbot's character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Recently, the risks brought about by the furniture arrangement in houses and offices, especially the problem of insufficient evacuation routes due to the falling of items during disasters such as earthquakes and the obstacles in the passageways, have become apparent. In addition, the furniture arrangement may also pose unexpected risks to infants. An efficient method for enhancing safety by early recognition and accurate notification of such potential risks is demanded.

Means for Solving the Problems

[0005] This invention provides a system that acquires video data from a monitoring device and processes it using AI-based image analysis. This system identifies the arrangement of objects and potential hazardous conditions, and based on the results, provides the user with notifications including specific improvement suggestions. In this way, it provides the user with a means to intelligently and automatically contribute to improving safety.

[0006] A "surveillance device" is an electronic device that has the function of capturing images of the surroundings and transmitting that data in real time.

[0007] "Video data" refers to visual information recorded as still images or videos, and is digital data acquired by a specific device.

[0008] "Artificial intelligence image analysis" is a technology that uses machine learning algorithms to automatically recognize and analyze specific patterns, objects, and environmental information from digital image data.

[0009] "Object position" refers to information indicating the arrangement of an object in three-dimensional space within a specific environment.

[0010] A "hazardous condition" is a state in which the arrangement or arrangement of objects within an environment is judged to potentially threaten safety in the event of a disaster or accident.

[0011] "Means of evaluation" refers to the process of determining whether a detected condition is safe or dangerous based on specific criteria, and deciding whether improvement is necessary.

[0012] A "notification" is the act of conveying specific information or a message to a user, or the message itself used for that purpose.

[0013] A "user" is an individual or group that utilizes this system, receives notifications, and takes action as needed. [Brief explanation of the drawing]

[0014] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the 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.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0028] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention relates to a system that uses AI technology to analyze video data acquired using a monitoring device, detects specific dangerous conditions, and notifies the user. The program processing of this system will be explained below using natural language.

[0036] First, the server continuously receives video data transmitted from the monitoring device. This allows it to always have access to the latest information about the monitored area. Next, the received video data is converted to an appropriate resolution and format and pre-processed to a format suitable for AI image analysis.

[0037] Next, the server uses an AI image analysis program to analyze the video data. This analysis identifies the location and state of specific objects and detects potential hazards such as heavy objects in high places or obstacles blocking pathways. Based on this information, the system assesses the risks in emergencies such as earthquakes.

[0038] The server then generates a notification message containing specific improvement suggestions based on the analysis results. For example, if objects are stacked in a particular location, it might create a notification such as, "We recommend moving the heavy object in this location." These notifications are sent to the user's electronic device via the terminal.

[0039] Users can check the situation based on the received notifications and adjust the placement of furniture and objects as needed. This action can mitigate potential hazards and improve the safety of living spaces and offices.

[0040] As a concrete example, when a monitoring system was used in an office, the server detected a large filing box stacked on a metal shelf. Based on this information, AI analysis determined that it posed a risk of falling during an earthquake and generated a notification saying, "Please move the filing box to a location below the shelf." Upon receiving this notification, the user can immediately move the filing box to a safe location, thereby mitigating the risk. In this way, the present invention aims to raise users' safety awareness and prevent accidents.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The server receives video data from the monitoring device in real time. The received data is recorded as a secure stream and used for subsequent processing.

[0044] Step 2:

[0045] The server preprocesses the received video data into a format that can be analyzed. This process includes adjusting the resolution and performing noise filtering.

[0046] Step 3:

[0047] The server inputs the pre-processed video data into an AI image analysis module. The artificial intelligence model used here performs object recognition and identifies the placement of furniture and other objects.

[0048] Step 4:

[0049] Based on the results of the AI ​​analysis, the server evaluates the location and condition of identified objects and determines potential hazards. For example, it assesses the presence of heavy objects at high altitudes and the degree of obstruction in passageways.

[0050] Step 5:

[0051] Based on the analysis results and evaluation, the server generates notifications to send to the user. These notifications are prioritized and presented in a message format that specifically outlines improvement suggestions.

[0052] Step 6:

[0053] The device receives notifications from the server and displays them to the user. Notifications are delivered to the user's device as push notifications or in-app messages.

[0054] Step 7:

[0055] The user reviews the notification and makes improvements to the furniture and object placement as needed. After improvements are made, the system is configured to rerun the process for re-evaluation.

[0056] (Example 1)

[0057] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0058] Conventional video surveillance systems lack the ability to automatically identify and notify of dangerous situations, posing a challenge in preventing risks and ensuring safety in emergency situations where rapid response is required. This results in users being unable to identify potential dangers in a timely manner and take necessary countermeasures.

[0059] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0060] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for converting the video information into an analyzable format, and means for identifying the location and hazardous state of an object using machine intelligence image analysis. This makes it possible to identify hazardous states in real time, immediately notify the user, quickly assess the risk, and take countermeasures.

[0061] A "surveillance device" is a device that visually observes a specific area and continuously acquires video information.

[0062] "Visual information" refers to visual data acquired by surveillance equipment, which is image or video data used for analysis.

[0063] "Machine intelligence image analysis" is a technology that uses deep learning and other advanced algorithms to identify objects in video information and determine their location and state.

[0064] A "hazardous situation" is a situation in which the location or state of an object identified in the video information could potentially pose a risk.

[0065] "Means of risk assessment" refers to the process of determining the degree of risk that an identified hazardous situation may pose and evaluating whether appropriate countermeasures are necessary.

[0066] A "means for generating and communicating attention" is a system that creates messages indicating warnings and countermeasures for users based on the results of a risk assessment, and then electronically sends these messages to the user.

[0067] This invention relates to a system that analyzes video information acquired from a monitoring device and notifies the user of identified dangerous conditions. This system consists of a server, a terminal, and a user, each playing a specific role.

[0068] The server continuously receives video information from monitoring devices using network communication software. The received video information is converted into an AI-analyzable format using an image processing library. This conversion includes resizing and changing the format of the images.

[0069] The converted video information is analyzed on the server using machine intelligence image analysis tools. Using deep learning frameworks such as TENSORFLOW® and PyTorch, the location of objects and their hazardous conditions are identified. For example, it can identify cases where an object is blocking a passageway or where a heavy object is placed at a high position.

[0070] Based on the analysis results, a risk assessment is performed, and specific notifications are generated to improve the hazardous situation. These notifications are sent to the user's electronic device via a terminal. The notification content includes specific instructions such as, "The passage is blocked by an object. Please move the object for safety."

[0071] Users receive notifications from their devices and check the situation on-site based on their content. By taking the instructed actions as needed, they can mitigate potential risks.

[0072] As a concrete example, consider its use in an office. If the server analyzes video information from a monitoring device and identifies a stack of document boxes on a metal shelf, it determines that there is a risk of them falling during an earthquake. A notification is generated and sent to the user saying, "Please move the document boxes under the shelf." This allows the user to move the document boxes to a safe location and mitigate the risk.

[0073] An example of a prompt message would be, "Please tell me how to build a notification system that detects hazardous materials from office security camera footage and assesses the risk."

[0074] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0075] Step 1:

[0076] The server receives video information from the monitoring device. The received video information is provided raw and in real time as a stream. This video information serves as input. Next, an image processing library is used to convert the video information into a format that can be analyzed. Specifically, the image resolution is resized to a size that the AI ​​model can process, and the format is converted to a standard format. As output of this process, image data in a format suitable for AI analysis is generated.

[0077] Step 2:

[0078] The server inputs pre-processed image data into an AI image analysis tool. It performs object recognition using deep learning models such as YOLO and ResNet. Calculations are performed on the input image data to detect the object's location and type. The object detection results are output, including information such as the object's type, location coordinates, and confidence score.

[0079] Step 3:

[0080] The server performs a risk assessment based on the detected object information. The specific input is information about the object's location and type obtained in the previous step. The risk assessment is performed using an expert system, taking into account the object's placement, human movement patterns, and potential hazards in emergencies. The assessment output includes potential hazard areas and priority for countermeasures.

[0081] Step 4:

[0082] The server generates notification messages for the user based on the risk assessment results. The assessment results and recommended actions for the user are used as input. Prompts from the generating AI model are utilized to create notifications that include specific improvement suggestions. Output notifications might include statements such as, "We recommend moving the heavy object located here."

[0083] Step 5:

[0084] The terminal receives notifications sent from the server and displays them to the user. The input is the notification message sent from the server, which is output to the user's electronic device as a push notification or email. Based on the notification content, the user checks the situation on site and takes the instructed actions as needed.

[0085] Step 6:

[0086] Based on the notification, users will identify the areas where risks have been identified. Specific actions might include moving heavy objects from shelves to a safer location. This can reduce potential safety risks and improve the safety of living and working environments.

[0087] (Application Example 1)

[0088] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0089] Industrial facilities present potential hazards arising from complex machine layouts and work environments, necessitating accident prevention measures. However, currently, there is no established method for detecting these hazards in real time and responding immediately, making efficient safety measures difficult. Furthermore, the lack of not only detection of hazards but also instructions for specific corrective measures hinders the rapid implementation of safety measures.

[0090] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0091] In this invention, the server includes means for receiving image information acquired from a monitoring device, means for converting the image information into an analyzable format, and means for identifying the location of an object and its hazardous status using machine learning image analysis. This makes it possible to detect potential hazardous conditions within industrial facilities in real time and to quickly notify operators of specific corrective measures.

[0092] A "monitoring device" is a device that captures the surrounding environment and continuously acquires image information.

[0093] "Image information" refers to data that represents the visual information of the surroundings, acquired by a monitoring device.

[0094] A "server" is a computer device that receives and processes information from multiple devices.

[0095] A "parsable format" refers to the data structure and format necessary for processing using machine learning techniques.

[0096] "Machine learning image analysis" is an analytical method that uses advanced technologies such as neural network algorithms to recognize objects in an image and evaluate the situation.

[0097] "Object position" refers to information that indicates the spatial arrangement of a specific object within the observed area.

[0098] A "hazardous situation" refers to any condition or arrangement in an industrial facility that could potentially threaten safety.

[0099] An "operator" is a person responsible for managing and controlling machinery and equipment and maintaining the safety of the facility.

[0100] A "notification" is an informational message generated based on the detected situation and sent to the operator.

[0101] An "information terminal" is a device used by operators to view received notifications.

[0102] The system program for implementing this invention consists of a monitoring device, a server, machine learning software, and a user information terminal.

[0103] The server continuously receives image information transmitted from monitoring devices. This image information shows the current arrangement and status of machinery and objects within the facility. The server first converts this information into a format that can be used for machine learning image analysis. This involves image format conversion and resolution adjustment.

[0104] Next, the server performs machine learning image analysis. Specifically, a neural network algorithm is used to recognize objects in the image and determine their location. After that, it identifies dangerous situations and identifies areas for improvement as needed. Frameworks such as TensorFlow are used for this analysis.

[0105] Next, the server generates a notification containing specific instructions to encourage corrective action based on the identified risk situation. This notification is sent as a push message to the operator's information terminal. The notification is communicated using services such as Firebase Cloud Messaging.

[0106] As a concrete example, consider a situation in a factory where a robotic arm is blocking a passageway and obstructing an emergency evacuation route. A monitoring device detects this situation, and a server performs real-time analysis. The analysis results are notified in the form of "We recommend adjusting the position of the robotic arm" and sent to the operator's information terminal. This enables a rapid response.

[0107] An example of a prompt to input into the generating AI model would be: "Explain the notification generation process for a system that uses AI to analyze video data inside a factory, detect dangerous situations, and generate notifications."

[0108] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0109] Step 1:

[0110] The server receives image information transmitted from the monitoring device. The input is real-time image data of the factory captured by the monitoring device. At this stage, the image data is retained in its original format. The output is the unprocessed image data stored in temporary storage for data retention.

[0111] Step 2:

[0112] The server converts received image information into a format suitable for analysis. The input is raw image data. Data processing is performed to change the image format to an appropriate format such as JPEG or PNG, and to reduce the resolution to a size suitable for machine learning models. The output is image data converted into a format suitable for analysis.

[0113] Step 3:

[0114] The server performs machine learning image analysis. The input is the image data transformed in the previous step. Here, the server applies a neural network model using a framework such as TensorFlow to recognize objects and identify hazardous situations within the image. The output is data on the location of objects and hazardous situations.

[0115] Step 4:

[0116] The server evaluates what needs to be improved based on the hazard situation and generates specific improvement instructions. The input is the location information of the identified object and hazard situation data. The data is analyzed to create an assessment of specific hazards, such as "the passage is blocked," and response instructions, such as "adjust the position of the robot arm." The output is notification message data, including the improvement instructions.

[0117] Step 5:

[0118] The server sends the generated notification to the operator's information terminal. The input is notification message data, including improvement instructions. Push notification services such as Firebase Cloud Messaging are used to send notifications to the operator's terminal in real time. The output is the notification message displayed on the operator's information terminal screen.

[0119] Step 6:

[0120] The operator checks the information notified on the terminal and makes adjustments to the machine layout, etc., according to the instructions. The input is the notification message displayed on the terminal. Based on the notification content, the operator checks the site and takes specific actions such as moving objects or adjusting the position of machinery. The output is the new layout within the facility after safety has been improved.

[0121] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0122] This invention provides a system that identifies hazardous elements in the environment and provides appropriate notifications to the user by combining video data acquisition from a monitoring device with AI image analysis. Furthermore, it incorporates an emotion engine that recognizes the user's emotional state and individually optimizes user interaction. The processing of this system's program is described in detail below in natural language.

[0123] First, the server receives video data acquired from the monitoring device and preprocesses it into a format suitable for analysis. The preprocessed data is then input into the AI ​​image analysis module to identify the placement of objects and potential hazards in the environment. In this identification process, deep learning algorithms are used to achieve highly accurate object recognition.

[0124] Next, the server evaluates the identified risk conditions based on the analysis results and creates improvement plans. In this process, it uses an emotion engine to simultaneously recognize the user's emotional state from their facial expressions and voice. If the user is feeling anxious or stressed, the server adjusts the content and tone of notifications accordingly and adds information to reassure the user.

[0125] Next, the server formats the generated notification in the most suitable format for the user and sends it to the user's digital device via the terminal. The notification is delivered using the most effective method, such as push notification or email. The content of this notification also takes into account the results of the emotion engine's analysis, enabling interaction that takes into account the user's emotional state.

[0126] As a concrete example, consider the use of the system in an office. If an employee is sitting at their desk with an anxious expression, the server will recognize this emotion and, in addition to the usual safety notification, send a notification that includes emotionally supportive content, such as, "For your safety, we will provide you with the most relaxed environment possible. Please let us know if you need any support." In this way, the system can provide the user with the most appropriate support.

[0127] The following describes the processing flow.

[0128] Step 1:

[0129] The server receives video data in real time from the monitoring devices. This allows it to collect basic data to understand the latest situation in the monitored area.

[0130] Step 2:

[0131] The server preprocesses the received video data, performing noise reduction and resolution adjustments, and converts it into a format suitable for AI image analysis.

[0132] Step 3:

[0133] The server inputs the pre-processed video data into an AI image analysis module. This AI module utilizes deep learning to perform object recognition and location determination with high accuracy.

[0134] Step 4:

[0135] The server assesses potential hazards based on the placement of objects identified from the analysis results. This assessment includes the risk of objects falling during an earthquake and the possibility of difficulty in evacuation due to obstacles in pathways.

[0136] Step 5:

[0137] The server simultaneously uses an emotion engine to recognize the user's emotional state. It analyzes input data from cameras and audio sensors to identify emotions from the user's facial expressions and voice.

[0138] Step 6:

[0139] The server generates notification content based on the risk assessment results and the user's emotional state. This notification includes not only conventional safety measures but also messages and improvement suggestions that take the user's feelings into consideration.

[0140] Step 7:

[0141] The device receives notifications generated from the server and displays them as push notifications on the user's electronic device. These notifications are delivered at a timing and tone that corresponds to the user's emotional state.

[0142] Step 8:

[0143] Users review the displayed notifications and take appropriate action to ensure the safety of the facility as needed. At this stage, emotions are re-evaluated, allowing for more appropriate support to be provided.

[0144] (Example 2)

[0145] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0146] Modern, high-performance monitoring systems are required not only to identify hazardous elements in the environment, but also to provide individually optimized notifications tailored to the user's emotional state. However, conventional technologies have struggled to achieve such flexible notifications that take user emotions into account. Therefore, a new approach is needed to improve both safety and user experience.

[0147] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0148] In this invention, the server includes a configuration for receiving video information acquired from a monitoring device, a configuration for converting the video information into an analyzable format, a configuration for identifying the arrangement of objects and dangerous conditions using machine learning image analysis, a configuration for recognizing the user's emotional state using an emotion analysis engine, and a configuration for generating and sending notifications to the user based on the evaluation and emotional state. This makes it possible to provide safe and individually optimized notifications that take the user's emotions into consideration.

[0149] A "monitoring device" is a device used to continuously monitor the environment and acquire video information.

[0150] "Visual information" refers to visual data provided by surveillance equipment, including image and video data.

[0151] "Machine learning image analysis" is a method of analyzing video information using machine learning techniques to identify the arrangement of objects and specific patterns.

[0152] An "emotion analysis engine" is a software configuration that determines a user's emotional state based on their facial expressions and voice data.

[0153] A "notification" is an informational message sent to the user based on the analysis results, which may include warnings or suggestions.

[0154] "Push-out notifications" are a notification method that displays information from the system in real time on the user's electronic device.

[0155] "Electronic communication" refers to the method of transmitting digital data between electronic devices, such as using email or messaging apps.

[0156] "Object arrangement" refers to information about the position and state of objects within an environment.

[0157] A "hazardous condition" refers to a situation or condition that could potentially pose a danger to the user or the environment.

[0158] This invention is a monitoring system that identifies hazardous elements in the environment and provides users with individually optimized notifications that respond to their emotions.

[0159] The server first receives video information in real time from the monitoring device. This data is preprocessed into an appropriate format before being analyzed as raw data. Specifically, processes such as data format conversion and noise reduction are performed. This prepares the image analysis module for high-precision analysis.

[0160] Next, using the processed data, the server identifies object placement and hazardous conditions within the environment through machine learning image analysis. Examples of algorithms used include deep learning frameworks such as TensorFlow and PyTorch, with YOLO and Faster R-CNN being used for object identification. During this process, objects identified as hazardous elements are labeled and recorded in a database.

[0161] Furthermore, the server uses an emotion analysis engine to recognize the user's emotional state from their facial expressions and voice data. Speech recognition technology (for example, technology that transcribes speech into text) and facial expression analysis software are used.

[0162] Based on these analysis results, the server generates a notification for the user and sends it to the user's electronic device via the terminal. The notification includes specific actions and support information tailored to the user's situation and emotions. This allows the user to confidently follow the instructions. Notifications are typically delivered to the user as push notifications or email.

[0163] For example, when this system is used in a business environment such as an office, it sends a notification prompting immediate action if a hazardous element is detected. Furthermore, when employees are experiencing stress, it provides emotionally supportive recommendations such as "relax and maintain your health."

[0164] An example of a prompt message might be, "Please tell me how to identify potential hazards in the meeting room using video analysis and how to set up notifications based on employee emotion recognition."

[0165] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0166] Step 1:

[0167] The server receives video information from the monitoring device in real time. This input data is typically in video stream format (e.g., MP4 or AVI). The received data is stored in a temporary cache as a preliminary step before analysis. This ensures that the data is always available for subsequent processing.

[0168] Step 2:

[0169] The server preprocesses the received video information into a format that can be analyzed. The input is video stream data, and the output is image data converted to a still image format (JPEG or PNG). This process involves data format conversion and noise reduction. Specifically, a Gaussian filter is applied to reduce noise and clarify the image.

[0170] Step 3:

[0171] The server performs machine learning image analysis using pre-processed data. The input is still image data, and the output is location information (bounding boxes and labels) of identified objects. Using TensorFlow or PyTorch, models such as YOLO and Faster R-CNN are used to accurately identify the location and type of objects.

[0172] Step 4:

[0173] The server evaluates the hazardous conditions within the environment based on the analysis results. The input is information on the placement of identified objects, and the output is a hazard assessment report. This assessment uses pre-defined safety standards and rules to identify high-risk situations and conditions.

[0174] Step 5:

[0175] The server analyzes the user's emotional state using an emotion analysis engine. Input is the user's facial expressions and voice data, and output is the emotional state (e.g., anxiety, relaxation). Speech recognition and facial expression analysis software are used to recognize the user's emotions according to their situation.

[0176] Step 6:

[0177] The server generates notifications based on the risk level and the user's emotions. Inputs are a risk assessment report and emotional state, while output is a customized notification message. The notification includes specific actions and emotional support to encourage the user to respond confidently.

[0178] Step 7:

[0179] The device sends notifications generated from the server to the user's electronic device. The input is a customized notification message, and the output is a notification received on a smartphone or PC. Notifications are delivered in the form of push notifications or email and are presented to the user in real time.

[0180] (Application Example 2)

[0181] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0182] Conventional monitoring systems have a simplistic and general approach to identifying risk factors and notifying users, making it difficult to respond flexibly to different situations and user emotional states. Therefore, challenges include a lack of real-time response capabilities to risks within the monitoring area and a lack of individualized stress reduction measures for users.

[0183] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0184] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for processing the video information into an analyzable format, means for identifying the location of objects and risk factors using artificial intelligence image analysis, means for recognizing the user's emotional state, means for performing evaluations and generating notifications based on the identified risk factors and emotional state, and means for transmitting notifications to the user's visual device. This enables real-time identification of risk factors and the provision of appropriate notifications according to the user's emotional state.

[0185] A "surveillance device" is a device used to acquire video information of a monitored area.

[0186] "Visual information" refers to visual data of the environment acquired from monitoring devices.

[0187] An "analyzable format" is a format that has been prepared in a way that is suitable for data analysis.

[0188] "Artificial intelligence image analysis" is a method that uses AI technology to identify the location of an object and potential hazards from video information.

[0189] "Object position" refers to the spatial location occupied by an object within the video information.

[0190] A "risk factor" is an element in the environment that could potentially lead to accidents or problems.

[0191] "Emotion recognition means" refers to a function that analyzes the user's facial expressions and voice from video information to infer and recognize their emotional state.

[0192] "Notifications" are warning or informational messages provided to users based on analysis and evaluation.

[0193] A "visual device" is a display device that a user can wear to directly receive visual information.

[0194] In the system that realizes this invention, a server plays a central role. The server directly receives video information acquired from the monitoring device and first converts that information into an analyzable format. Specifically, it performs preprocessing to optimize the video data, such as adjusting the resolution and removing noise. The software used here could be libraries or tools that are widely used for data processing.

[0195] Next, the server performs artificial intelligence image analysis. Here, deep learning frameworks such as TensorFlow and PyTorch are used to accurately identify the location of objects and potential hazards within the image data. In this image analysis phase, pre-trained models are utilized, and it is also possible to retrain the models in real time as needed.

[0196] In addition, the server uses emotion recognition methods to analyze the user's emotional state. Here, facial recognition technology and voice analysis technology are utilized to determine emotions from the user's facial expressions and vocalizations. Specific tools that can be used include OpenCV and Microsoft® Azure® Emotion API.

[0197] Based on the analysis results, the server generates a notification for the user and sends it to their visual device. This notification instantly appears on the user's wearable device, such as smart glasses. It functions as a push notification, prompting immediate action in response to potential hazards.

[0198] As a concrete example, consider a security officer patrolling a factory at night. If the officer senses danger and detects a suspicious person who has infiltrated the surveillance area, a warning message will immediately appear in the user's field of view. At this time, the user's emotional state will also be taken into consideration, and if necessary, a kind message such as "Please remain calm" will be displayed simultaneously.

[0199] An example of a prompt message might be: "Check the status of windows and doors from the current video data, and immediately output a warning if they are open. If the user is feeling stressed, generate a message to help them relax." This prompt ensures that appropriate information is provided at the right time.

[0200] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0201] Step 1:

[0202] The server receives video information in real time from the monitoring device. It receives raw video data of the monitoring area as input and converts the data into an analyzable format. This process involves pre-processing such as image resolution adjustment and noise reduction to format the data so that it can be easily analyzed by deep learning algorithms. The output is clear video data with pre-processing completed.

[0203] Step 2:

[0204] The server inputs pre-processed video data into an AI image analysis module. The input is formatted video data. The server then analyzes the video data using deep learning models such as TensorFlow or PyTorch to determine the location of objects and identify hazardous factors. A convolutional neural network is used for data processing, and the output consists of object feature data and a list of hazardous factors.

[0205] Step 3:

[0206] The server performs emotion recognition using the user's video and audio data. It uses the user's real-time video and audio data as input. Here, OpenCV and the Microsoft Azure Emotion API are used to infer emotions from facial and voice data, and the emotion engine determines the user's stress and anxiety. The output is the user's emotional state data.

[0207] Step 4:

[0208] The server generates notifications considering the analysis results and emotional state. Inputs include object feature data, a list of risk factors, and emotional state data. Based on the analysis results, a notification is generated suggesting a specific action. Depending on the emotional state, additional messages to reassure the user may also be included. The output is the generated notification data.

[0209] Step 5:

[0210] The server sends the generated notification to the user's visual device. The input is the generated notification data. Here, a push notification protocol is used so that the message appears immediately in the user's field of view. The output is the warning and follow-up message displayed on the user's visual device. This step allows the user to quickly understand the situation and take appropriate action.

[0211] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0212] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0213] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0214] [Second Embodiment]

[0215] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0216] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0217] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0218] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0219] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0220] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0221] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0222] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0223] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0224] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0225] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0226] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0227] This invention relates to a system that uses AI technology to analyze video data acquired using a monitoring device, detects specific dangerous conditions, and notifies the user. The program processing of this system will be explained below using natural language.

[0228] First, the server continuously receives video data transmitted from the monitoring device. This allows it to always have access to the latest information about the monitored area. Next, the received video data is converted to an appropriate resolution and format and pre-processed to a format suitable for AI image analysis.

[0229] Next, the server uses an AI image analysis program to analyze the video data. This analysis identifies the location and state of specific objects and detects potential hazards such as heavy objects in high places or obstacles blocking pathways. Based on this information, the system assesses the risks in emergencies such as earthquakes.

[0230] The server then generates a notification message containing specific improvement suggestions based on the analysis results. For example, if objects are stacked in a particular location, it might create a notification such as, "We recommend moving the heavy object in this location." These notifications are sent to the user's electronic device via the terminal.

[0231] Users can check the situation based on the received notifications and adjust the placement of furniture and objects as needed. This action can mitigate potential hazards and improve the safety of living spaces and offices.

[0232] As a concrete example, when a monitoring system was used in an office, the server detected a large filing box stacked on a metal shelf. Based on this information, AI analysis determined that it posed a risk of falling during an earthquake and generated a notification saying, "Please move the filing box to a location below the shelf." Upon receiving this notification, the user can immediately move the filing box to a safe location, thereby mitigating the risk. In this way, the present invention aims to raise users' safety awareness and prevent accidents.

[0233] The following describes the processing flow.

[0234] Step 1:

[0235] The server receives video data from the monitoring device in real time. The received data is recorded as a secure stream and used for subsequent processing.

[0236] Step 2:

[0237] The server preprocesses the received video data into a format that can be analyzed. This process includes adjusting the resolution and performing noise filtering.

[0238] Step 3:

[0239] The server inputs the pre-processed video data into an AI image analysis module. The artificial intelligence model used here performs object recognition and identifies the placement of furniture and other objects.

[0240] Step 4:

[0241] Based on the results of the AI ​​analysis, the server evaluates the location and condition of identified objects and determines potential hazards. For example, it assesses the presence of heavy objects at high altitudes and the degree of obstruction in passageways.

[0242] Step 5:

[0243] Based on the analysis results and evaluation, the server generates notifications to send to the user. These notifications are prioritized and presented in a message format that specifically outlines improvement suggestions.

[0244] Step 6:

[0245] The device receives notifications from the server and displays them to the user. Notifications are delivered to the user's device as push notifications or in-app messages.

[0246] Step 7:

[0247] The user reviews the notification and makes improvements to the furniture and object placement as needed. After improvements are made, the system is configured to rerun the process for re-evaluation.

[0248] (Example 1)

[0249] Next, we will describe Example 1. 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."

[0250] Conventional video surveillance systems lack the ability to automatically identify and notify of dangerous situations, posing a challenge in preventing risks and ensuring safety in emergency situations where rapid response is required. This results in users being unable to identify potential dangers in a timely manner and take necessary countermeasures.

[0251] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0252] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for converting the video information into an analyzable format, and means for identifying the location and hazardous state of an object using machine intelligence image analysis. This makes it possible to identify hazardous states in real time, immediately notify the user, quickly assess the risk, and take countermeasures.

[0253] A "surveillance device" is a device that visually observes a specific area and continuously acquires video information.

[0254] "Visual information" refers to visual data acquired by surveillance equipment, which is image or video data used for analysis.

[0255] "Machine intelligence image analysis" is a technology that uses deep learning and other advanced algorithms to identify objects in video information and determine their location and state.

[0256] A "hazardous situation" is a situation in which the location or state of an object identified in the video information could potentially pose a risk.

[0257] "Means of risk assessment" refers to the process of determining the degree of risk that an identified hazardous situation may pose and evaluating whether appropriate countermeasures are necessary.

[0258] A "means for generating and communicating attention" is a system that creates messages indicating warnings and countermeasures for users based on the results of a risk assessment, and then electronically sends these messages to the user.

[0259] This invention relates to a system that analyzes video information acquired from a monitoring device and notifies the user of identified dangerous conditions. This system consists of a server, a terminal, and a user, each playing a specific role.

[0260] The server continuously receives video information from monitoring devices using network communication software. The received video information is converted into an AI-analyzable format using an image processing library. This conversion includes resizing and changing the format of the images.

[0261] The converted video information is analyzed on a server using machine intelligence image analysis tools. Deep learning frameworks such as TensorFlow and PyTorch are used to identify the location of objects and their hazardous conditions. For example, this can identify situations where an object is blocking a passageway or where a heavy object is placed at a high position.

[0262] Based on the analysis results, a risk assessment is performed, and specific notifications are generated to improve the hazardous situation. These notifications are sent to the user's electronic device via a terminal. The notification content includes specific instructions such as, "The passage is blocked by an object. Please move the object for safety."

[0263] Users receive notifications from their devices and check the situation on-site based on their content. By taking the instructed actions as needed, they can mitigate potential risks.

[0264] As a concrete example, consider its use in an office. If the server analyzes video information from a monitoring device and identifies a stack of document boxes on a metal shelf, it determines that there is a risk of them falling during an earthquake. A notification is generated and sent to the user saying, "Please move the document boxes under the shelf." This allows the user to move the document boxes to a safe location and mitigate the risk.

[0265] An example of a prompt message would be, "Please tell me how to build a notification system that detects hazardous materials from office security camera footage and assesses the risk."

[0266] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0267] Step 1:

[0268] The server receives video information from the monitoring device. The received video information is provided raw and in real time as a stream. This video information serves as input. Next, an image processing library is used to convert the video information into a format that can be analyzed. Specifically, the image resolution is resized to a size that the AI ​​model can process, and the format is converted to a standard format. As output of this process, image data in a format suitable for AI analysis is generated.

[0269] Step 2:

[0270] The server inputs pre-processed image data into an AI image analysis tool. It performs object recognition using deep learning models such as YOLO and ResNet. Calculations are performed on the input image data to detect the object's location and type. The object detection results are output, including information such as the object's type, location coordinates, and confidence score.

[0271] Step 3:

[0272] The server performs a risk assessment based on the detected object information. The specific input is information about the object's location and type obtained in the previous step. The risk assessment is performed using an expert system, taking into account the object's placement, human movement patterns, and potential hazards in emergencies. The assessment output includes potential hazard areas and priority for countermeasures.

[0273] Step 4:

[0274] The server generates notification messages for the user based on the risk assessment results. The assessment results and recommended actions for the user are used as input. Prompts from the generating AI model are utilized to create notifications that include specific improvement suggestions. Output notifications might include statements such as, "We recommend moving the heavy object located here."

[0275] Step 5:

[0276] The terminal receives notifications sent from the server and displays them to the user. The input is the notification message sent from the server, which is output to the user's electronic device as a push notification or email. Based on the notification content, the user checks the situation on site and takes the instructed actions as needed.

[0277] Step 6:

[0278] Based on the notification, users will identify the areas where risks have been identified. Specific actions might include moving heavy objects from shelves to a safer location. This can reduce potential safety risks and improve the safety of living and working environments.

[0279] (Application Example 1)

[0280] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0281] In industrial facilities, there are potential dangerous situations arising from complex machine arrangements and working environments, and accident prevention measures are required for them. However, at present, methods for detecting these dangers in real time and responding immediately have not been established, and it is difficult to implement efficient safety measures. Furthermore, there is a problem that rapid safety measures cannot be taken because not only the discovery of dangerous situations but also instructions for specific improvement measures for them are lacking.

[0282] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0283] In this invention, the server includes means for receiving image information acquired from a monitoring device, means for converting the image information into an analyzable format, and means for identifying the position of an object and a dangerous situation using machine learning image analysis. As a result, it becomes possible to detect potential dangerous situations in an industrial facility in real time and notify an operator of prompt and specific improvement measures.

[0284] The "monitoring device" is a device for capturing the surrounding situation and continuously acquiring image information.

[0285] The "image information" is data representing the surrounding visual information acquired by the monitoring device.

[0286] The "server" is a computer device that receives information from a plurality of devices and performs processing.

[0287] The "analyzable format" is a data structure and format necessary for performing processing by machine learning techniques.

[0288] The "machine learning image analysis" is an analysis method for recognizing an object in an image and evaluating the situation using advanced techniques such as neural network algorithms.

[0289] The "position of an object" is information indicating the spatial arrangement of a specific object in the observation target.

[0290] A "hazardous situation" refers to any condition or arrangement in an industrial facility that could potentially threaten safety.

[0291] An "operator" is a person responsible for managing and controlling machinery and equipment and maintaining the safety of the facility.

[0292] A "notification" is an informational message generated based on the detected situation and sent to the operator.

[0293] An "information terminal" is a device used by operators to view received notifications.

[0294] The system program for implementing this invention consists of a monitoring device, a server, machine learning software, and a user information terminal.

[0295] The server continuously receives image information transmitted from monitoring devices. This image information shows the current arrangement and status of machinery and objects within the facility. The server first converts this information into a format that can be used for machine learning image analysis. This involves image format conversion and resolution adjustment.

[0296] Next, the server performs machine learning image analysis. Specifically, a neural network algorithm is used to recognize objects in the image and determine their location. After that, it identifies dangerous situations and identifies areas for improvement as needed. Frameworks such as TensorFlow are used for this analysis.

[0297] Next, the server generates a notification containing specific instructions to encourage corrective action based on the identified risk situation. This notification is sent as a push message to the operator's information terminal. The notification is communicated using services such as Firebase Cloud Messaging.

[0298] As a concrete example, consider a situation in a factory where a robotic arm is blocking a passageway and obstructing an emergency evacuation route. A monitoring device detects this situation, and a server performs real-time analysis. The analysis results are notified in the form of "We recommend adjusting the position of the robotic arm" and sent to the operator's information terminal. This enables a rapid response.

[0299] An example of a prompt to input into the generating AI model would be: "Explain the notification generation process for a system that uses AI to analyze video data inside a factory, detect dangerous situations, and generate notifications."

[0300] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0301] Step 1:

[0302] The server receives image information transmitted from the monitoring device. The input is real-time image data of the factory captured by the monitoring device. At this stage, the image data is retained in its original format. The output is the unprocessed image data stored in temporary storage for data retention.

[0303] Step 2:

[0304] The server converts received image information into a format suitable for analysis. The input is raw image data. Data processing is performed to change the image format to an appropriate format such as JPEG or PNG, and to reduce the resolution to a size suitable for machine learning models. The output is image data converted into a format suitable for analysis.

[0305] Step 3:

[0306] The server performs machine learning image analysis. The input is the image data converted in the previous step. Here, the server applies a neural network model using a framework such as TensorFlow to perform object recognition in the image and identify dangerous situations. The output is the position information of the object and data related to the dangerous situation.

[0307] Step 4:

[0308] The server evaluates what needs to be improved based on the dangerous situation and generates specific improvement instructions. The input is the position information of the identified object and the dangerous situation data. Analyze the data to create an evaluation of specific dangers such as "the passage is blocked" and response instructions such as "adjust the position of the robotic arm". The output is the notification message data including the improvement instructions.

[0309] Step 5:

[0310] The server sends the generated notification to the operator's information terminal. The input is the notification message data including the improvement instructions. Utilize a push notification service such as Firebase Cloud Messaging to notify the operator's terminal in real time. The output is the notification message displayed on the screen of the operator's information terminal.

[0311] Step 6:

[0312] The operator checks the information notified on the terminal and makes adjustments such as machine placement according to the instructions. The input is the notification message displayed on the terminal. The operator checks the site based on the notification content and makes specific responses such as moving the object or adjusting the position of the machine. The output is the new layout situation within the facility after the safety has been improved.

[0313] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion specific model 59 and perform specific processing using the user's emotion.

[0314] This invention provides a system that identifies hazardous elements in the environment and provides appropriate notifications to the user by combining video data acquisition from a monitoring device with AI image analysis. Furthermore, it incorporates an emotion engine that recognizes the user's emotional state and individually optimizes user interaction. The processing of this system's program is described in detail below in natural language.

[0315] First, the server receives video data acquired from the monitoring device and preprocesses it into a format suitable for analysis. The preprocessed data is then input into the AI ​​image analysis module to identify the placement of objects and potential hazards in the environment. In this identification process, deep learning algorithms are used to achieve highly accurate object recognition.

[0316] Next, the server evaluates the identified risk conditions based on the analysis results and creates improvement plans. In this process, it uses an emotion engine to simultaneously recognize the user's emotional state from their facial expressions and voice. If the user is feeling anxious or stressed, the server adjusts the content and tone of notifications accordingly and adds information to reassure the user.

[0317] Next, the server formats the generated notification in the most suitable format for the user and sends it to the user's digital device via the terminal. The notification is delivered using the most effective method, such as push notification or email. The content of this notification also takes into account the results of the emotion engine's analysis, enabling interaction that takes into account the user's emotional state.

[0318] As a concrete example, consider the use of the system in an office. If an employee is sitting at their desk with an anxious expression, the server will recognize this emotion and, in addition to the usual safety notification, send a notification that includes emotionally supportive content, such as, "For your safety, we will provide you with the most relaxed environment possible. Please let us know if you need any support." In this way, the system can provide the user with the most appropriate support.

[0319] The following describes the processing flow.

[0320] Step 1:

[0321] The server receives video data in real time from the monitoring devices. This allows it to collect basic data to understand the latest situation in the monitored area.

[0322] Step 2:

[0323] The server preprocesses the received video data, performing noise reduction and resolution adjustments, and converts it into a format suitable for AI image analysis.

[0324] Step 3:

[0325] The server inputs the pre-processed video data into an AI image analysis module. This AI module utilizes deep learning to perform object recognition and location determination with high accuracy.

[0326] Step 4:

[0327] The server assesses potential hazards based on the placement of objects identified from the analysis results. This assessment includes the risk of objects falling during an earthquake and the possibility of difficulty in evacuation due to obstacles in pathways.

[0328] Step 5:

[0329] The server simultaneously uses an emotion engine to recognize the user's emotional state. It analyzes input data from cameras and audio sensors to identify emotions from the user's facial expressions and voice.

[0330] Step 6:

[0331] The server generates notification content based on the risk assessment results and the user's emotional state. This notification includes not only conventional safety measures but also messages and improvement suggestions that take the user's feelings into consideration.

[0332] Step 7:

[0333] The device receives notifications generated from the server and displays them as push notifications on the user's electronic device. These notifications are delivered at a timing and tone that corresponds to the user's emotional state.

[0334] Step 8:

[0335] Users review the displayed notifications and take appropriate action to ensure the safety of the facility as needed. At this stage, emotions are re-evaluated, allowing for more appropriate support to be provided.

[0336] (Example 2)

[0337] Next, we will describe Example 2. 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".

[0338] Modern, high-performance monitoring systems are required not only to identify hazardous elements in the environment, but also to provide individually optimized notifications tailored to the user's emotional state. However, conventional technologies have struggled to achieve such flexible notifications that take user emotions into account. Therefore, a new approach is needed to improve both safety and user experience.

[0339] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0340] In this invention, the server includes a configuration for receiving video information acquired from a monitoring device, a configuration for converting the video information into an analyzable format, a configuration for identifying the arrangement of objects and dangerous conditions using machine learning image analysis, a configuration for recognizing the user's emotional state using an emotion analysis engine, and a configuration for generating and sending notifications to the user based on the evaluation and emotional state. This makes it possible to provide safe and individually optimized notifications that take the user's emotions into consideration.

[0341] A "monitoring device" is a device used to continuously monitor the environment and acquire video information.

[0342] "Visual information" refers to visual data provided by surveillance equipment, including image and video data.

[0343] "Machine learning image analysis" is a method of analyzing video information using machine learning techniques to identify the arrangement of objects and specific patterns.

[0344] An "emotion analysis engine" is a software configuration that determines a user's emotional state based on their facial expressions and voice data.

[0345] A "notification" is an informational message sent to the user based on the analysis results, which may include warnings or suggestions.

[0346] "Push-out notifications" are a notification method that displays information from the system in real time on the user's electronic device.

[0347] "Electronic communication" refers to the method of transmitting digital data between electronic devices, such as using email or messaging apps.

[0348] "Object arrangement" refers to information about the position and state of objects within an environment.

[0349] A "hazardous condition" refers to a situation or condition that could potentially pose a danger to the user or the environment.

[0350] This invention is a monitoring system that identifies hazardous elements in the environment and provides users with individually optimized notifications that respond to their emotions.

[0351] The server first receives video information in real time from the monitoring device. This data is preprocessed into an appropriate format before being analyzed as raw data. Specifically, processes such as data format conversion and noise reduction are performed. This prepares the image analysis module for high-precision analysis.

[0352] Next, using the processed data, the server identifies object placement and hazardous conditions within the environment through machine learning image analysis. Examples of algorithms used include deep learning frameworks such as TensorFlow and PyTorch, with YOLO and Faster R-CNN being used for object identification. During this process, objects identified as hazardous elements are labeled and recorded in a database.

[0353] Furthermore, the server uses an emotion analysis engine to recognize the user's emotional state from their facial expressions and voice data. Speech recognition technology (for example, technology that transcribes speech into text) and facial expression analysis software are used.

[0354] Based on these analysis results, the server generates a notification for the user and sends it to the user's electronic device via the terminal. The notification includes specific actions and support information tailored to the user's situation and emotions. This allows the user to confidently follow the instructions. Notifications are typically delivered to the user as push notifications or email.

[0355] For example, when this system is used in a business environment such as an office, it sends a notification prompting immediate action if a hazardous element is detected. Furthermore, when employees are experiencing stress, it provides emotionally supportive recommendations such as "relax and maintain your health."

[0356] An example of a prompt message might be, "Please tell me how to identify potential hazards in the meeting room using video analysis and how to set up notifications based on employee emotion recognition."

[0357] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0358] Step 1:

[0359] The server receives video information from the monitoring device in real time. This input data is typically in video stream format (e.g., MP4 or AVI). The received data is stored in a temporary cache as a preliminary step before analysis. This ensures that the data is always available for subsequent processing.

[0360] Step 2:

[0361] The server preprocesses the received video information into a format that can be analyzed. The input is video stream data, and the output is image data converted to a still image format (JPEG or PNG). This process involves data format conversion and noise reduction. Specifically, a Gaussian filter is applied to reduce noise and clarify the image.

[0362] Step 3:

[0363] The server performs machine learning image analysis using pre-processed data. The input is still image data, and the output is location information (bounding boxes and labels) of identified objects. Using TensorFlow or PyTorch, models such as YOLO and Faster R-CNN are used to accurately identify the location and type of objects.

[0364] Step 4:

[0365] The server evaluates the hazardous conditions within the environment based on the analysis results. The input is information on the placement of identified objects, and the output is a hazard assessment report. This assessment uses pre-defined safety standards and rules to identify high-risk situations and conditions.

[0366] Step 5:

[0367] The server analyzes the user's emotional state using an emotion analysis engine. Input is the user's facial expressions and voice data, and output is the emotional state (e.g., anxiety, relaxation). Speech recognition and facial expression analysis software are used to recognize the user's emotions according to their situation.

[0368] Step 6:

[0369] The server generates notifications based on the risk level and the user's emotions. Inputs are a risk assessment report and emotional state, while output is a customized notification message. The notification includes specific actions and emotional support to encourage the user to respond confidently.

[0370] Step 7:

[0371] The device sends notifications generated from the server to the user's electronic device. The input is a customized notification message, and the output is a notification received on a smartphone or PC. Notifications are delivered in the form of push notifications or email and are presented to the user in real time.

[0372] (Application Example 2)

[0373] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0374] Conventional monitoring systems have a simplistic and general approach to identifying risk factors and notifying users, making it difficult to respond flexibly to different situations and user emotional states. Therefore, challenges include a lack of real-time response capabilities to risks within the monitoring area and a lack of individualized stress reduction measures for users.

[0375] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0376] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for processing the video information into an analyzable format, means for identifying the location of objects and risk factors using artificial intelligence image analysis, means for recognizing the user's emotional state, means for performing evaluations and generating notifications based on the identified risk factors and emotional state, and means for transmitting notifications to the user's visual device. This enables real-time identification of risk factors and the provision of appropriate notifications according to the user's emotional state.

[0377] A "surveillance device" is a device used to acquire video information of a monitored area.

[0378] "Visual information" refers to visual data of the environment acquired from monitoring devices.

[0379] An "analyzable format" is a format that has been prepared in a way that is suitable for data analysis.

[0380] "Artificial intelligence image analysis" is a method that uses AI technology to identify the location of an object and potential hazards from video information.

[0381] "Object position" refers to the spatial location occupied by an object within the video information.

[0382] A "risk factor" is an element in the environment that could potentially lead to accidents or problems.

[0383] "Emotion recognition means" refers to a function that analyzes the user's facial expressions and voice from video information to infer and recognize their emotional state.

[0384] "Notifications" are warning or informational messages provided to users based on analysis and evaluation.

[0385] A "visual device" is a display device that a user can wear to directly receive visual information.

[0386] In the system that realizes this invention, a server plays a central role. The server directly receives video information acquired from the monitoring device and first converts that information into an analyzable format. Specifically, it performs preprocessing to optimize the video data, such as adjusting the resolution and removing noise. The software used here could be libraries or tools that are widely used for data processing.

[0387] Next, the server performs artificial intelligence image analysis. Here, deep learning frameworks such as TensorFlow and PyTorch are used to accurately identify the location of objects and potential hazards within the image data. In this image analysis phase, pre-trained models are utilized, and it is also possible to retrain the models in real time as needed.

[0388] In addition, the server uses emotion recognition methods to analyze the user's emotional state. Here, facial recognition technology and voice analysis technology are utilized to determine emotions from the user's facial expressions and vocalizations. Specific tools that can be used include OpenCV and the Microsoft Azure Emotion API.

[0389] Based on the analysis results, the server generates a notification for the user and sends it to their visual device. This notification instantly appears on the user's wearable device, such as smart glasses. It functions as a push notification, prompting immediate action in response to potential hazards.

[0390] As a concrete example, consider a security officer patrolling a factory at night. If the officer senses danger and detects a suspicious person who has infiltrated the surveillance area, a warning message will immediately appear in the user's field of view. At this time, the user's emotional state will also be taken into consideration, and if necessary, a kind message such as "Please remain calm" will be displayed simultaneously.

[0391] An example of a prompt message might be: "Check the status of windows and doors from the current video data, and immediately output a warning if they are open. If the user is feeling stressed, generate a message to help them relax." This prompt ensures that appropriate information is provided at the right time.

[0392] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0393] Step 1:

[0394] The server receives video information in real time from the monitoring device. It receives raw video data of the monitoring area as input and converts the data into an analyzable format. This process involves pre-processing such as image resolution adjustment and noise reduction to format the data so that it can be easily analyzed by deep learning algorithms. The output is clear video data with pre-processing completed.

[0395] Step 2:

[0396] The server inputs pre-processed video data into an AI image analysis module. The input is formatted video data. The server then analyzes the video data using deep learning models such as TensorFlow or PyTorch to determine the location of objects and identify hazardous factors. A convolutional neural network is used for data processing, and the output consists of object feature data and a list of hazardous factors.

[0397] Step 3:

[0398] The server performs emotion recognition using the user's video and audio data. It uses the user's real-time video and audio data as input. Here, OpenCV and the Microsoft Azure Emotion API are used to infer emotions from facial and voice data, and the emotion engine determines the user's stress and anxiety. The output is the user's emotional state data.

[0399] Step 4:

[0400] The server generates notifications considering the analysis results and emotional state. Inputs include object feature data, a list of risk factors, and emotional state data. Based on the analysis results, a notification is generated suggesting a specific action. Depending on the emotional state, additional messages to reassure the user may also be included. The output is the generated notification data.

[0401] Step 5:

[0402] The server sends the generated notification to the user's visual device. The input is the generated notification data. Here, a push notification protocol is used so that the message appears immediately in the user's field of view. The output is the warning and follow-up message displayed on the user's visual device. This step allows the user to quickly understand the situation and take appropriate action.

[0403] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0404] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0405] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0406] [Third Embodiment]

[0407] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0408] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0409] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0410] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0411] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0412] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0413] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0414] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0415] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0416] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0417] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0418] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0419] This invention relates to a system that uses AI technology to analyze video data acquired using a monitoring device, detects specific dangerous conditions, and notifies the user. The program processing of this system will be explained below using natural language.

[0420] First, the server continuously receives video data transmitted from the monitoring device. This allows it to always have access to the latest information about the monitored area. Next, the received video data is converted to an appropriate resolution and format and pre-processed to a format suitable for AI image analysis.

[0421] Next, the server uses an AI image analysis program to analyze the video data. This analysis identifies the location and state of specific objects and detects potential hazards such as heavy objects in high places or obstacles blocking pathways. Based on this information, the system assesses the risks in emergencies such as earthquakes.

[0422] The server then generates a notification message containing specific improvement suggestions based on the analysis results. For example, if objects are stacked in a particular location, it might create a notification such as, "We recommend moving the heavy object in this location." These notifications are sent to the user's electronic device via the terminal.

[0423] Users can check the situation based on the received notifications and adjust the placement of furniture and objects as needed. This action can mitigate potential hazards and improve the safety of living spaces and offices.

[0424] As a concrete example, when a monitoring system was used in an office, the server detected a large filing box stacked on a metal shelf. Based on this information, AI analysis determined that it posed a risk of falling during an earthquake and generated a notification saying, "Please move the filing box to a location below the shelf." Upon receiving this notification, the user can immediately move the filing box to a safe location, thereby mitigating the risk. In this way, the present invention aims to raise users' safety awareness and prevent accidents.

[0425] The following describes the processing flow.

[0426] Step 1:

[0427] The server receives video data from the monitoring device in real time. The received data is recorded as a secure stream and used for subsequent processing.

[0428] Step 2:

[0429] The server preprocesses the received video data into a format that can be analyzed. This process includes adjusting the resolution and performing noise filtering.

[0430] Step 3:

[0431] The server inputs the pre-processed video data into an AI image analysis module. The artificial intelligence model used here performs object recognition and identifies the placement of furniture and other objects.

[0432] Step 4:

[0433] Based on the results of the AI ​​analysis, the server evaluates the location and condition of identified objects and determines potential hazards. For example, it assesses the presence of heavy objects at high altitudes and the degree of obstruction in passageways.

[0434] Step 5:

[0435] Based on the analysis results and evaluation, the server generates notifications to send to the user. These notifications are prioritized and presented in a message format that specifically outlines improvement suggestions.

[0436] Step 6:

[0437] The device receives notifications from the server and displays them to the user. Notifications are delivered to the user's device as push notifications or in-app messages.

[0438] Step 7:

[0439] The user reviews the notification and makes improvements to the furniture and object placement as needed. After improvements are made, the system is configured to rerun the process for re-evaluation.

[0440] (Example 1)

[0441] Next, we will describe Example 1. 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."

[0442] Conventional video surveillance systems lack the ability to automatically identify and notify of dangerous situations, posing a challenge in preventing risks and ensuring safety in emergency situations where rapid response is required. This results in users being unable to identify potential dangers in a timely manner and take necessary countermeasures.

[0443] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0444] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for converting the video information into an analyzable format, and means for identifying the location and hazardous state of an object using machine intelligence image analysis. This makes it possible to identify hazardous states in real time, immediately notify the user, quickly assess the risk, and take countermeasures.

[0445] A "surveillance device" is a device that visually observes a specific area and continuously acquires video information.

[0446] "Visual information" refers to visual data acquired by surveillance equipment, which is image or video data used for analysis.

[0447] "Machine intelligence image analysis" is a technology that uses deep learning and other advanced algorithms to identify objects in video information and determine their location and state.

[0448] A "hazardous situation" is a situation in which the location or state of an object identified in the video information could potentially pose a risk.

[0449] "Means of risk assessment" refers to the process of determining the degree of risk that an identified hazardous situation may pose and evaluating whether appropriate countermeasures are necessary.

[0450] A "means for generating and communicating attention" is a system that creates messages indicating warnings and countermeasures for users based on the results of a risk assessment, and then electronically sends these messages to the user.

[0451] This invention relates to a system that analyzes video information acquired from a monitoring device and notifies the user of identified dangerous conditions. This system consists of a server, a terminal, and a user, each playing a specific role.

[0452] The server continuously receives video information from monitoring devices using network communication software. The received video information is converted into an AI-analyzable format using an image processing library. This conversion includes resizing and changing the format of the images.

[0453] The converted video information is analyzed on a server using machine intelligence image analysis tools. Deep learning frameworks such as TensorFlow and PyTorch are used to identify the location of objects and their hazardous conditions. For example, this can identify situations where an object is blocking a passageway or where a heavy object is placed at a high position.

[0454] Based on the analysis results, a risk assessment is performed, and specific notifications are generated to improve the hazardous situation. These notifications are sent to the user's electronic device via a terminal. The notification content includes specific instructions such as, "The passage is blocked by an object. Please move the object for safety."

[0455] Users receive notifications from their devices and check the situation on-site based on their content. By taking the instructed actions as needed, they can mitigate potential risks.

[0456] As a concrete example, consider its use in an office. If the server analyzes video information from a monitoring device and identifies a stack of document boxes on a metal shelf, it determines that there is a risk of them falling during an earthquake. A notification is generated and sent to the user saying, "Please move the document boxes under the shelf." This allows the user to move the document boxes to a safe location and mitigate the risk.

[0457] An example of a prompt message would be, "Please tell me how to build a notification system that detects hazardous materials from office security camera footage and assesses the risk."

[0458] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0459] Step 1:

[0460] The server receives video information from the monitoring device. The received video information is provided raw and in real time as a stream. This video information serves as input. Next, an image processing library is used to convert the video information into a format that can be analyzed. Specifically, the image resolution is resized to a size that the AI ​​model can process, and the format is converted to a standard format. As output of this process, image data in a format suitable for AI analysis is generated.

[0461] Step 2:

[0462] The server inputs pre-processed image data into an AI image analysis tool. It performs object recognition using deep learning models such as YOLO and ResNet. Calculations are performed on the input image data to detect the object's location and type. The object detection results are output, including information such as the object's type, location coordinates, and confidence score.

[0463] Step 3:

[0464] The server performs a risk assessment based on the detected object information. The specific input is information about the object's location and type obtained in the previous step. The risk assessment is performed using an expert system, taking into account the object's placement, human movement patterns, and potential hazards in emergencies. The assessment output includes potential hazard areas and priority for countermeasures.

[0465] Step 4:

[0466] The server generates notification messages for the user based on the risk assessment results. The assessment results and recommended actions for the user are used as input. Prompts from the generating AI model are utilized to create notifications that include specific improvement suggestions. Output notifications might include statements such as, "We recommend moving the heavy object located here."

[0467] Step 5:

[0468] The terminal receives notifications sent from the server and displays them to the user. The input is the notification message sent from the server, which is output to the user's electronic device as a push notification or email. Based on the notification content, the user checks the situation on site and takes the instructed actions as needed.

[0469] Step 6:

[0470] Based on the notification, users will identify the areas where risks have been identified. Specific actions might include moving heavy objects from shelves to a safer location. This can reduce potential safety risks and improve the safety of living and working environments.

[0471] (Application Example 1)

[0472] Next, we will explain Application Example 1. In the following explanation, 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."

[0473] Industrial facilities present potential hazards arising from complex machine layouts and work environments, necessitating accident prevention measures. However, currently, there is no established method for detecting these hazards in real time and responding immediately, making efficient safety measures difficult. Furthermore, the lack of not only detection of hazards but also instructions for specific corrective measures hinders the rapid implementation of safety measures.

[0474] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0475] In this invention, the server includes means for receiving image information acquired from a monitoring device, means for converting the image information into an analyzable format, and means for identifying the location of an object and its hazardous status using machine learning image analysis. This makes it possible to detect potential hazardous conditions within industrial facilities in real time and to quickly notify operators of specific corrective measures.

[0476] A "monitoring device" is a device that captures the surrounding environment and continuously acquires image information.

[0477] "Image information" refers to data that represents the visual information of the surroundings, acquired by a monitoring device.

[0478] A "server" is a computer device that receives and processes information from multiple devices.

[0479] A "parsable format" refers to the data structure and format necessary for processing using machine learning techniques.

[0480] "Machine learning image analysis" is an analytical method that uses advanced technologies such as neural network algorithms to recognize objects in an image and evaluate the situation.

[0481] "Object position" refers to information that indicates the spatial arrangement of a specific object within the observed area.

[0482] A "hazardous situation" refers to any condition or arrangement in an industrial facility that could potentially threaten safety.

[0483] An "operator" is a person responsible for managing and controlling machinery and equipment and maintaining the safety of the facility.

[0484] A "notification" is an informational message generated based on the detected situation and sent to the operator.

[0485] An "information terminal" is a device used by operators to view received notifications.

[0486] The system program for implementing this invention consists of a monitoring device, a server, machine learning software, and a user information terminal.

[0487] The server continuously receives image information transmitted from monitoring devices. This image information shows the current arrangement and status of machinery and objects within the facility. The server first converts this information into a format that can be used for machine learning image analysis. This involves image format conversion and resolution adjustment.

[0488] Next, the server performs machine learning image analysis. Specifically, a neural network algorithm is used to recognize objects in the image and determine their location. After that, it identifies dangerous situations and identifies areas for improvement as needed. Frameworks such as TensorFlow are used for this analysis.

[0489] Next, the server generates a notification containing specific instructions to encourage corrective action based on the identified risk situation. This notification is sent as a push message to the operator's information terminal. The notification is communicated using services such as Firebase Cloud Messaging.

[0490] As a concrete example, consider a situation in a factory where a robotic arm is blocking a passageway and obstructing an emergency evacuation route. A monitoring device detects this situation, and a server performs real-time analysis. The analysis results are notified in the form of "We recommend adjusting the position of the robotic arm" and sent to the operator's information terminal. This enables a rapid response.

[0491] An example of a prompt to input into the generating AI model would be: "Explain the notification generation process for a system that uses AI to analyze video data inside a factory, detect dangerous situations, and generate notifications."

[0492] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0493] Step 1:

[0494] The server receives image information transmitted from the monitoring device. The input is real-time image data of the factory captured by the monitoring device. At this stage, the image data is retained in its original format. The output is the unprocessed image data stored in temporary storage for data retention.

[0495] Step 2:

[0496] The server converts received image information into a format suitable for analysis. The input is raw image data. Data processing is performed to change the image format to an appropriate format such as JPEG or PNG, and to reduce the resolution to a size suitable for machine learning models. The output is image data converted into a format suitable for analysis.

[0497] Step 3:

[0498] The server performs machine learning image analysis. The input is the image data transformed in the previous step. Here, the server applies a neural network model using a framework such as TensorFlow to recognize objects and identify hazardous situations within the image. The output is data on the location of objects and hazardous situations.

[0499] Step 4:

[0500] The server evaluates what needs to be improved based on the hazard situation and generates specific improvement instructions. The input is the location information of the identified object and hazard situation data. The data is analyzed to create an assessment of specific hazards, such as "the passage is blocked," and response instructions, such as "adjust the position of the robot arm." The output is notification message data, including the improvement instructions.

[0501] Step 5:

[0502] The server sends the generated notification to the operator's information terminal. The input is notification message data, including improvement instructions. Push notification services such as Firebase Cloud Messaging are used to send notifications to the operator's terminal in real time. The output is the notification message displayed on the operator's information terminal screen.

[0503] Step 6:

[0504] The operator checks the information notified on the terminal and makes adjustments to the machine layout, etc., according to the instructions. The input is the notification message displayed on the terminal. Based on the notification content, the operator checks the site and takes specific actions such as moving objects or adjusting the position of machinery. The output is the new layout within the facility after safety has been improved.

[0505] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0506] This invention provides a system that identifies hazardous elements in the environment and provides appropriate notifications to the user by combining video data acquisition from a monitoring device with AI image analysis. Furthermore, it incorporates an emotion engine that recognizes the user's emotional state and individually optimizes user interaction. The processing of this system's program is described in detail below in natural language.

[0507] First, the server receives video data acquired from the monitoring device and preprocesses it into a format suitable for analysis. The preprocessed data is then input into the AI ​​image analysis module to identify the placement of objects and potential hazards in the environment. In this identification process, deep learning algorithms are used to achieve highly accurate object recognition.

[0508] Next, the server evaluates the identified risk conditions based on the analysis results and creates improvement plans. In this process, it uses an emotion engine to simultaneously recognize the user's emotional state from their facial expressions and voice. If the user is feeling anxious or stressed, the server adjusts the content and tone of notifications accordingly and adds information to reassure the user.

[0509] Next, the server formats the generated notification in the most suitable format for the user and sends it to the user's digital device via the terminal. The notification is delivered using the most effective method, such as push notification or email. The content of this notification also takes into account the results of the emotion engine's analysis, enabling interaction that takes into account the user's emotional state.

[0510] As a concrete example, consider the use of the system in an office. If an employee is sitting at their desk with an anxious expression, the server will recognize this emotion and, in addition to the usual safety notification, send a notification that includes emotionally supportive content, such as, "For your safety, we will provide you with the most relaxed environment possible. Please let us know if you need any support." In this way, the system can provide the user with the most appropriate support.

[0511] The following describes the processing flow.

[0512] Step 1:

[0513] The server receives video data in real time from the monitoring devices. This allows it to collect basic data to understand the latest situation in the monitored area.

[0514] Step 2:

[0515] The server preprocesses the received video data, performing noise reduction and resolution adjustments, and converts it into a format suitable for AI image analysis.

[0516] Step 3:

[0517] The server inputs the pre-processed video data into an AI image analysis module. This AI module utilizes deep learning to perform object recognition and location determination with high accuracy.

[0518] Step 4:

[0519] The server assesses potential hazards based on the placement of objects identified from the analysis results. This assessment includes the risk of objects falling during an earthquake and the possibility of difficulty in evacuation due to obstacles in pathways.

[0520] Step 5:

[0521] The server simultaneously uses an emotion engine to recognize the user's emotional state. It analyzes input data from cameras and audio sensors to identify emotions from the user's facial expressions and voice.

[0522] Step 6:

[0523] The server generates notification content based on the risk assessment results and the user's emotional state. This notification includes not only conventional safety measures but also messages and improvement suggestions that take the user's feelings into consideration.

[0524] Step 7:

[0525] The device receives notifications generated from the server and displays them as push notifications on the user's electronic device. These notifications are delivered at a timing and tone that corresponds to the user's emotional state.

[0526] Step 8:

[0527] Users review the displayed notifications and take appropriate action to ensure the safety of the facility as needed. At this stage, emotions are re-evaluated, allowing for more appropriate support to be provided.

[0528] (Example 2)

[0529] Next, we will describe Example 2. 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."

[0530] Modern, high-performance monitoring systems are required not only to identify hazardous elements in the environment, but also to provide individually optimized notifications tailored to the user's emotional state. However, conventional technologies have struggled to achieve such flexible notifications that take user emotions into account. Therefore, a new approach is needed to improve both safety and user experience.

[0531] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0532] In this invention, the server includes a configuration for receiving video information acquired from a monitoring device, a configuration for converting the video information into an analyzable format, a configuration for identifying the arrangement of objects and dangerous conditions using machine learning image analysis, a configuration for recognizing the user's emotional state using an emotion analysis engine, and a configuration for generating and sending notifications to the user based on the evaluation and emotional state. This makes it possible to provide safe and individually optimized notifications that take the user's emotions into consideration.

[0533] A "monitoring device" is a device used to continuously monitor the environment and acquire video information.

[0534] "Visual information" refers to visual data provided by surveillance equipment, including image and video data.

[0535] "Machine learning image analysis" is a method of analyzing video information using machine learning techniques to identify the arrangement of objects and specific patterns.

[0536] An "emotion analysis engine" is a software configuration that determines a user's emotional state based on their facial expressions and voice data.

[0537] A "notification" is an informational message sent to the user based on the analysis results, which may include warnings or suggestions.

[0538] "Push-out notifications" are a notification method that displays information from the system in real time on the user's electronic device.

[0539] "Electronic communication" refers to the method of transmitting digital data between electronic devices, such as using email or messaging apps.

[0540] "Object arrangement" refers to information about the position and state of objects within an environment.

[0541] A "hazardous condition" refers to a situation or condition that could potentially pose a danger to the user or the environment.

[0542] This invention is a monitoring system that identifies hazardous elements in the environment and provides users with individually optimized notifications that respond to their emotions.

[0543] The server first receives video information in real time from the monitoring device. This data is preprocessed into an appropriate format before being analyzed as raw data. Specifically, processes such as data format conversion and noise reduction are performed. This prepares the image analysis module for high-precision analysis.

[0544] Next, using the processed data, the server identifies object placement and hazardous conditions within the environment through machine learning image analysis. Examples of algorithms used include deep learning frameworks such as TensorFlow and PyTorch, with YOLO and Faster R-CNN being used for object identification. During this process, objects identified as hazardous elements are labeled and recorded in a database.

[0545] Furthermore, the server uses an emotion analysis engine to recognize the user's emotional state from their facial expressions and voice data. Speech recognition technology (for example, technology that transcribes speech into text) and facial expression analysis software are used.

[0546] Based on these analysis results, the server generates a notification for the user and sends it to the user's electronic device via the terminal. The notification includes specific actions and support information tailored to the user's situation and emotions. This allows the user to confidently follow the instructions. Notifications are typically delivered to the user as push notifications or email.

[0547] For example, when this system is used in a business environment such as an office, it sends a notification prompting immediate action if a hazardous element is detected. Furthermore, when employees are experiencing stress, it provides emotionally supportive recommendations such as "relax and maintain your health."

[0548] An example of a prompt message might be, "Please tell me how to identify potential hazards in the meeting room using video analysis and how to set up notifications based on employee emotion recognition."

[0549] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0550] Step 1:

[0551] The server receives video information from the monitoring device in real time. This input data is typically in video stream format (e.g., MP4 or AVI). The received data is stored in a temporary cache as a preliminary step before analysis. This ensures that the data is always available for subsequent processing.

[0552] Step 2:

[0553] The server preprocesses the received video information into a format that can be analyzed. The input is video stream data, and the output is image data converted to a still image format (JPEG or PNG). This process involves data format conversion and noise reduction. Specifically, a Gaussian filter is applied to reduce noise and clarify the image.

[0554] Step 3:

[0555] The server performs machine learning image analysis using pre-processed data. The input is still image data, and the output is location information (bounding boxes and labels) of identified objects. Using TensorFlow or PyTorch, models such as YOLO and Faster R-CNN are used to accurately identify the location and type of objects.

[0556] Step 4:

[0557] The server evaluates the hazardous conditions within the environment based on the analysis results. The input is information on the placement of identified objects, and the output is a hazard assessment report. This assessment uses pre-defined safety standards and rules to identify high-risk situations and conditions.

[0558] Step 5:

[0559] The server analyzes the user's emotional state using an emotion analysis engine. Input is the user's facial expressions and voice data, and output is the emotional state (e.g., anxiety, relaxation). Speech recognition and facial expression analysis software are used to recognize the user's emotions according to their situation.

[0560] Step 6:

[0561] The server generates notifications based on the risk level and the user's emotions. Inputs are a risk assessment report and emotional state, while output is a customized notification message. The notification includes specific actions and emotional support to encourage the user to respond confidently.

[0562] Step 7:

[0563] The device sends notifications generated from the server to the user's electronic device. The input is a customized notification message, and the output is a notification received on a smartphone or PC. Notifications are delivered in the form of push notifications or email and are presented to the user in real time.

[0564] (Application Example 2)

[0565] Next, we will explain application example 2. In the following explanation, 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."

[0566] Conventional monitoring systems have a simplistic and general approach to identifying risk factors and notifying users, making it difficult to respond flexibly to different situations and user emotional states. Therefore, challenges include a lack of real-time response capabilities to risks within the monitoring area and a lack of individualized stress reduction measures for users.

[0567] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0568] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for processing the video information into an analyzable format, means for identifying the location of objects and risk factors using artificial intelligence image analysis, means for recognizing the user's emotional state, means for performing evaluations and generating notifications based on the identified risk factors and emotional state, and means for transmitting notifications to the user's visual device. This enables real-time identification of risk factors and the provision of appropriate notifications according to the user's emotional state.

[0569] A "surveillance device" is a device used to acquire video information of a monitored area.

[0570] "Visual information" refers to visual data of the environment acquired from monitoring devices.

[0571] An "analyzable format" is a format that has been prepared in a way that is suitable for data analysis.

[0572] "Artificial intelligence image analysis" is a method that uses AI technology to identify the location of an object and potential hazards from video information.

[0573] "Object position" refers to the spatial location occupied by an object within the video information.

[0574] A "risk factor" is an element in the environment that could potentially lead to accidents or problems.

[0575] "Emotion recognition means" refers to a function that analyzes the user's facial expressions and voice from video information to infer and recognize their emotional state.

[0576] "Notifications" are warning or informational messages provided to users based on analysis and evaluation.

[0577] A "visual device" is a display device that a user can wear to directly receive visual information.

[0578] In the system that realizes this invention, a server plays a central role. The server directly receives video information acquired from the monitoring device and first converts that information into an analyzable format. Specifically, it performs preprocessing to optimize the video data, such as adjusting the resolution and removing noise. The software used here could be libraries or tools that are widely used for data processing.

[0579] Next, the server performs artificial intelligence image analysis. Here, deep learning frameworks such as TensorFlow and PyTorch are used to accurately identify the location of objects and potential hazards within the image data. In this image analysis phase, pre-trained models are utilized, and it is also possible to retrain the models in real time as needed.

[0580] In addition, the server uses emotion recognition methods to analyze the user's emotional state. Here, facial recognition technology and voice analysis technology are utilized to determine emotions from the user's facial expressions and vocalizations. Specific tools that can be used include OpenCV and the Microsoft Azure Emotion API.

[0581] Based on the analysis results, the server generates a notification for the user and sends it to their visual device. This notification instantly appears on the user's wearable device, such as smart glasses. It functions as a push notification, prompting immediate action in response to potential hazards.

[0582] As a concrete example, consider a security officer patrolling a factory at night. If the officer senses danger and detects a suspicious person who has infiltrated the surveillance area, a warning message will immediately appear in the user's field of view. At this time, the user's emotional state will also be taken into consideration, and if necessary, a kind message such as "Please remain calm" will be displayed simultaneously.

[0583] An example of a prompt message might be: "Check the status of windows and doors from the current video data, and immediately output a warning if they are open. If the user is feeling stressed, generate a message to help them relax." This prompt ensures that appropriate information is provided at the right time.

[0584] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0585] Step 1:

[0586] The server receives video information in real time from the monitoring device. It receives raw video data of the monitoring area as input and converts the data into an analyzable format. This process involves pre-processing such as image resolution adjustment and noise reduction to format the data so that it can be easily analyzed by deep learning algorithms. The output is clear video data with pre-processing completed.

[0587] Step 2:

[0588] The server inputs pre-processed video data into an AI image analysis module. The input is formatted video data. The server then analyzes the video data using deep learning models such as TensorFlow or PyTorch to determine the location of objects and identify hazardous factors. A convolutional neural network is used for data processing, and the output consists of object feature data and a list of hazardous factors.

[0589] Step 3:

[0590] The server performs emotion recognition using the user's video and audio data. It uses the user's real-time video and audio data as input. Here, OpenCV and the Microsoft Azure Emotion API are used to infer emotions from facial and voice data, and the emotion engine determines the user's stress and anxiety. The output is the user's emotional state data.

[0591] Step 4:

[0592] The server generates notifications considering the analysis results and emotional state. Inputs include object feature data, a list of risk factors, and emotional state data. Based on the analysis results, a notification is generated suggesting a specific action. Depending on the emotional state, additional messages to reassure the user may also be included. The output is the generated notification data.

[0593] Step 5:

[0594] The server sends the generated notification to the user's visual device. The input is the generated notification data. Here, a push notification protocol is used so that the message appears immediately in the user's field of view. The output is the warning and follow-up message displayed on the user's visual device. This step allows the user to quickly understand the situation and take appropriate action.

[0595] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0596] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0597] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0598] [Fourth Embodiment]

[0599] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0600] As shown in Figure 7, the 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.

[0601] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0602] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0603] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0604] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0605] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0606] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0607] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0608] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0609] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0610] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0611] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0612] This invention relates to a system that uses AI technology to analyze video data acquired using a monitoring device, detects specific dangerous conditions, and notifies the user. The program processing of this system will be explained below using natural language.

[0613] First, the server continuously receives video data transmitted from the monitoring device. This allows it to always have access to the latest information about the monitored area. Next, the received video data is converted to an appropriate resolution and format and pre-processed to a format suitable for AI image analysis.

[0614] Next, the server uses an AI image analysis program to analyze the video data. This analysis identifies the location and state of specific objects and detects potential hazards such as heavy objects in high places or obstacles blocking pathways. Based on this information, the system assesses the risks in emergencies such as earthquakes.

[0615] The server then generates a notification message containing specific improvement suggestions based on the analysis results. For example, if objects are stacked in a particular location, it might create a notification such as, "We recommend moving the heavy object in this location." These notifications are sent to the user's electronic device via the terminal.

[0616] Users can check the situation based on the received notifications and adjust the placement of furniture and objects as needed. This action can mitigate potential hazards and improve the safety of living spaces and offices.

[0617] As a concrete example, when a monitoring system was used in an office, the server detected a large filing box stacked on a metal shelf. Based on this information, AI analysis determined that it posed a risk of falling during an earthquake and generated a notification saying, "Please move the filing box to a location below the shelf." Upon receiving this notification, the user can immediately move the filing box to a safe location, thereby mitigating the risk. In this way, the present invention aims to raise users' safety awareness and prevent accidents.

[0618] The following describes the processing flow.

[0619] Step 1:

[0620] The server receives video data from the monitoring device in real time. The received data is recorded as a secure stream and used for subsequent processing.

[0621] Step 2:

[0622] The server preprocesses the received video data into a format that can be analyzed. This process includes adjusting the resolution and performing noise filtering.

[0623] Step 3:

[0624] The server inputs the pre-processed video data into an AI image analysis module. The artificial intelligence model used here performs object recognition and identifies the placement of furniture and other objects.

[0625] Step 4:

[0626] Based on the results of the AI ​​analysis, the server evaluates the location and condition of identified objects and determines potential hazards. For example, it assesses the presence of heavy objects at high altitudes and the degree of obstruction in passageways.

[0627] Step 5:

[0628] Based on the analysis results and evaluation, the server generates notifications to send to the user. These notifications are prioritized and presented in a message format that specifically outlines improvement suggestions.

[0629] Step 6:

[0630] The device receives notifications from the server and displays them to the user. Notifications are delivered to the user's device as push notifications or in-app messages.

[0631] Step 7:

[0632] The user reviews the notification and makes improvements to the furniture and object placement as needed. After improvements are made, the system is configured to rerun the process for re-evaluation.

[0633] (Example 1)

[0634] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0635] Conventional video surveillance systems lack the ability to automatically identify and notify of dangerous situations, posing a challenge in preventing risks and ensuring safety in emergency situations where rapid response is required. This results in users being unable to identify potential dangers in a timely manner and take necessary countermeasures.

[0636] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0637] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for converting the video information into an analyzable format, and means for identifying the location and hazardous state of an object using machine intelligence image analysis. This makes it possible to identify hazardous states in real time, immediately notify the user, quickly assess the risk, and take countermeasures.

[0638] A "surveillance device" is a device that visually observes a specific area and continuously acquires video information.

[0639] "Visual information" refers to visual data acquired by surveillance equipment, which is image or video data used for analysis.

[0640] "Machine intelligence image analysis" is a technology that uses deep learning and other advanced algorithms to identify objects in video information and determine their location and state.

[0641] A "hazardous situation" is a situation in which the location or state of an object identified in the video information could potentially pose a risk.

[0642] "Means of risk assessment" refers to the process of determining the degree of risk that an identified hazardous situation may pose and evaluating whether appropriate countermeasures are necessary.

[0643] A "means for generating and communicating attention" is a system that creates messages indicating warnings and countermeasures for users based on the results of a risk assessment, and then electronically sends these messages to the user.

[0644] This invention relates to a system that analyzes video information acquired from a monitoring device and notifies the user of identified dangerous conditions. This system consists of a server, a terminal, and a user, each playing a specific role.

[0645] The server continuously receives video information from monitoring devices using network communication software. The received video information is converted into an AI-analyzable format using an image processing library. This conversion includes resizing and changing the format of the images.

[0646] The converted video information is analyzed on a server using machine intelligence image analysis tools. Deep learning frameworks such as TensorFlow and PyTorch are used to identify the location of objects and their hazardous conditions. For example, this can identify situations where an object is blocking a passageway or where a heavy object is placed at a high position.

[0647] Based on the analysis results, a risk assessment is performed, and specific notifications are generated to improve the hazardous situation. These notifications are sent to the user's electronic device via a terminal. The notification content includes specific instructions such as, "The passage is blocked by an object. Please move the object for safety."

[0648] Users receive notifications from their devices and check the situation on-site based on their content. By taking the instructed actions as needed, they can mitigate potential risks.

[0649] As a concrete example, consider its use in an office. If the server analyzes video information from a monitoring device and identifies a stack of document boxes on a metal shelf, it determines that there is a risk of them falling during an earthquake. A notification is generated and sent to the user saying, "Please move the document boxes under the shelf." This allows the user to move the document boxes to a safe location and mitigate the risk.

[0650] An example of a prompt message would be, "Please tell me how to build a notification system that detects hazardous materials from office security camera footage and assesses the risk."

[0651] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0652] Step 1:

[0653] The server receives video information from the monitoring device. The received video information is provided raw and in real time as a stream. This video information serves as input. Next, an image processing library is used to convert the video information into a format that can be analyzed. Specifically, the image resolution is resized to a size that the AI ​​model can process, and the format is converted to a standard format. As output of this process, image data in a format suitable for AI analysis is generated.

[0654] Step 2:

[0655] The server inputs pre-processed image data into an AI image analysis tool. It performs object recognition using deep learning models such as YOLO and ResNet. Calculations are performed on the input image data to detect the object's location and type. The object detection results are output, including information such as the object's type, location coordinates, and confidence score.

[0656] Step 3:

[0657] The server performs a risk assessment based on the detected object information. The specific input is information about the object's location and type obtained in the previous step. The risk assessment is performed using an expert system, taking into account the object's placement, human movement patterns, and potential hazards in emergencies. The assessment output includes potential hazard areas and priority for countermeasures.

[0658] Step 4:

[0659] The server generates notification messages for the user based on the risk assessment results. The assessment results and recommended actions for the user are used as input. Prompts from the generating AI model are utilized to create notifications that include specific improvement suggestions. Output notifications might include statements such as, "We recommend moving the heavy object located here."

[0660] Step 5:

[0661] The terminal receives notifications sent from the server and displays them to the user. The input is the notification message sent from the server, which is output to the user's electronic device as a push notification or email. Based on the notification content, the user checks the situation on site and takes the instructed actions as needed.

[0662] Step 6:

[0663] Based on the notification, users will identify the areas where risks have been identified. Specific actions might include moving heavy objects from shelves to a safer location. This can reduce potential safety risks and improve the safety of living and working environments.

[0664] (Application Example 1)

[0665] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0666] Industrial facilities present potential hazards arising from complex machine layouts and work environments, necessitating accident prevention measures. However, currently, there is no established method for detecting these hazards in real time and responding immediately, making efficient safety measures difficult. Furthermore, the lack of not only detection of hazards but also instructions for specific corrective measures hinders the rapid implementation of safety measures.

[0667] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0668] In this invention, the server includes means for receiving image information acquired from a monitoring device, means for converting the image information into an analyzable format, and means for identifying the location of an object and its hazardous status using machine learning image analysis. This makes it possible to detect potential hazardous conditions within industrial facilities in real time and to quickly notify operators of specific corrective measures.

[0669] A "monitoring device" is a device that captures the surrounding environment and continuously acquires image information.

[0670] "Image information" refers to data that represents the visual information of the surroundings, acquired by a monitoring device.

[0671] A "server" is a computer device that receives and processes information from multiple devices.

[0672] A "parsable format" refers to the data structure and format necessary for processing using machine learning techniques.

[0673] "Machine learning image analysis" is an analytical method that uses advanced technologies such as neural network algorithms to recognize objects in an image and evaluate the situation.

[0674] "Object position" refers to information that indicates the spatial arrangement of a specific object within the observed area.

[0675] A "hazardous situation" refers to any condition or arrangement in an industrial facility that could potentially threaten safety.

[0676] An "operator" is a person responsible for managing and controlling machinery and equipment and maintaining the safety of the facility.

[0677] A "notification" is an informational message generated based on the detected situation and sent to the operator.

[0678] An "information terminal" is a device used by operators to view received notifications.

[0679] The system program for implementing this invention consists of a monitoring device, a server, machine learning software, and a user information terminal.

[0680] The server continuously receives image information transmitted from monitoring devices. This image information shows the current arrangement and status of machinery and objects within the facility. The server first converts this information into a format that can be used for machine learning image analysis. This involves image format conversion and resolution adjustment.

[0681] Next, the server performs machine learning image analysis. Specifically, a neural network algorithm is used to recognize objects in the image and determine their location. After that, it identifies dangerous situations and identifies areas for improvement as needed. Frameworks such as TensorFlow are used for this analysis.

[0682] Next, the server generates a notification containing specific instructions to encourage corrective action based on the identified risk situation. This notification is sent as a push message to the operator's information terminal. The notification is communicated using services such as Firebase Cloud Messaging.

[0683] As a concrete example, consider a situation in a factory where a robotic arm is blocking a passageway and obstructing an emergency evacuation route. A monitoring device detects this situation, and a server performs real-time analysis. The analysis results are notified in the form of "We recommend adjusting the position of the robotic arm" and sent to the operator's information terminal. This enables a rapid response.

[0684] An example of a prompt to input into the generating AI model would be: "Explain the notification generation process for a system that uses AI to analyze video data inside a factory, detect dangerous situations, and generate notifications."

[0685] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0686] Step 1:

[0687] The server receives image information transmitted from the monitoring device. The input is real-time image data of the factory captured by the monitoring device. At this stage, the image data is retained in its original format. The output is the unprocessed image data stored in temporary storage for data retention.

[0688] Step 2:

[0689] The server converts received image information into a format suitable for analysis. The input is raw image data. Data processing is performed to change the image format to an appropriate format such as JPEG or PNG, and to reduce the resolution to a size suitable for machine learning models. The output is image data converted into a format suitable for analysis.

[0690] Step 3:

[0691] The server performs machine learning image analysis. The input is the image data transformed in the previous step. Here, the server applies a neural network model using a framework such as TensorFlow to recognize objects and identify hazardous situations within the image. The output is data on the location of objects and hazardous situations.

[0692] Step 4:

[0693] The server evaluates what needs to be improved based on the hazard situation and generates specific improvement instructions. The input is the location information of the identified object and hazard situation data. The data is analyzed to create an assessment of specific hazards, such as "the passage is blocked," and response instructions, such as "adjust the position of the robot arm." The output is notification message data, including the improvement instructions.

[0694] Step 5:

[0695] The server sends the generated notification to the operator's information terminal. The input is notification message data, including improvement instructions. Push notification services such as Firebase Cloud Messaging are used to send notifications to the operator's terminal in real time. The output is the notification message displayed on the operator's information terminal screen.

[0696] Step 6:

[0697] The operator checks the information notified on the terminal and makes adjustments to the machine layout, etc., according to the instructions. The input is the notification message displayed on the terminal. Based on the notification content, the operator checks the site and takes specific actions such as moving objects or adjusting the position of machinery. The output is the new layout within the facility after safety has been improved.

[0698] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0699] This invention provides a system that identifies hazardous elements in the environment and provides appropriate notifications to the user by combining video data acquisition from a monitoring device with AI image analysis. Furthermore, it incorporates an emotion engine that recognizes the user's emotional state and individually optimizes user interaction. The processing of this system's program is described in detail below in natural language.

[0700] First, the server receives video data acquired from the monitoring device and preprocesses it into a format suitable for analysis. The preprocessed data is then input into the AI ​​image analysis module to identify the placement of objects and potential hazards in the environment. In this identification process, deep learning algorithms are used to achieve highly accurate object recognition.

[0701] Next, the server evaluates the identified risk conditions based on the analysis results and creates improvement plans. In this process, it uses an emotion engine to simultaneously recognize the user's emotional state from their facial expressions and voice. If the user is feeling anxious or stressed, the server adjusts the content and tone of notifications accordingly and adds information to reassure the user.

[0702] Next, the server formats the generated notification in the most suitable format for the user and sends it to the user's digital device via the terminal. The notification is delivered using the most effective method, such as push notification or email. The content of this notification also takes into account the results of the emotion engine's analysis, enabling interaction that takes into account the user's emotional state.

[0703] As a concrete example, consider the use of the system in an office. If an employee is sitting at their desk with an anxious expression, the server will recognize this emotion and, in addition to the usual safety notification, send a notification that includes emotionally supportive content, such as, "For your safety, we will provide you with the most relaxed environment possible. Please let us know if you need any support." In this way, the system can provide the user with the most appropriate support.

[0704] The following describes the processing flow.

[0705] Step 1:

[0706] The server receives video data in real time from the monitoring devices. This allows it to collect basic data to understand the latest situation in the monitored area.

[0707] Step 2:

[0708] The server preprocesses the received video data, performing noise reduction and resolution adjustments, and converts it into a format suitable for AI image analysis.

[0709] Step 3:

[0710] The server inputs the pre-processed video data into an AI image analysis module. This AI module utilizes deep learning to perform object recognition and location determination with high accuracy.

[0711] Step 4:

[0712] The server assesses potential hazards based on the placement of objects identified from the analysis results. This assessment includes the risk of objects falling during an earthquake and the possibility of difficulty in evacuation due to obstacles in pathways.

[0713] Step 5:

[0714] The server simultaneously uses an emotion engine to recognize the user's emotional state. It analyzes input data from cameras and audio sensors to identify emotions from the user's facial expressions and voice.

[0715] Step 6:

[0716] The server generates notification content based on the risk assessment results and the user's emotional state. This notification includes not only conventional safety measures but also messages and improvement suggestions that take the user's feelings into consideration.

[0717] Step 7:

[0718] The device receives notifications generated from the server and displays them as push notifications on the user's electronic device. These notifications are delivered at a timing and tone that corresponds to the user's emotional state.

[0719] Step 8:

[0720] Users review the displayed notifications and take appropriate action to ensure the safety of the facility as needed. At this stage, emotions are re-evaluated, allowing for more appropriate support to be provided.

[0721] (Example 2)

[0722] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0723] Modern, high-performance monitoring systems are required not only to identify hazardous elements in the environment, but also to provide individually optimized notifications tailored to the user's emotional state. However, conventional technologies have struggled to achieve such flexible notifications that take user emotions into account. Therefore, a new approach is needed to improve both safety and user experience.

[0724] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0725] In this invention, the server includes a configuration for receiving video information acquired from a monitoring device, a configuration for converting the video information into an analyzable format, a configuration for identifying the arrangement of objects and dangerous conditions using machine learning image analysis, a configuration for recognizing the user's emotional state using an emotion analysis engine, and a configuration for generating and sending notifications to the user based on the evaluation and emotional state. This makes it possible to provide safe and individually optimized notifications that take the user's emotions into consideration.

[0726] A "monitoring device" is a device used to continuously monitor the environment and acquire video information.

[0727] "Visual information" refers to visual data provided by surveillance equipment, including image and video data.

[0728] "Machine learning image analysis" is a method of analyzing video information using machine learning techniques to identify the arrangement of objects and specific patterns.

[0729] An "emotion analysis engine" is a software configuration that determines a user's emotional state based on their facial expressions and voice data.

[0730] A "notification" is an informational message sent to the user based on the analysis results, which may include warnings or suggestions.

[0731] "Push-out notifications" are a notification method that displays information from the system in real time on the user's electronic device.

[0732] "Electronic communication" refers to the method of transmitting digital data between electronic devices, such as using email or messaging apps.

[0733] "Object arrangement" refers to information about the position and state of objects within an environment.

[0734] A "hazardous condition" refers to a situation or condition that could potentially pose a danger to the user or the environment.

[0735] This invention is a monitoring system that identifies hazardous elements in the environment and provides users with individually optimized notifications that respond to their emotions.

[0736] The server first receives video information in real time from the monitoring device. This data is preprocessed into an appropriate format before being analyzed as raw data. Specifically, processes such as data format conversion and noise reduction are performed. This prepares the image analysis module for high-precision analysis.

[0737] Next, using the processed data, the server identifies object placement and hazardous conditions within the environment through machine learning image analysis. Examples of algorithms used include deep learning frameworks such as TensorFlow and PyTorch, with YOLO and Faster R-CNN being used for object identification. During this process, objects identified as hazardous elements are labeled and recorded in a database.

[0738] Furthermore, the server uses an emotion analysis engine to recognize the user's emotional state from their facial expressions and voice data. Speech recognition technology (for example, technology that transcribes speech into text) and facial expression analysis software are used.

[0739] Based on these analysis results, the server generates a notification for the user and sends it to the user's electronic device via the terminal. The notification includes specific actions and support information tailored to the user's situation and emotions. This allows the user to confidently follow the instructions. Notifications are typically delivered to the user as push notifications or email.

[0740] For example, when this system is used in a business environment such as an office, it sends a notification prompting immediate action if a hazardous element is detected. Furthermore, when employees are experiencing stress, it provides emotionally supportive recommendations such as "relax and maintain your health."

[0741] An example of a prompt message might be, "Please tell me how to identify potential hazards in the meeting room using video analysis and how to set up notifications based on employee emotion recognition."

[0742] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0743] Step 1:

[0744] The server receives video information from the monitoring device in real time. This input data is typically in video stream format (e.g., MP4 or AVI). The received data is stored in a temporary cache as a preliminary step before analysis. This ensures that the data is always available for subsequent processing.

[0745] Step 2:

[0746] The server preprocesses the received video information into a format that can be analyzed. The input is video stream data, and the output is image data converted to a still image format (JPEG or PNG). This process involves data format conversion and noise reduction. Specifically, a Gaussian filter is applied to reduce noise and clarify the image.

[0747] Step 3:

[0748] The server performs machine learning image analysis using pre-processed data. The input is still image data, and the output is location information (bounding boxes and labels) of identified objects. Using TensorFlow or PyTorch, models such as YOLO and Faster R-CNN are used to accurately identify the location and type of objects.

[0749] Step 4:

[0750] The server evaluates the hazardous conditions within the environment based on the analysis results. The input is information on the placement of identified objects, and the output is a hazard assessment report. This assessment uses pre-defined safety standards and rules to identify high-risk situations and conditions.

[0751] Step 5:

[0752] The server analyzes the user's emotional state using an emotion analysis engine. Input is the user's facial expressions and voice data, and output is the emotional state (e.g., anxiety, relaxation). Speech recognition and facial expression analysis software are used to recognize the user's emotions according to their situation.

[0753] Step 6:

[0754] The server generates notifications based on the risk level and the user's emotions. Inputs are a risk assessment report and emotional state, while output is a customized notification message. The notification includes specific actions and emotional support to encourage the user to respond confidently.

[0755] Step 7:

[0756] The device sends notifications generated from the server to the user's electronic device. The input is a customized notification message, and the output is a notification received on a smartphone or PC. Notifications are delivered in the form of push notifications or email and are presented to the user in real time.

[0757] (Application Example 2)

[0758] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0759] Conventional monitoring systems have a simplistic and general approach to identifying risk factors and notifying users, making it difficult to respond flexibly to different situations and user emotional states. Therefore, challenges include a lack of real-time response capabilities to risks within the monitoring area and a lack of individualized stress reduction measures for users.

[0760] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0761] In this invention, the server includes means for receiving video information acquired from a monitoring device, means for processing the video information into an analyzable format, means for identifying the location of objects and risk factors using artificial intelligence image analysis, means for recognizing the user's emotional state, means for performing evaluations and generating notifications based on the identified risk factors and emotional state, and means for transmitting notifications to the user's visual device. This enables real-time identification of risk factors and the provision of appropriate notifications according to the user's emotional state.

[0762] A "surveillance device" is a device used to acquire video information of a monitored area.

[0763] "Visual information" refers to visual data of the environment acquired from monitoring devices.

[0764] An "analyzable format" is a format that has been prepared in a way that is suitable for data analysis.

[0765] "Artificial intelligence image analysis" is a method that uses AI technology to identify the location of an object and potential hazards from video information.

[0766] "Object position" refers to the spatial location occupied by an object within the video information.

[0767] A "risk factor" is an element in the environment that could potentially lead to accidents or problems.

[0768] "Emotion recognition means" refers to a function that analyzes the user's facial expressions and voice from video information to infer and recognize their emotional state.

[0769] "Notifications" are warning or informational messages provided to users based on analysis and evaluation.

[0770] A "visual device" is a display device that a user can wear to directly receive visual information.

[0771] In the system that realizes this invention, a server plays a central role. The server directly receives video information acquired from the monitoring device and first converts that information into an analyzable format. Specifically, it performs preprocessing to optimize the video data, such as adjusting the resolution and removing noise. The software used here could be libraries or tools that are widely used for data processing.

[0772] Next, the server performs artificial intelligence image analysis. Here, deep learning frameworks such as TensorFlow and PyTorch are used to accurately identify the location of objects and potential hazards within the image data. In this image analysis phase, pre-trained models are utilized, and it is also possible to retrain the models in real time as needed.

[0773] In addition, the server uses emotion recognition methods to analyze the user's emotional state. Here, facial recognition technology and voice analysis technology are utilized to determine emotions from the user's facial expressions and vocalizations. Specific tools that can be used include OpenCV and the Microsoft Azure Emotion API.

[0774] Based on the analysis results, the server generates a notification for the user and sends it to their visual device. This notification instantly appears on the user's wearable device, such as smart glasses. It functions as a push notification, prompting immediate action in response to potential hazards.

[0775] As a concrete example, consider a security officer patrolling a factory at night. If the officer senses danger and detects a suspicious person who has infiltrated the surveillance area, a warning message will immediately appear in the user's field of view. At this time, the user's emotional state will also be taken into consideration, and if necessary, a kind message such as "Please remain calm" will be displayed simultaneously.

[0776] An example of a prompt message might be: "Check the status of windows and doors from the current video data, and immediately output a warning if they are open. If the user is feeling stressed, generate a message to help them relax." This prompt ensures that appropriate information is provided at the right time.

[0777] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0778] Step 1:

[0779] The server receives video information in real time from the monitoring device. It receives raw video data of the monitoring area as input and converts the data into an analyzable format. This process involves pre-processing such as image resolution adjustment and noise reduction to format the data so that it can be easily analyzed by deep learning algorithms. The output is clear video data with pre-processing completed.

[0780] Step 2:

[0781] The server inputs pre-processed video data into an AI image analysis module. The input is formatted video data. The server then analyzes the video data using deep learning models such as TensorFlow or PyTorch to determine the location of objects and identify hazardous factors. A convolutional neural network is used for data processing, and the output consists of object feature data and a list of hazardous factors.

[0782] Step 3:

[0783] The server performs emotion recognition using the user's video and audio data. It uses the user's real-time video and audio data as input. Here, OpenCV and the Microsoft Azure Emotion API are used to infer emotions from facial and voice data, and the emotion engine determines the user's stress and anxiety. The output is the user's emotional state data.

[0784] Step 4:

[0785] The server generates notifications considering the analysis results and emotional state. Inputs include object feature data, a list of risk factors, and emotional state data. Based on the analysis results, a notification is generated suggesting a specific action. Depending on the emotional state, additional messages to reassure the user may also be included. The output is the generated notification data.

[0786] Step 5:

[0787] The server sends the generated notification to the user's visual device. The input is the generated notification data. Here, a push notification protocol is used so that the message appears immediately in the user's field of view. The output is the warning and follow-up message displayed on the user's visual device. This step allows the user to quickly understand the situation and take appropriate action.

[0788] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0789] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0790] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0791] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0792] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0793] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0794] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0795] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0796] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0797] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0798] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0799] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0800] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0801] 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.

[0802] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0803] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0804] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0805] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0806] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0807] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0808] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0809] The following is further disclosed regarding the embodiments described above.

[0810] (Claim 1)

[0811] A means for receiving video data acquired from a monitoring device,

[0812] Means for processing the aforementioned video data into an analyzable format,

[0813] A means of identifying the location and hazardous state of an object using artificial intelligence image analysis,

[0814] A means of evaluating areas for improvement based on identified hazardous conditions,

[0815] A means for generating and sending a notification to the user based on the aforementioned evaluation,

[0816] A system that includes this.

[0817] (Claim 2)

[0818] The artificial intelligence image analysis system according to claim 1 is characterized in that it recognizes objects using a deep learning algorithm.

[0819] (Claim 3)

[0820] The system according to claim 1, characterized in that the notification is sent to the user's electronic terminal as a push notification or email.

[0821] "Example 1"

[0822] (Claim 1)

[0823] A means for receiving video information acquired from a monitoring device,

[0824] means for converting the aforementioned video information into an analyzable format,

[0825] A means for identifying the position and hazardous state of an object using machine intelligence image analysis,

[0826] A means of assessing risk based on identified hazardous conditions,

[0827] A means for generating and communicating alerts to the user based on the aforementioned evaluation,

[0828] A device that includes this.

[0829] (Claim 2)

[0830] The apparatus according to claim 1, characterized in that the machine intelligence image analysis uses a deep learning method to identify objects.

[0831] (Claim 3)

[0832] The apparatus according to claim 1, characterized in that the aforementioned notice is transmitted to the user's electronic device as a push notification or electronic communication.

[0833] "Application Example 1"

[0834] (Claim 1)

[0835] Means for receiving image information acquired from a monitoring device,

[0836] Means for converting the aforementioned image information into an analyzable format,

[0837] A means for identifying the location and hazardous conditions of an object using machine learning image analysis,

[0838] A means of evaluating what needs to be improved based on identified hazardous situations,

[0839] A means for generating and distributing notifications to operators based on the aforementioned evaluation,

[0840] A means by which an operator who receives a notification can provide instructions for adjusting the layout of machinery within the factory,

[0841] A system that includes this.

[0842] (Claim 2)

[0843] The machine learning image analysis is characterized by performing object recognition using a neural network algorithm, as described in claim 1.

[0844] (Claim 3)

[0845] The system according to claim 1, characterized in that the notification is sent as a push message to the operator's information terminal.

[0846] "Example 2 of combining an emotion engine"

[0847] (Claim 1)

[0848] A configuration that receives video information acquired from a monitoring device,

[0849] A configuration for converting the aforementioned video information into an analyzable format,

[0850] A configuration that uses machine learning image analysis to identify the arrangement of objects and their hazardous state,

[0851] A configuration that evaluates areas for improvement based on identified hazardous conditions,

[0852] A configuration that recognizes the user's emotional state using an emotion analysis engine,

[0853] A configuration that generates and sends notifications to the user based on the aforementioned evaluation and emotional state,

[0854] A system that includes this.

[0855] (Claim 2)

[0856] The machine learning image analysis is characterized by using a phase detection learning algorithm to identify objects, as described in claim 1.

[0857] (Claim 3)

[0858] The system according to claim 1, characterized in that the notification is transmitted to the user's electronic device as a push notification or electronic communication.

[0859] "Application example 2 when combining with an emotional engine"

[0860] (Claim 1)

[0861] A means for receiving video information acquired from a monitoring device,

[0862] Means for processing the aforementioned video information into an analyzable format,

[0863] A means of identifying the location and hazardous factors of an object using artificial intelligence image analysis,

[0864] An emotion recognition means for recognizing the user's emotional state,

[0865] A means for conducting an assessment and generating a notification based on identified risk factors and emotional states,

[0866] Means for transmitting the aforementioned notification to the user's visual device,

[0867] A system that includes this.

[0868] (Claim 2)

[0869] The artificial intelligence image analysis system according to claim 1 is characterized in that it uses a deep learning algorithm to identify objects.

[0870] (Claim 3)

[0871] The system according to claim 1, characterized in that the notification is displayed in real time on the user's wearable visual device. [Explanation of symbols]

[0872] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for receiving video data acquired from a monitoring device, Means for processing the aforementioned video data into an analyzable format, A means of identifying the location and hazardous state of an object using artificial intelligence image analysis, A means of evaluating areas for improvement based on identified hazardous conditions, A means for generating and sending a notification to the user based on the aforementioned evaluation, A system that includes this.

2. The artificial intelligence image analysis system according to claim 1 is characterized in that it recognizes objects using a deep learning algorithm.

3. The system according to claim 1, characterized in that the notification is sent to the user's electronic terminal as a push notification or email.

Images