System

Abstract

To provide a system for accurately and objectively evaluating the fatigue degree of a driver, and for reducing the risk of an operation based on the evaluation.SOLUTION: The driver takes a picture of his / her face using the terminal, and the image is transmitted to the server. The generation AI on the server extracts feature amounts such as eye redness, eyelid motion, and facial tone, and evaluates the degree of tiredness on a scale of 0 to 100. When the fatigue degree exceeds the reference value, an alert is transmitted to the terminal and the operation manager to urge the stop of the operation and the plan change.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Driving errors and accidents caused by driver fatigue have become a problem. Accidents caused by poor health and chronic illnesses are increasing, particularly with the aging of the population, threatening the safety of driving. Conventional methods often rely on simple visual checks or self-reporting, making accurate assessment of fatigue difficult. Furthermore, these methods are prone to subjective feedback from the driver, making them unreliable. The present invention aims to provide a system for accurately and objectively assessing driver fatigue and reducing driving risks based on the assessment. [Means for solving the problem]

[0005] The present invention solves the aforementioned problems with a system that includes a means for capturing a driver's facial image, a means for transmitting the captured facial image data to a server, a means for using a generation AI on the server to evaluate the driver's fatigue level from the facial image data, and a means for issuing an alert based on the evaluation results. Specifically, the driver uses a device to capture a photo of their face, and the image is sent to the server. The generation AI on the server extracts features from the image data, such as eye redness, eyelid movement, and facial color, and evaluates the driver's fatigue level on a scale of 0 to 100. If the driver's fatigue level exceeds a threshold, an alert is sent to the device and the operation manager, prompting them to stop operation or change the plan. In addition, the captured facial image data and evaluation results are stored in a database, and the generation AI continues to learn based on the accumulated data. This enables more accurate fatigue assessments, contributing to improved operation safety.

[0006] "Driver" means a person engaged in the business of operating or driving a vehicle or means of transport.

[0007] "Facial image" refers to digital image data of the driver's face.

[0008] "Means for taking photographs" refers to means for obtaining facial images using a camera or a photographing device.

[0009] The "transmitting means" refers to a communication means for transmitting the captured face image data to the server.

[0010] "Server" means a computer system that receives, processes, and stores data over a network.

[0011] "Generative AI" refers to a model for performing a specific task using artificial intelligence technology, and in the present invention refers to a model for assessing fatigue level from facial image data.

[0012] "Fatigue level" is a numerical indicator that shows how much a driver's physical condition and attention have deteriorated.

[0013] The "means of evaluation" refers to a means for calculating fatigue levels from facial image data captured using a generation AI.

[0014] The "means for issuing an alert" refers to a means for issuing a warning when the fatigue level exceeds a reference value.

[0015] A "database" is a system that systematically stores and manages large amounts of data and allows them to be searched and retrieved as needed.

[0016] An "operations manager" is a person responsible for managing the work of drivers and operation plans.

[0017] "Learning" is the process by which generative AI improves its performance based on past data. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0019] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0022] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0023] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0024] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0031] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0038] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0039] The present invention relates to a system for assessing driver fatigue levels and ensuring operational safety. This system includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the driver's fatigue level using a generation AI on the server, and issuing an alert based on the results.

[0040] Server Roles and Operations

[0041] The server is responsible for the core processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Second, it saves the image data in a database. This saved data is used not only for subsequent fatigue assessments but also for training the generation AI.

[0042] The server then inputs the image data into the generation AI, which uses a pre-trained model to extract features from the image, such as bloodshot eyes, eyelid movement, and facial color. Based on the extracted features, the AI ​​evaluates the driver's fatigue level on a scale of 0 to 100. If the result exceeds the threshold, the server generates an alert and sends it to the device and the operation manager.

[0043] Terminal roles and processing

[0044] The terminal is a device used by the driver, and performs the following process. First, the driver uses the terminal to take a picture of his or her own face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. Next, the terminal sends the captured face image data to a server. This transmission is usually performed via a network.

[0045] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[0046] User Roles and Actions

[0047] Users mainly play two roles: drivers and operation managers. Drivers must take facial images using a terminal before and after work, which provides the necessary data to the server.

[0048] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0049] Specific example explanation

[0050] For example, driver A takes a picture of his face on his device before starting work. At this time, the device's camera activates and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0051] By using the above system and process, the present invention can efficiently and objectively evaluate the fatigue level of drivers and improve operational safety. This system contributes to reducing accidents and improving the working environment.

[0052] The processing flow will be explained below.

[0053] Specific processing steps of the program

[0054] Server Processing

[0055] Step 1:

[0056] The server receives the facial image data sent from the terminal. Specifically, the image data sent as an HTTP request is processed by the server's reception process and temporarily stored in memory.

[0057] Step 2:

[0058] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (e.g., transmission time, driver ID) in the database.

[0059] Step 3:

[0060] The server inputs the saved image data into the generative AI. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI model.

[0061] Step 4:

[0062] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[0063] Step 5:

[0064] The server evaluates the fatigue level and checks the results. If the evaluation result exceeds the reference value, an alert is generated.

[0065] Step 6:

[0066] The server sends the generated alert information to the terminal and the operation manager. Specifically, it executes a process to send an alert notification to the terminal and the manager's device.

[0067] Terminal handling

[0068] Step 1:

[0069] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[0070] Step 2:

[0071] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[0072] Step 3:

[0073] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[0074] Step 4:

[0075] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[0076] Step 5:

[0077] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[0078] User Action

[0079] Step 1:

[0080] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[0081] Step 2:

[0082] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[0083] Step 3:

[0084] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[0085] Step 4:

[0086] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[0087] Example 1

[0088] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0089] Driver fatigue is one of the major causes of traffic accidents, and its prevention and management are important. Conventional methods often rely on subjective self-reporting or simple checklists, making it difficult to provide an immediate, objective assessment. This creates a risk that highly fatigued drivers may continue driving. This invention aims to provide a system that uses facial images of drivers to quickly and accurately assess their fatigue level and improve driving safety.

[0090] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0091] In this invention, the server includes means for receiving facial image data and storing it in a database, means for inputting the facial image data into a generative AI model and extracting features such as eye congestion, eyelid movement, and facial color to evaluate fatigue level, and means for generating an alert if the fatigue level exceeds a reference value based on the evaluation result and sending the alert to the terminals of the driver and manager. This enables quick and accurate fatigue level evaluation based on the facial image of the driver, thereby improving driving safety.

[0092] "Driver" means a person who drives a vehicle professionally.

[0093] "Facial Image" refers to a photograph or digital image of a driving personnel's face.

[0094] "Terminal" refers to an electronic device used by a driver that has camera and communication capabilities.

[0095] "Network" refers to a communications infrastructure for transmitting data, including the Internet and mobile communications.

[0096] "Server" refers to a computer system for receiving, processing, and storing data over a network.

[0097] A "database" refers to a system for efficiently storing, retrieving, and managing data.

[0098] A "generative AI model" refers to a mathematical model that is trained to perform a specific task using artificial intelligence techniques.

[0099] "Features" are attributes and information extracted from facial images, including eye congestion, eyelid movement, and facial color.

[0100] "Fatigue level" refers to the degree of fatigue of the driver expressed numerically.

[0101] "Alert" refers to a warning or caution message sent to an operator or manager.

[0102] "Manager" refers to the person responsible for supervising and managing the health of drivers and the safety of operations.

[0103] This invention relates to a system for assessing the fatigue level of drivers and ensuring driving safety. Specifically, it includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the fatigue level using a generation AI on the server, and issuing an alert based on the results.

[0104] System Configuration

[0105] Device Role

[0106] The terminal is an electronic device used by the driver and has the following functions:

[0107] Camera function: Taking a facial image of the driver, for example, using the camera on a smartphone or tablet.

[0108] Communication function: Sends captured facial image data to a server via a network (Internet or mobile communications).

[0109] Display function: Displays alerts received from the server to the driver.

[0110] Server Roles

[0111] The server is responsible for the core processing and has the following functions:

[0112] Data reception and storage function: The server receives the facial image data sent from the device, temporarily stores it in memory, and then saves it in a database. The database used could be, for example, MySQL or PostgreSQL.

[0113] Generative AI model: The server inputs facial image data into a generative AI model (e.g., a model using TensorFlow or PyTorch) to extract features such as eye redness, eyelid movement, and facial color. Based on these features, the server then evaluates the level of fatigue on a scale of 0 to 100.

[0114] Alert generation and notification: If the evaluation result exceeds the reference value, the server generates alert information and sends it to the terminals of the driver and manager.

[0115] User Roles

[0116] Drivers use a device to take a picture of their face and send it to the server before and after work, providing the server with the necessary data.

[0117] Administrators monitor the alert information sent from the server, and when an alert occurs, they recheck the health status of the drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[0118] Specific examples

[0119] For example, driver A takes a facial image on his device before work. The device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into a generative AI model to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the manager's device, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0120] Prompt Sentence Examples

[0121] Below is an example of a prompt for configuring the generative AI model used in this system.

[0122] We would like to develop an AI system that uses facial images of drivers to assess their fatigue level. We would like you to create an AI model that extracts eye congestion, eyelid movement, and facial color from the images and evaluates fatigue level on a scale of 0 to 100.

[0123] This system and process will enable efficient and objective evaluation of driver fatigue levels and improve operational safety, contributing to reducing accidents and improving the working environment.

[0124] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0125] Step 1: Take an image (device processing)

[0126] Specific explanation: The driver activates the device's camera to take a facial image. The device uses the camera function to capture the image and temporarily stores it in its internal memory. At this time, the device checks that the image resolution, brightness, etc. are appropriate.

[0127] Input and output: The input is the driver's face, and the output is the captured facial image data.

[0128] Specific operation: The driver launches the device's camera app, positions their face on the screen, and presses the shutter button.

[0129] Step 2: Data transmission (terminal processing)

[0130] Specifically, the device sends the captured facial image data to the server via the network. This transmission is performed as encrypted data to ensure the security of the communication.

[0131] Input and output: The input is the captured facial image data, and the output is the data sent to the server.

[0132] Specific operation: After taking a photo, the device automatically encrypts the image data and sends a POST request to the server via the Internet.

[0133] Step 3: Data reception and storage (server processing)

[0134] Specific explanation: The server receives the facial image data sent from the device and temporarily stores it in memory. This image data is then saved in a database. The saved data is used for fatigue evaluation and as learning data for the generation AI.

[0135] Input and output: The input is the facial image data sent from the terminal, and the output is the data stored in the database.

[0136] Specific operation: The server decodes the received image data, stores it in temporary memory, and then stores it in the "face image" table of the database.

[0137] Step 4: Feature extraction and fatigue evaluation (server processing)

[0138] Specifically, the server inputs facial image data obtained from a database into a generative AI model. The generative AI model extracts features from the image, such as bloodshot eyes, eyelid movement, and facial color. Then, based on the extracted features, the AI ​​evaluates the fatigue level on a scale of 0 to 100.

[0139] Input and output: The input is face image data obtained from the database, and the output is the fatigue score.

[0140] Specific operation: The server inputs image data into the TensorFlow model, and evaluates and integrates the eye bloodshot score, eyelid open / close score, and facial color score.

[0141] Step 5: Alert generation and notification (server processing)

[0142] Specific explanation: If the evaluation result exceeds the reference value, the server generates an alert information, which is sent to the driver's terminal and the manager's terminal.

[0143] Inputs and Outputs: The input is the fatigue score, and the output is the alert sent to the driver and manager.

[0144] Specific operation: When the evaluation score exceeds a threshold value (e.g., 80), the server generates an alert message and sends a push notification to the devices of the driver and manager.

[0145] Step 6: Receiving and displaying alerts (terminal processing)

[0146] Specific explanation: The device will receive the alert message as soon as it is sent and display an alert to the driver, which will include specific instructions such as "take a break."

[0147] Input and Output: The input is the alert message sent from the server, and the output is the alert displayed to the driver.

[0148] Specific operation: The driver's device receives a push notification and displays the message, "You are highly fatigued. Please take a break."

[0149] Step 7: Check the alert and respond (user action)

[0150] Specific explanation: Users (drivers and managers) ensure that drivers check the alerts on their devices and take breaks if necessary. Managers recheck the health status of drivers, and if necessary, change the operation plan and arrange for a substitute driver.

[0151] Input and output: Input is an alert message sent from a terminal or server, and output is a response such as a change in the operation plan or a driver's break.

[0152] Specific behavior:

[0153] The driver checks the alert, stops the vehicle in a safe place, and takes a break.

[0154] The administrator checks the driver's status on the system's management screen and, if necessary, contacts other drivers to arrange a replacement.

[0155] (Application example 1)

[0156] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0157] Conventional driver fatigue assessment systems have difficulty assessing driver fatigue in real time, making it difficult to issue alerts in a timely manner. Furthermore, fatigue assessments cannot be performed while driving, which in turn makes it difficult to ensure safety. Therefore, there is a need for a system that can assess driver fatigue in real time while driving and immediately notify drivers of alerts.

[0158] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0159] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for periodically capturing a facial image of the driver using smart glasses and transmitting the captured facial image data to the server in real time, and means for displaying an audio or visual alert on the smart glasses based on the evaluation result. This makes it possible to evaluate the fatigue level in real time even while driving and to immediately notify the driver of the alert.

[0160] "Operator" means an individual responsible for operating or driving a vehicle.

[0161] "Facial image" refers to video data showing the driver's face.

[0162] "Server" refers to the computer system that receives and stores facial image data and analyzes the data using a generative AI model.

[0163] "Fatigue level" is an index that indicates the degree of physical and mental fatigue of drivers.

[0164] "Generative AI" is an artificial intelligence model that learns from large amounts of data and uses the results to analyze and predict new data.

[0165] An "alert" is a warning message that is issued to call attention when a set reference value is exceeded.

[0166] "Smart glasses" are eyeglass-type devices that display digital information and connect to the Internet and applications.

[0167] "Real-time" is a term that refers to processing or transmission of information immediately, without delay.

[0168] "Voice alert" refers to a function that issues warnings or notifications by voice.

[0169] "Visual alert" refers to a function that provides warnings or notifications through visual means such as a display or LED.

[0170] The present invention relates to a system that acquires facial images of drivers through smart glasses, transmits the images to a server in real time, evaluates their fatigue level using a generative AI model, and issues audio and visual alerts.

[0171] System configuration

[0172] Server Roles and Operations

[0173] The server receives, stores, and generates AI to evaluate the level of fatigue and generates an alert if the level exceeds the threshold. The main hardware and software used are as follows:

[0174] Hardware: High-performance server machine

[0175] Software: Python, generative AI models, database systems

[0176] The server first receives the facial image data sent from the smart glasses and temporarily stores it in a database. It then inputs the image data into a generative AI model to extract features and evaluate the level of fatigue based on information such as bloodshot eyes, eyelid movement, and facial color. If the evaluation result exceeds a threshold, an alert is generated and sent to the driver's smart glasses and the operations manager.

[0177] The role and processing of smart glasses

[0178] The smart glasses used by drivers will:

[0179] Photographing: Take regular photographs of your face using a camera

[0180] Send: Send the captured facial image data to the server in real time

[0181] Alert Indication: Audio and visual indication of alerts sent from the server

[0182] The main hardware and software used are:

[0183] Hardware: Smart glasses (camera, microphone, display)

[0184] Software: Photo and transmission application, alert display application

[0185] User Roles and Actions

[0186] Users have two main roles: Driver and Operations Manager:

[0187] Driver: Wears smart glasses, takes a facial image and sends it to the server. This process is automatic while driving.

[0188] Operation manager: Monitors alert information sent from the server and makes changes to operation plans or arranges for substitute drivers as necessary.

[0189] Specific examples

[0190] For example, driver A puts on his smart glasses and begins driving. The smart glasses periodically take images of A's face and send the data to a server in real time. The server uses a generative AI model to evaluate A's fatigue level, and if the evaluation result exceeds the standard value, it issues a voice alert to A's smart glasses saying, "Your fatigue level is high. Please take a break." A similar alert is also sent to the operations manager, who checks the situation and takes action as necessary.

[0191] Examples of prompt statements

[0192] "Rate the fatigue level of the person in this image on a scale of 0-100."

[0193] The above is the details of each part in the embodiment of the present invention.

[0194] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0195] Step 1:

[0196] The driver wears smart glasses, and the camera in the smart glasses periodically captures facial images.

[0197] Input: Camera image

[0198] Data processing: Smart glasses camera captures driver's face

[0199] Output: Face image data

[0200] Step 2:

[0201] The captured facial image data is temporarily stored in the internal memory of the smart glasses.

[0202] Input: Facial image data

[0203] Data processing: Temporary storage of data

[0204] Output: Facial image data stored in internal memory

[0205] Step 3:

[0206] Facial image data is transmitted to a server in real time.

[0207] Input: Facial image data stored in the internal memory

[0208] Data processing: Sending image data

[0209] Output: Facial image data arriving at the server

[0210] Step 4:

[0211] The server receives the facial image data and temporarily stores it in memory.

[0212] Input: Facial image data that arrives at the server

[0213] Data processing: storing in memory

[0214] Output: Temporarily stored face image data

[0215] Step 5:

[0216] The server stores the facial image data in a database and inputs the stored data into a generative AI model.

[0217] Input: Temporarily stored facial image data

[0218] Data processing: storing in a database and inputting into a generative AI model

[0219] Output: Facial image data input to the generative AI model

[0220] Step 6:

[0221] The generative AI model extracts features from facial image data and assesses fatigue levels based on information such as eye congestion, eyelid movement, and facial color.

[0222] Input: Facial image data

[0223] Data calculation: feature extraction and fatigue evaluation

[0224] Output: Fatigue evaluation results

[0225] Step 7:

[0226] If the evaluation result exceeds the reference value, the server generates alert information and sends the alert to the driver's smart glasses and the operation manager.

[0227] Input: Fatigue evaluation result

[0228] Data Calculation: Alert Generation

[0229] Output: Alert information

[0230] Step 8:

[0231] The driver's smart glasses receive the alert information and display audio and visual alerts.

[0232] Input: Alert information

[0233] Data processing: Alert display

[0234] Output: Audio and visual alerts

[0235] Specific examples

[0236] For example, driver A puts on his smart glasses and steps 1 to 8 are carried out. An example of a prompt sentence at this time is "Please rate the fatigue level of the person in this image on a scale of 0 to 100." This series of steps makes it possible to assess fatigue levels in real time while driving, ensuring safety.

[0237] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0238] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure driving safety. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generation AI on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[0239] Server Roles and Operations

[0240] The server is responsible for the core processing. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion evaluations, but also for training the generative AI.

[0241] The server then inputs the image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the image, such as eye redness, eyelid movement, and facial color, and evaluates fatigue levels on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear.

[0242] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the device and the operation manager.

[0243] Terminal roles and processing

[0244] The terminal is a device used by the driver, and performs the following process: First, the driver uses the terminal to take a picture of his or her face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. The terminal then transmits the captured face image data to a server. Transmission is usually performed via a network.

[0245] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[0246] User Roles and Actions

[0247] Users mainly play two roles: drivers and operation managers. Drivers always use a device to take a facial image before and after work, which provides the necessary data to the server.

[0248] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0249] Specific example explanation

[0250] For example, driver A takes a facial image on his device before starting work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion. If the evaluation reveals that fatigue level exceeds the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0251] With the above system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[0252] The processing flow will be explained below.

[0253] Server Processing

[0254] Step 1:

[0255] The server receives the facial image data sent from the terminal. Specifically, it receives the image data sent as an HTTP request and temporarily stores it in memory.

[0256] Step 2:

[0257] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (such as transmission time and driver ID) in the database.

[0258] Step 3:

[0259] The server inputs the saved image data into the generative AI and emotion engine. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI and emotion engine.

[0260] Step 4:

[0261] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[0262] Step 5:

[0263] The server uses an emotion engine to analyze emotions from facial image data, specifically identifying emotions such as joy, anger, sadness, surprise, and fear, and converting them into numerical data.

[0264] Step 6:

[0265] The server then integrates the extracted fatigue level and emotion data to make a comprehensive assessment. For example, if fatigue is high and strong stress or anger is also detected, the server will assess the risk as high.

[0266] Step 7:

[0267] If the evaluation result exceeds the reference value, the server generates an alert message that includes the driver ID and the evaluation result.

[0268] Step 8:

[0269] The server sends the generated alert information to the terminal and the operation manager. Specifically, it sends the alert notification as an HTTP request to the terminal and the manager's device.

[0270] Terminal handling

[0271] Step 1:

[0272] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[0273] Step 2:

[0274] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[0275] Step 3:

[0276] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[0277] Step 4:

[0278] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[0279] Step 5:

[0280] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[0281] User Action

[0282] Step 1:

[0283] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[0284] Step 2:

[0285] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[0286] Step 3:

[0287] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[0288] Step 4:

[0289] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[0290] Example 2

[0291] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0292] Conventional driver health management systems mainly rely on self-reporting or some biometric sensors, making it difficult to accurately assess fatigue levels and emotions. Furthermore, it is difficult to issue warnings at the appropriate time, making it difficult to ensure operational safety. To address this issue, the present invention aims to provide a system that uses facial images of drivers to assess their fatigue levels and emotions and take appropriate countermeasures accurately and quickly.

[0293] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0294] In this invention, the server includes a means for capturing a facial image of the driver, a means for transmitting the captured facial image data to a remote computing device, a means for using artificial intelligence on the computing device to evaluate the fatigue level from the facial image data, and a means for issuing an alarm based on the evaluation result, thereby enabling accurate evaluation of the fatigue level and emotions of the driver based on the facial image and improving driving safety.

[0295] "Operator" means a person responsible for operating a vehicle.

[0296] The "face image" is image data of the driver's face.

[0297] "Capturing means" refers to a camera or other recording device used to capture a facial image.

[0298] The "transmitting means" refers to a communication device or software for transferring captured facial image data to a remote computing device.

[0299] A "server" is a type of computing device that is a data processing device that receives, stores, and evaluates facial image data.

[0300] "Computing device" refers to a device that contains hardware and software for processing data.

[0301] "Fatigue level" is an indicator of the driver's fatigue state and is evaluated using a generative AI model.

[0302] "Generative AI" refers to an artificial intelligence model that has been pre-trained to assess fatigue levels from facial image data.

[0303] "Evaluation results" are the analysis results of fatigue levels and emotions obtained by the generative AI and emotion engine.

[0304] An "alert" is a warning or notification issued to a driver or operations manager based on the evaluation results.

[0305] "Means for issuing an alert" refers to a method or device for generating an alert based on the evaluation results and notifying the driver or operations manager.

[0306] The "storage device" refers to a data storage device for storing photographed face image data and evaluation results.

[0307] "Means for assessing emotions" refers to software or systems for identifying emotions of drivers from facial image data.

[0308] "Comprehensive assessment" is a process of integrating the results of fatigue and emotional assessment to comprehensively judge the driver's condition.

[0309] "Terminal" refers to an electronic device used by a driver, and is equipped with a camera and communication functions.

[0310] "Operations manager" means a person responsible for managing the driving staff and the operational status of the vehicle.

[0311] "Means of notification" refers to the method or device for communicating an alert to an operator or dispatcher.

[0312] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure safe driving. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generative AI model on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[0313] The server is the core device responsible for the main processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. Database and storage systems such as MySQL and Amazon S3 are used for this purpose. The image data is also used for subsequent fatigue and emotion assessments, as well as for training the generative AI model.

[0314] The server then inputs the received image data into the generation AI and emotion engine. The generation AI uses pre-trained PyTorch and TensorFlow to extract fatigue characteristics from the image (such as bloodshot eyes, eyelid movement, and facial color). At the same time, the emotion engine uses OpenFace and the Microsoft Emotion API to analyze facial expressions from the facial image and identify emotions such as joy, anger, sadness, surprise, and fear. The evaluation results are displayed on a scale of 0 to 100.

[0315] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert using services such as "Twilio" and sends it to the device and the operation manager. This enables a prompt response according to the driver's condition.

[0316] The terminal is a device that is directly used by the driver. First, the terminal uses its camera function to capture a facial image of the driver and temporarily stores it in its internal memory. Then, it sends this captured facial image data to a server via the "HTTP" or "HTTPS" protocol. When an alert is sent from the server, the terminal immediately receives the information and displays the alert to the driver. This display is done through audio or a pop-up.

[0317] Users are mainly divided into drivers and operation managers. Drivers are required to take facial images using a terminal before and after work. This allows the system to always obtain the latest necessary data. Operation managers monitor alert information sent from the server. If an alert occurs, the operation manager reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0318] As a specific example, driver A takes a facial image on a device before work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion.

[0319] Here are some example prompts to input to a generative AI model:

[0320] "Please rate the level of fatigue of driver A based on the facial image."

[0321] "Please rate the emotions of driver A based on the facial image."

[0322] If the evaluation reveals that fatigue levels exceed the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to Person A's device. At the same time, an alert is also sent to the driver's device, allowing the driver to urge Person A to take a break and arrange for a substitute driver if necessary.

[0323] With this system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[0324] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0325] Step 1:

[0326] The terminal captures a facial image of the driver.

[0327] Specific operation: The device activates the camera and plays a voice prompt saying, "Please face forward." When the driver follows the instructions and takes a facial image, the image data is saved in the device's internal memory.

[0328] Input: Face image captured from the camera

[0329] Output: Facial image data stored in internal memory

[0330] Step 2:

[0331] The terminal transmits the facial image data to the server.

[0332] Specific operation: The device makes an HTTP request and sends the captured face image data to the server's API endpoint. If the request is successfully received, the device displays a notification saying "Image sent successfully."

[0333] Input: Facial image data stored in the internal memory

[0334] Output: Sending face image data via HTTP request to the server

[0335] Step 3:

[0336] The server receives and stores the facial image data.

[0337] Specific operation: When the server receives the facial image data, it records a "reception completed" log and saves the data to storage such as Amazon S3. If the save is successful, it updates the status to "save completed."

[0338] Input: Facial image data sent from the device

[0339] Output: Facial image data stored in storage (e.g. Amazon S3)

[0340] Step 4:

[0341] The server inputs the facial image data into the generation AI and emotion engine.

[0342] How it works: The server reads the stored facial image data and inputs it into a generative AI model using PyTorch or TensorFlow, and also into an emotion engine using OpenFace or the Microsoft Emotion API.

[0343] Input: Facial image data stored in storage (e.g. Amazon S3)

[0344] Output: Input data to the generative AI model and emotion engine

[0345] Step 5:

[0346] The server evaluates fatigue and emotion.

[0347] How it works: The server uses a generative AI model to assess fatigue levels and an emotion engine to identify the driver's emotions. The assessment results are displayed on a scale of 0 to 100.

[0348] Input: Input data to the generative AI model and emotion engine

[0349] Output: Fatigue rating score and emotion rating score

[0350] Step 6:

[0351] The server aggregates the evaluation results and generates alerts.

[0352] Specific operation: The server integrates the fatigue and emotion evaluation scores, and if the threshold is exceeded, generates an alert with the message "Caution: High fatigue, unstable emotions." After generating the alert, the status is updated to "Alert generation completed."

[0353] Input: Fatigue rating score and emotion rating score

[0354] Output: Alert information for drivers and operation managers

[0355] Step 7:

[0356] The server sends the alert information to the terminal and the operation manager.

[0357] Specific operation: The server uses "Twilio" to send the alert information to the driver's device and the operation manager. If the transmission is successful, the status is updated to "Successful transmission."

[0358] Input: Generated alert information

[0359] Output: Alert information sent to drivers and dispatchers

[0360] Step 8:

[0361] The device displays an alert.

[0362] Specific operation: The device notifies the driver of the received alert information with a voice notification and a pop-up. When the notification is displayed, the device updates the status to "Alert display completed."

[0363] Input: Alert information sent from the server

[0364] Output: Alert display to the driver

[0365] Step 9:

[0366] The user (operation manager) implements countermeasures.

[0367] Specific actions: The operations manager checks the alert, promptly urges the driver to take a break, and arranges for a substitute driver if necessary. After completing the countermeasures, the status is updated to "Response Completed."

[0368] Input: Alert information sent to drivers and dispatchers

[0369] Output: Implementing countermeasures such as changing the operation plan or arranging for a substitute driver

[0370] (Application example 2)

[0371] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0372] The challenge is to provide a system that can reliably monitor the fatigue level and emotional state of drivers and improve safety. In particular, previous systems only issued alerts to the driver, and cooperation with the operation manager and the autonomous driving system was insufficient. Furthermore, unless the alert was displayed to the driver quickly and visually, there were limitations to improving safety.

[0373] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0374] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server via a communication device, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for displaying the alert information on the driver's visualization display device, and means for notifying the driving system based on the evaluation result and adjusting the control of automated driving. This enables real-time monitoring of the driver's fatigue level and emotional state, enabling quick responses to improve safety.

[0375] "Driver" means a person engaged in the business of driving a vehicle.

[0376] A "face image" is image data that captures facial features.

[0377] "Communication device" is a general term for hardware or software for transmitting and receiving data.

[0378] A "server" is a computer system that stores, processes, and distributes data over a network.

[0379] "Generative AI" is a form of artificial intelligence that uses pre-trained models to extract, generate, or classify specific information from data.

[0380] "Fatigue level" is an index that quantitatively indicates the degree of fatigue of a person.

[0381] An "alert" is a notification that warns of an abnormal condition or alerts you to something.

[0382] A "visual display device" is a device for visually displaying information to a user.

[0383] "Driving system" is a general term for hardware and software used to manage and control the operation of a vehicle.

[0384] "Evaluation results" are judgments made based on data analysis.

[0385] A "database" is a system for systematically storing, managing, and searching data.

[0386] The present invention relates to a system for evaluating the fatigue level and emotions of a driver and supporting safe driving. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS ... will be specifically described below.

[0387] Server Roles and Operations

[0388] The server plays a central role in processing and analyzing the main data. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion assessments, but also for training the generative AI.

[0389] The server inputs image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the facial image, such as eye redness, eyelid movement, and facial color, and evaluates the fatigue level on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear. Specifically, it is recommended to use machine learning frameworks such as TensorFlow and PyTorch.

[0390] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the driver's visualization display device and the operation manager. Furthermore, the server notifies the driving system based on the evaluation result and adjusts the autonomous driving control.

[0391] Terminal roles and processing

[0392] The terminal is a device used by the driver and performs the following process: First, the driver takes a picture of his / her face using the camera on the terminal. The terminal temporarily stores the captured facial image data in its internal memory and then transmits it to a server via the network.

[0393] When an alert is sent from the server, the terminal immediately receives it and displays it on the driver's visualization display device, prompting the driver to reconsider safety checks.

[0394] User Roles and Actions

[0395] Users mainly play the roles of drivers and operation managers. Drivers always use the device to take facial images before and after work, which provides the necessary data to the server. Operation managers monitor alert information sent from the server, and when an alert occurs, they reconfirm the health status of drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[0396] Specific example explanation

[0397] For example, when a driver wears smart glasses, the built-in camera automatically captures facial images at regular intervals. The image data is sent to a server in real time, and the server uses a generative AI model to evaluate the driver's fatigue level and emotions. If the driver's fatigue level exceeds the threshold, an alert will appear on the smart glasses' display, prompting the driver to take a break. At the same time, the autonomous driving system will be notified, improving driving safety.

[0398] An example prompt might be "input_image: captured_image.jpg, evaluate: [fatigue_level, emotional_state]", which inputs a facial image into a generative AI model and evaluates fatigue level and emotion.

[0399] As described above, the system of the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby realizing safe driving. This system contributes to reducing accidents and improving the working environment.

[0400] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0401] Step 1:

[0402] The terminal takes a facial image of the driver. Specifically, it activates the built-in camera of the terminal and captures the driver's face. The camera is operated as input, and facial image data (image file) is obtained as output.

[0403] Step 2:

[0404] The device temporarily stores the captured facial image data in its internal memory. It receives facial image data as input and stores the data in its memory as output.

[0405] Step 3:

[0406] The device sends the facial image data to the server. It sends an HTTP POST request to the specified URL on the server via the network and uploads the facial image data. The input is the facial image data, and the output is a success response to the server that the data was sent.

[0407] Step 4:

[0408] The server inputs the received facial image data into a generative AI model. Specifically, the server uses machine learning frameworks such as TensorFlow and PyTorch to input data into a pre-trained model. The input is facial image data, and the output is the extracted features (eye redness, eyelid movement, facial color, etc.).

[0409] Step 5:

[0410] The server evaluates the driver's fatigue level based on the extracted features. The evaluation scale is 0 to 100, and the generation AI performs the calculations. The input is the extracted features, and the output is an evaluation score of the fatigue level.

[0411] Step 6:

[0412] The server inputs facial image data into the emotion engine, which identifies emotions such as joy, anger, sadness, surprise, and fear from the image. The input is the facial image data, and the output is the identified emotion data.

[0413] Step 7:

[0414] The server integrates the fatigue evaluation score and emotional data to perform a comprehensive evaluation. Based on this, it generates an alert if the threshold is exceeded. The input is the fatigue evaluation score and emotional data, and the output is alert information.

[0415] Step 8:

[0416] The server sends the generated alert information to the terminal. The alert is displayed on the visualization display device of the driver. The input is the alert information, and the output is the alert displayed on the terminal.

[0417] Step 9:

[0418] The server also sends an alert to the dispatcher based on the evaluation results. The notification from the server is sent to the dispatcher's terminal in a remote location, and the manager monitors it. The input is the alert information, and the output is the alert display on the dispatcher's terminal.

[0419] Step 10:

[0420] The server notifies the driving system based on the evaluation results and adjusts the autonomous driving control. Specifically, it changes the parameters of the autonomous driving system based on the evaluation results to ensure safe driving. The input is the evaluation results, and the output is the adjusted autonomous driving control parameters.

[0421] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0422] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0423] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0424] [Second embodiment]

[0425] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0426] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0427] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0428] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0429] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0430] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0431] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0432] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0433] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0434] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0435] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0436] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0437] The present invention relates to a system for assessing driver fatigue levels and ensuring operational safety. This system includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the driver's fatigue level using a generation AI on the server, and issuing an alert based on the results.

[0438] Server Roles and Operations

[0439] The server is responsible for the core processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Second, it saves the image data in a database. This saved data is used not only for subsequent fatigue assessments but also for training the generation AI.

[0440] The server then inputs the image data into the generation AI, which uses a pre-trained model to extract features from the image, such as bloodshot eyes, eyelid movement, and facial color. Based on the extracted features, the AI ​​evaluates the driver's fatigue level on a scale of 0 to 100. If the result exceeds the threshold, the server generates an alert and sends it to the device and the operation manager.

[0441] Terminal roles and processing

[0442] The terminal is a device used by the driver, and performs the following process. First, the driver uses the terminal to take a picture of his or her own face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. Next, the terminal sends the captured face image data to a server. This transmission is usually performed via a network.

[0443] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[0444] User Roles and Actions

[0445] Users mainly play two roles: drivers and operation managers. Drivers must take facial images using a terminal before and after work, which provides the necessary data to the server.

[0446] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0447] Specific example explanation

[0448] For example, driver A takes a picture of his face on his device before starting work. At this time, the device's camera activates and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0449] By using the above system and process, the present invention can efficiently and objectively evaluate the fatigue level of drivers and improve operational safety. This system contributes to reducing accidents and improving the working environment.

[0450] The processing flow will be explained below.

[0451] Specific processing steps of the program

[0452] Server Processing

[0453] Step 1:

[0454] The server receives the facial image data sent from the terminal. Specifically, the image data sent as an HTTP request is processed by the server's reception process and temporarily stored in memory.

[0455] Step 2:

[0456] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (e.g., transmission time, driver ID) in the database.

[0457] Step 3:

[0458] The server inputs the saved image data into the generative AI. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI model.

[0459] Step 4:

[0460] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[0461] Step 5:

[0462] The server evaluates the fatigue level and checks the results. If the evaluation result exceeds the reference value, an alert is generated.

[0463] Step 6:

[0464] The server sends the generated alert information to the terminal and the operation manager. Specifically, it executes a process to send an alert notification to the terminal and the manager's device.

[0465] Terminal handling

[0466] Step 1:

[0467] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[0468] Step 2:

[0469] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[0470] Step 3:

[0471] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[0472] Step 4:

[0473] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[0474] Step 5:

[0475] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[0476] User Action

[0477] Step 1:

[0478] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[0479] Step 2:

[0480] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[0481] Step 3:

[0482] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[0483] Step 4:

[0484] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[0485] Example 1

[0486] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0487] Driver fatigue is one of the major causes of traffic accidents, and its prevention and management are important. Conventional methods often rely on subjective self-reporting or simple checklists, making it difficult to provide an immediate, objective assessment. This creates a risk that highly fatigued drivers may continue driving. This invention aims to provide a system that uses facial images of drivers to quickly and accurately assess their fatigue level and improve driving safety.

[0488] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0489] In this invention, the server includes means for receiving facial image data and storing it in a database, means for inputting the facial image data into a generative AI model and extracting features such as eye congestion, eyelid movement, and facial color to evaluate fatigue level, and means for generating an alert if the fatigue level exceeds a reference value based on the evaluation result and sending the alert to the terminals of the driver and manager. This enables quick and accurate fatigue level evaluation based on the facial image of the driver, thereby improving driving safety.

[0490] "Driver" means a person who drives a vehicle professionally.

[0491] "Facial Image" refers to a photograph or digital image of a driving personnel's face.

[0492] "Terminal" refers to an electronic device used by a driver that has camera and communication capabilities.

[0493] "Network" refers to a communications infrastructure for transmitting data, including the Internet and mobile communications.

[0494] "Server" refers to a computer system for receiving, processing, and storing data over a network.

[0495] A "database" refers to a system for efficiently storing, retrieving, and managing data.

[0496] A "generative AI model" refers to a mathematical model that is trained to perform a specific task using artificial intelligence techniques.

[0497] "Features" are attributes and information extracted from facial images, including eye congestion, eyelid movement, and facial color.

[0498] "Fatigue level" refers to the degree of fatigue of the driver expressed numerically.

[0499] "Alert" refers to a warning or caution message sent to an operator or manager.

[0500] "Manager" refers to the person responsible for supervising and managing the health of drivers and the safety of operations.

[0501] This invention relates to a system for assessing the fatigue level of drivers and ensuring driving safety. Specifically, it includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the fatigue level using a generation AI on the server, and issuing an alert based on the results.

[0502] System Configuration

[0503] Device Role

[0504] The terminal is an electronic device used by the driver and has the following functions:

[0505] Camera function: Taking a facial image of the driver, for example, using the camera on a smartphone or tablet.

[0506] Communication function: Sends captured facial image data to a server via a network (Internet or mobile communications).

[0507] Display function: Displays alerts received from the server to the driver.

[0508] Server Roles

[0509] The server is responsible for the core processing and has the following functions:

[0510] Data reception and storage function: The server receives the facial image data sent from the device, temporarily stores it in memory, and then saves it in a database. The database used could be, for example, MySQL or PostgreSQL.

[0511] Generative AI model: The server inputs facial image data into a generative AI model (e.g., a model using TensorFlow or PyTorch) to extract features such as eye redness, eyelid movement, and facial color. Based on these features, the server then evaluates the level of fatigue on a scale of 0 to 100.

[0512] Alert generation and notification: If the evaluation result exceeds the reference value, the server generates alert information and sends it to the terminals of the driver and manager.

[0513] User Roles

[0514] Drivers use a device to take a picture of their face and send it to the server before and after work, providing the server with the necessary data.

[0515] Administrators monitor the alert information sent from the server, and when an alert occurs, they recheck the health status of the drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[0516] Specific examples

[0517] For example, driver A takes a facial image on his device before work. The device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into a generative AI model to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the manager's device, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0518] Prompt Sentence Examples

[0519] Below is an example of a prompt for configuring the generative AI model used in this system.

[0520] We would like to develop an AI system that uses facial images of drivers to assess their fatigue level. We would like you to create an AI model that extracts eye congestion, eyelid movement, and facial color from the images and evaluates fatigue level on a scale of 0 to 100.

[0521] This system and process will enable efficient and objective evaluation of driver fatigue levels and improve operational safety, contributing to reducing accidents and improving the working environment.

[0522] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0523] Step 1: Take an image (device processing)

[0524] Specific explanation: The driver activates the device's camera to take a facial image. The device uses the camera function to capture the image and temporarily stores it in its internal memory. At this time, the device checks that the image resolution, brightness, etc. are appropriate.

[0525] Input and output: The input is the driver's face, and the output is the captured facial image data.

[0526] Specific operation: The driver launches the device's camera app, positions their face on the screen, and presses the shutter button.

[0527] Step 2: Data transmission (terminal processing)

[0528] Specifically, the device sends the captured facial image data to the server via the network. This transmission is performed as encrypted data to ensure the security of the communication.

[0529] Input and output: The input is the captured facial image data, and the output is the data sent to the server.

[0530] Specific operation: After taking a photo, the device automatically encrypts the image data and sends a POST request to the server via the Internet.

[0531] Step 3: Data reception and storage (server processing)

[0532] Specific explanation: The server receives the facial image data sent from the device and temporarily stores it in memory. This image data is then saved in a database. The saved data is used for fatigue evaluation and as learning data for the generation AI.

[0533] Input and output: The input is the facial image data sent from the terminal, and the output is the data stored in the database.

[0534] Specific operation: The server decodes the received image data, stores it in temporary memory, and then stores it in the "face image" table of the database.

[0535] Step 4: Feature extraction and fatigue evaluation (server processing)

[0536] Specifically, the server inputs facial image data obtained from a database into a generative AI model. The generative AI model extracts features from the image, such as bloodshot eyes, eyelid movement, and facial color. Then, based on the extracted features, the AI ​​evaluates the fatigue level on a scale of 0 to 100.

[0537] Input and output: The input is face image data obtained from the database, and the output is the fatigue score.

[0538] Specific operation: The server inputs image data into the TensorFlow model, and evaluates and integrates the eye bloodshot score, eyelid open / close score, and facial color score.

[0539] Step 5: Alert generation and notification (server processing)

[0540] Specific explanation: If the evaluation result exceeds the reference value, the server generates an alert information, which is sent to the driver's terminal and the manager's terminal.

[0541] Inputs and Outputs: The input is the fatigue score, and the output is the alert sent to the driver and manager.

[0542] Specific operation: When the evaluation score exceeds a threshold value (e.g., 80), the server generates an alert message and sends a push notification to the devices of the driver and manager.

[0543] Step 6: Receiving and displaying alerts (terminal processing)

[0544] Specific explanation: The device will receive the alert message as soon as it is sent and display an alert to the driver, which will include specific instructions such as "take a break."

[0545] Input and Output: The input is the alert message sent from the server, and the output is the alert displayed to the driver.

[0546] Specific operation: The driver's device receives a push notification and displays the message, "You are highly fatigued. Please take a break."

[0547] Step 7: Check the alert and respond (user action)

[0548] Specific explanation: Users (drivers and managers) ensure that drivers check the alerts on their devices and take breaks if necessary. Managers recheck the health status of drivers, and if necessary, change the operation plan and arrange for a substitute driver.

[0549] Input and output: Input is an alert message sent from a terminal or server, and output is a response such as a change in the operation plan or a driver's break.

[0550] Specific behavior:

[0551] The driver checks the alert, stops the vehicle in a safe place, and takes a break.

[0552] The administrator checks the driver's status on the system's management screen and, if necessary, contacts other drivers to arrange a replacement.

[0553] (Application example 1)

[0554] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0555] Conventional driver fatigue assessment systems have difficulty assessing driver fatigue in real time, making it difficult to issue alerts in a timely manner. Furthermore, fatigue assessments cannot be performed while driving, which in turn makes it difficult to ensure safety. Therefore, there is a need for a system that can assess driver fatigue in real time while driving and immediately notify drivers of alerts.

[0556] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0557] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for periodically capturing a facial image of the driver using smart glasses and transmitting the captured facial image data to the server in real time, and means for displaying an audio or visual alert on the smart glasses based on the evaluation result. This makes it possible to evaluate the fatigue level in real time even while driving and to immediately notify the driver of the alert.

[0558] "Operator" means an individual responsible for operating or driving a vehicle.

[0559] "Facial image" refers to video data showing the driver's face.

[0560] "Server" refers to the computer system that receives and stores facial image data and analyzes the data using a generative AI model.

[0561] "Fatigue level" is an index that indicates the degree of physical and mental fatigue of drivers.

[0562] "Generative AI" is an artificial intelligence model that learns from large amounts of data and uses the results to analyze and predict new data.

[0563] An "alert" is a warning message that is issued to call attention when a set reference value is exceeded.

[0564] "Smart glasses" are eyeglass-type devices that display digital information and connect to the Internet and applications.

[0565] "Real-time" is a term that refers to processing or transmission of information immediately, without delay.

[0566] "Voice alert" refers to a function that issues warnings or notifications by voice.

[0567] "Visual alert" refers to a function that provides warnings or notifications through visual means such as a display or LED.

[0568] The present invention relates to a system that acquires facial images of drivers through smart glasses, transmits the images to a server in real time, evaluates their fatigue level using a generative AI model, and issues audio and visual alerts.

[0569] System configuration

[0570] Server Roles and Operations

[0571] The server receives, stores, and generates AI to evaluate the level of fatigue and generates an alert if the level exceeds the threshold. The main hardware and software used are as follows:

[0572] Hardware: High-performance server machine

[0573] Software: Python, generative AI models, database systems

[0574] The server first receives the facial image data sent from the smart glasses and temporarily stores it in a database. It then inputs the image data into a generative AI model to extract features and evaluate the level of fatigue based on information such as bloodshot eyes, eyelid movement, and facial color. If the evaluation result exceeds a threshold, an alert is generated and sent to the driver's smart glasses and the operations manager.

[0575] The role and processing of smart glasses

[0576] The smart glasses used by drivers will:

[0577] Photographing: Take regular photographs of your face using a camera

[0578] Send: Send the captured facial image data to the server in real time

[0579] Alert Indication: Audio and visual indication of alerts sent from the server

[0580] The main hardware and software used are:

[0581] Hardware: Smart glasses (camera, microphone, display)

[0582] Software: Photo and transmission application, alert display application

[0583] User Roles and Actions

[0584] Users have two main roles: Driver and Operations Manager:

[0585] Driver: Wears smart glasses, takes a facial image and sends it to the server. This process is automatic while driving.

[0586] Operation manager: Monitors alert information sent from the server and makes changes to operation plans or arranges for substitute drivers as necessary.

[0587] Specific examples

[0588] For example, driver A puts on his smart glasses and begins driving. The smart glasses periodically take images of A's face and send the data to a server in real time. The server uses a generative AI model to evaluate A's fatigue level, and if the evaluation result exceeds the standard value, it issues a voice alert to A's smart glasses saying, "Your fatigue level is high. Please take a break." A similar alert is also sent to the operations manager, who checks the situation and takes action as necessary.

[0589] Examples of prompt statements

[0590] "Rate the fatigue level of the person in this image on a scale of 0-100."

[0591] The above is the details of each part in the embodiment of the present invention.

[0592] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0593] Step 1:

[0594] The driver wears smart glasses, and the camera in the smart glasses periodically captures facial images.

[0595] Input: Camera image

[0596] Data processing: Smart glasses camera captures driver's face

[0597] Output: Face image data

[0598] Step 2:

[0599] The captured facial image data is temporarily stored in the internal memory of the smart glasses.

[0600] Input: Facial image data

[0601] Data processing: Temporary storage of data

[0602] Output: Facial image data stored in internal memory

[0603] Step 3:

[0604] Facial image data is transmitted to a server in real time.

[0605] Input: Facial image data stored in the internal memory

[0606] Data processing: Sending image data

[0607] Output: Facial image data arriving at the server

[0608] Step 4:

[0609] The server receives the facial image data and temporarily stores it in memory.

[0610] Input: Facial image data that arrives at the server

[0611] Data processing: storing in memory

[0612] Output: Temporarily stored face image data

[0613] Step 5:

[0614] The server stores the facial image data in a database and inputs the stored data into a generative AI model.

[0615] Input: Temporarily stored facial image data

[0616] Data processing: storing in a database and inputting into a generative AI model

[0617] Output: Facial image data input to the generative AI model

[0618] Step 6:

[0619] The generative AI model extracts features from facial image data and assesses fatigue levels based on information such as eye congestion, eyelid movement, and facial color.

[0620] Input: Facial image data

[0621] Data calculation: feature extraction and fatigue evaluation

[0622] Output: Fatigue evaluation results

[0623] Step 7:

[0624] If the evaluation result exceeds the reference value, the server generates alert information and sends the alert to the driver's smart glasses and the operation manager.

[0625] Input: Fatigue evaluation result

[0626] Data Calculation: Alert Generation

[0627] Output: Alert information

[0628] Step 8:

[0629] The driver's smart glasses receive the alert information and display audio and visual alerts.

[0630] Input: Alert information

[0631] Data processing: Alert display

[0632] Output: Audio and visual alerts

[0633] Specific examples

[0634] For example, driver A puts on his smart glasses and steps 1 to 8 are carried out. An example of a prompt sentence at this time is "Please rate the fatigue level of the person in this image on a scale of 0 to 100." This series of steps makes it possible to assess fatigue levels in real time while driving, ensuring safety.

[0635] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0636] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure driving safety. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generation AI on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[0637] Server Roles and Operations

[0638] The server is responsible for the core processing. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion evaluations, but also for training the generative AI.

[0639] The server then inputs the image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the image, such as eye redness, eyelid movement, and facial color, and evaluates fatigue levels on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear.

[0640] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the device and the operation manager.

[0641] Terminal roles and processing

[0642] The terminal is a device used by the driver, and performs the following process: First, the driver uses the terminal to take a picture of his or her face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. The terminal then transmits the captured face image data to a server. Transmission is usually performed via a network.

[0643] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[0644] User Roles and Actions

[0645] Users mainly play two roles: drivers and operation managers. Drivers always use a device to take a facial image before and after work, which provides the necessary data to the server.

[0646] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0647] Specific example explanation

[0648] For example, driver A takes a facial image on his device before starting work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion. If the evaluation reveals that fatigue level exceeds the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0649] With the above system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[0650] The processing flow will be explained below.

[0651] Server Processing

[0652] Step 1:

[0653] The server receives the facial image data sent from the terminal. Specifically, it receives the image data sent as an HTTP request and temporarily stores it in memory.

[0654] Step 2:

[0655] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (such as transmission time and driver ID) in the database.

[0656] Step 3:

[0657] The server inputs the saved image data into the generative AI and emotion engine. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI and emotion engine.

[0658] Step 4:

[0659] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[0660] Step 5:

[0661] The server uses an emotion engine to analyze emotions from facial image data, specifically identifying emotions such as joy, anger, sadness, surprise, and fear, and converting them into numerical data.

[0662] Step 6:

[0663] The server then integrates the extracted fatigue level and emotion data to make a comprehensive assessment. For example, if fatigue is high and strong stress or anger is also detected, the server will assess the risk as high.

[0664] Step 7:

[0665] If the evaluation result exceeds the reference value, the server generates an alert message that includes the driver ID and the evaluation result.

[0666] Step 8:

[0667] The server sends the generated alert information to the terminal and the operation manager. Specifically, it sends the alert notification as an HTTP request to the terminal and the manager's device.

[0668] Terminal handling

[0669] Step 1:

[0670] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[0671] Step 2:

[0672] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[0673] Step 3:

[0674] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[0675] Step 4:

[0676] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[0677] Step 5:

[0678] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[0679] User Action

[0680] Step 1:

[0681] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[0682] Step 2:

[0683] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[0684] Step 3:

[0685] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[0686] Step 4:

[0687] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[0688] Example 2

[0689] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0690] Conventional driver health management systems mainly rely on self-reporting or some biometric sensors, making it difficult to accurately assess fatigue levels and emotions. Furthermore, it is difficult to issue warnings at the appropriate time, making it difficult to ensure operational safety. To address this issue, the present invention aims to provide a system that uses facial images of drivers to assess their fatigue levels and emotions and take appropriate countermeasures accurately and quickly.

[0691] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0692] In this invention, the server includes a means for capturing a facial image of the driver, a means for transmitting the captured facial image data to a remote computing device, a means for using artificial intelligence on the computing device to evaluate the fatigue level from the facial image data, and a means for issuing an alarm based on the evaluation result, thereby enabling accurate evaluation of the fatigue level and emotions of the driver based on the facial image and improving driving safety.

[0693] "Operator" means a person responsible for operating a vehicle.

[0694] The "face image" is image data of the driver's face.

[0695] "Capturing means" refers to a camera or other recording device used to capture a facial image.

[0696] The "transmitting means" refers to a communication device or software for transferring captured facial image data to a remote computing device.

[0697] A "server" is a type of computing device that is a data processing device that receives, stores, and evaluates facial image data.

[0698] "Computing device" refers to a device that contains hardware and software for processing data.

[0699] "Fatigue level" is an indicator of the driver's fatigue state and is evaluated using a generative AI model.

[0700] "Generative AI" refers to an artificial intelligence model that has been pre-trained to assess fatigue levels from facial image data.

[0701] "Evaluation results" are the analysis results of fatigue levels and emotions obtained by the generative AI and emotion engine.

[0702] An "alert" is a warning or notification issued to a driver or operations manager based on the evaluation results.

[0703] "Means for issuing an alert" refers to a method or device for generating an alert based on the evaluation results and notifying the driver or operations manager.

[0704] The "storage device" refers to a data storage device for storing photographed face image data and evaluation results.

[0705] "Means for assessing emotions" refers to software or systems for identifying emotions of drivers from facial image data.

[0706] "Comprehensive assessment" is a process of integrating the results of fatigue and emotional assessment to comprehensively judge the driver's condition.

[0707] "Terminal" refers to an electronic device used by a driver, and is equipped with a camera and communication functions.

[0708] "Operations manager" means a person responsible for managing the driving staff and the operational status of the vehicle.

[0709] "Means of notification" refers to the method or device for communicating an alert to an operator or dispatcher.

[0710] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure safe driving. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generative AI model on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[0711] The server is the core device responsible for the main processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. Database and storage systems such as MySQL and Amazon S3 are used for this purpose. The image data is also used for subsequent fatigue and emotion assessments, as well as for training the generative AI model.

[0712] The server then inputs the received image data into the generation AI and emotion engine. The generation AI uses pre-trained PyTorch and TensorFlow to extract fatigue characteristics from the image (such as bloodshot eyes, eyelid movement, and facial color). At the same time, the emotion engine uses OpenFace and the Microsoft Emotion API to analyze facial expressions from the facial image and identify emotions such as joy, anger, sadness, surprise, and fear. The evaluation results are displayed on a scale of 0 to 100.

[0713] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert using services such as "Twilio" and sends it to the device and the operation manager. This enables a prompt response according to the driver's condition.

[0714] The terminal is a device that is directly used by the driver. First, the terminal uses its camera function to capture a facial image of the driver and temporarily stores it in its internal memory. Then, it sends this captured facial image data to a server via the "HTTP" or "HTTPS" protocol. When an alert is sent from the server, the terminal immediately receives the information and displays the alert to the driver. This display is done through audio or a pop-up.

[0715] Users are mainly divided into drivers and operation managers. Drivers are required to take facial images using a terminal before and after work. This allows the system to always obtain the latest necessary data. Operation managers monitor alert information sent from the server. If an alert occurs, the operation manager reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0716] As a specific example, driver A takes a facial image on a device before work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion.

[0717] Here are some example prompts to input to a generative AI model:

[0718] "Please rate the level of fatigue of driver A based on the facial image."

[0719] "Please rate the emotions of driver A based on the facial image."

[0720] If the evaluation reveals that fatigue levels exceed the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to Person A's device. At the same time, an alert is also sent to the driver's device, allowing the driver to urge Person A to take a break and arrange for a substitute driver if necessary.

[0721] With this system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[0722] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0723] Step 1:

[0724] The terminal captures a facial image of the driver.

[0725] Specific operation: The device activates the camera and plays a voice prompt saying, "Please face forward." When the driver follows the instructions and takes a facial image, the image data is saved in the device's internal memory.

[0726] Input: Face image captured from the camera

[0727] Output: Facial image data stored in internal memory

[0728] Step 2:

[0729] The terminal transmits the facial image data to the server.

[0730] Specific operation: The device makes an HTTP request and sends the captured face image data to the server's API endpoint. If the request is successfully received, the device displays a notification saying "Image sent successfully."

[0731] Input: Facial image data stored in the internal memory

[0732] Output: Sending face image data via HTTP request to the server

[0733] Step 3:

[0734] The server receives and stores the facial image data.

[0735] Specific operation: When the server receives the facial image data, it records a "reception completed" log and saves the data to storage such as Amazon S3. If the save is successful, it updates the status to "save completed."

[0736] Input: Facial image data sent from the device

[0737] Output: Facial image data stored in storage (e.g. Amazon S3)

[0738] Step 4:

[0739] The server inputs the facial image data into the generation AI and emotion engine.

[0740] How it works: The server reads the stored facial image data and inputs it into a generative AI model using PyTorch or TensorFlow, and also into an emotion engine using OpenFace or the Microsoft Emotion API.

[0741] Input: Facial image data stored in storage (e.g. Amazon S3)

[0742] Output: Input data to the generative AI model and emotion engine

[0743] Step 5:

[0744] The server evaluates fatigue and emotion.

[0745] How it works: The server uses a generative AI model to assess fatigue levels and an emotion engine to identify the driver's emotions. The assessment results are displayed on a scale of 0 to 100.

[0746] Input: Input data to the generative AI model and emotion engine

[0747] Output: Fatigue rating score and emotion rating score

[0748] Step 6:

[0749] The server aggregates the evaluation results and generates alerts.

[0750] Specific operation: The server integrates the fatigue and emotion evaluation scores, and if the threshold is exceeded, generates an alert with the message "Caution: High fatigue, unstable emotions." After generating the alert, the status is updated to "Alert generation completed."

[0751] Input: Fatigue rating score and emotion rating score

[0752] Output: Alert information for drivers and operation managers

[0753] Step 7:

[0754] The server sends the alert information to the terminal and the operation manager.

[0755] Specific operation: The server uses "Twilio" to send the alert information to the driver's device and the operation manager. If the transmission is successful, the status is updated to "Successful transmission."

[0756] Input: Generated alert information

[0757] Output: Alert information sent to drivers and dispatchers

[0758] Step 8:

[0759] The device displays an alert.

[0760] Specific operation: The device notifies the driver of the received alert information with a voice notification and a pop-up. When the notification is displayed, the device updates the status to "Alert display completed."

[0761] Input: Alert information sent from the server

[0762] Output: Alert display to the driver

[0763] Step 9:

[0764] The user (operation manager) implements countermeasures.

[0765] Specific actions: The operations manager checks the alert, promptly urges the driver to take a break, and arranges for a substitute driver if necessary. After completing the countermeasures, the status is updated to "Response Completed."

[0766] Input: Alert information sent to drivers and dispatchers

[0767] Output: Implementing countermeasures such as changing the operation plan or arranging for a substitute driver

[0768] (Application example 2)

[0769] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0770] The challenge is to provide a system that can reliably monitor the fatigue level and emotional state of drivers and improve safety. In particular, previous systems only issued alerts to the driver, and cooperation with the operation manager and the autonomous driving system was insufficient. Furthermore, unless the alert was displayed to the driver quickly and visually, there were limitations to improving safety.

[0771] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0772] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server via a communication device, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for displaying the alert information on the driver's visualization display device, and means for notifying the driving system based on the evaluation result and adjusting the control of automated driving. This enables real-time monitoring of the driver's fatigue level and emotional state, enabling quick responses to improve safety.

[0773] "Driver" means a person engaged in the business of driving a vehicle.

[0774] A "face image" is image data that captures facial features.

[0775] "Communication device" is a general term for hardware or software for transmitting and receiving data.

[0776] A "server" is a computer system that stores, processes, and distributes data over a network.

[0777] "Generative AI" is a form of artificial intelligence that uses pre-trained models to extract, generate, or classify specific information from data.

[0778] "Fatigue level" is an index that quantitatively indicates the degree of fatigue of a person.

[0779] An "alert" is a notification that warns of an abnormal condition or alerts you to something.

[0780] A "visual display device" is a device for visually displaying information to a user.

[0781] "Driving system" is a general term for hardware and software used to manage and control the operation of a vehicle.

[0782] "Evaluation results" are judgments made based on data analysis.

[0783] A "database" is a system for systematically storing, managing, and searching data.

[0784] The present invention relates to a system for evaluating the fatigue level and emotions of a driver and supporting safe driving. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS ... will be specifically described below.

[0785] Server Roles and Operations

[0786] The server plays a central role in processing and analyzing the main data. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion assessments, but also for training the generative AI.

[0787] The server inputs image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the facial image, such as eye redness, eyelid movement, and facial color, and evaluates the fatigue level on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear. Specifically, it is recommended to use machine learning frameworks such as TensorFlow and PyTorch.

[0788] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the driver's visualization display device and the operation manager. Furthermore, the server notifies the driving system based on the evaluation result and adjusts the autonomous driving control.

[0789] Terminal roles and processing

[0790] The terminal is a device used by the driver and performs the following process: First, the driver takes a picture of his / her face using the camera on the terminal. The terminal temporarily stores the captured facial image data in its internal memory and then transmits it to a server via the network.

[0791] When an alert is sent from the server, the terminal immediately receives it and displays it on the driver's visualization display device, prompting the driver to reconsider safety checks.

[0792] User Roles and Actions

[0793] Users mainly play the roles of drivers and operation managers. Drivers always use the device to take facial images before and after work, which provides the necessary data to the server. Operation managers monitor alert information sent from the server, and when an alert occurs, they reconfirm the health status of drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[0794] Specific example explanation

[0795] For example, when a driver wears smart glasses, the built-in camera automatically captures facial images at regular intervals. The image data is sent to a server in real time, and the server uses a generative AI model to evaluate the driver's fatigue level and emotions. If the driver's fatigue level exceeds the threshold, an alert will appear on the smart glasses' display, prompting the driver to take a break. At the same time, the autonomous driving system will be notified, improving driving safety.

[0796] An example prompt might be "input_image: captured_image.jpg, evaluate: [fatigue_level, emotional_state]", which inputs a facial image into a generative AI model and evaluates fatigue level and emotion.

[0797] As described above, the system of the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby realizing safe driving. This system contributes to reducing accidents and improving the working environment.

[0798] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0799] Step 1:

[0800] The terminal takes a facial image of the driver. Specifically, it activates the built-in camera of the terminal and captures the driver's face. The camera is operated as input, and facial image data (image file) is obtained as output.

[0801] Step 2:

[0802] The device temporarily stores the captured facial image data in its internal memory. It receives facial image data as input and stores the data in its memory as output.

[0803] Step 3:

[0804] The device sends the facial image data to the server. It sends an HTTP POST request to the specified URL on the server via the network and uploads the facial image data. The input is the facial image data, and the output is a success response to the server that the data was sent.

[0805] Step 4:

[0806] The server inputs the received facial image data into a generative AI model. Specifically, the server uses machine learning frameworks such as TensorFlow and PyTorch to input data into a pre-trained model. The input is facial image data, and the output is the extracted features (eye redness, eyelid movement, facial color, etc.).

[0807] Step 5:

[0808] The server evaluates the driver's fatigue level based on the extracted features. The evaluation scale is 0 to 100, and the generation AI performs the calculations. The input is the extracted features, and the output is an evaluation score of the fatigue level.

[0809] Step 6:

[0810] The server inputs facial image data into the emotion engine, which identifies emotions such as joy, anger, sadness, surprise, and fear from the image. The input is the facial image data, and the output is the identified emotion data.

[0811] Step 7:

[0812] The server integrates the fatigue evaluation score and emotional data to perform a comprehensive evaluation. Based on this, it generates an alert if the threshold is exceeded. The input is the fatigue evaluation score and emotional data, and the output is alert information.

[0813] Step 8:

[0814] The server sends the generated alert information to the terminal. The alert is displayed on the visualization display device of the driver. The input is the alert information, and the output is the alert displayed on the terminal.

[0815] Step 9:

[0816] The server also sends an alert to the dispatcher based on the evaluation results. The notification from the server is sent to the dispatcher's terminal in a remote location, and the manager monitors it. The input is the alert information, and the output is the alert display on the dispatcher's terminal.

[0817] Step 10:

[0818] The server notifies the driving system based on the evaluation results and adjusts the autonomous driving control. Specifically, it changes the parameters of the autonomous driving system based on the evaluation results to ensure safe driving. The input is the evaluation results, and the output is the adjusted autonomous driving control parameters.

[0819] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0820] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0821] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0822] [Third embodiment]

[0823] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0824] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0825] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0826] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0827] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0828] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0829] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0830] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0831] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0832] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0833] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0834] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0835] The present invention relates to a system for assessing driver fatigue levels and ensuring operational safety. This system includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the driver's fatigue level using a generation AI on the server, and issuing an alert based on the results.

[0836] Server Roles and Operations

[0837] The server is responsible for the core processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Second, it saves the image data in a database. This saved data is used not only for subsequent fatigue assessments but also for training the generation AI.

[0838] The server then inputs the image data into the generation AI, which uses a pre-trained model to extract features from the image, such as bloodshot eyes, eyelid movement, and facial color. Based on the extracted features, the AI ​​evaluates the driver's fatigue level on a scale of 0 to 100. If the result exceeds the threshold, the server generates an alert and sends it to the device and the operation manager.

[0839] Terminal roles and processing

[0840] The terminal is a device used by the driver, and performs the following process. First, the driver uses the terminal to take a picture of his or her own face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. Next, the terminal sends the captured face image data to a server. This transmission is usually performed via a network.

[0841] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[0842] User Roles and Actions

[0843] Users mainly play two roles: drivers and operation managers. Drivers must take facial images using a terminal before and after work, which provides the necessary data to the server.

[0844] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[0845] Specific example explanation

[0846] For example, driver A takes a picture of his face on his device before starting work. At this time, the device's camera activates and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0847] By using the above system and process, the present invention can efficiently and objectively evaluate the fatigue level of drivers and improve operational safety. This system contributes to reducing accidents and improving the working environment.

[0848] The processing flow will be explained below.

[0849] Specific processing steps of the program

[0850] Server Processing

[0851] Step 1:

[0852] The server receives the facial image data sent from the terminal. Specifically, the image data sent as an HTTP request is processed by the server's reception process and temporarily stored in memory.

[0853] Step 2:

[0854] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (e.g., transmission time, driver ID) in the database.

[0855] Step 3:

[0856] The server inputs the saved image data into the generative AI. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI model.

[0857] Step 4:

[0858] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[0859] Step 5:

[0860] The server evaluates the fatigue level and checks the results. If the evaluation result exceeds the reference value, an alert is generated.

[0861] Step 6:

[0862] The server sends the generated alert information to the terminal and the operation manager. Specifically, it executes a process to send an alert notification to the terminal and the manager's device.

[0863] Terminal handling

[0864] Step 1:

[0865] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[0866] Step 2:

[0867] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[0868] Step 3:

[0869] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[0870] Step 4:

[0871] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[0872] Step 5:

[0873] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[0874] User Action

[0875] Step 1:

[0876] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[0877] Step 2:

[0878] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[0879] Step 3:

[0880] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[0881] Step 4:

[0882] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[0883] Example 1

[0884] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0885] Driver fatigue is one of the major causes of traffic accidents, and its prevention and management are important. Conventional methods often rely on subjective self-reporting or simple checklists, making it difficult to provide an immediate, objective assessment. This creates a risk that highly fatigued drivers may continue driving. This invention aims to provide a system that uses facial images of drivers to quickly and accurately assess their fatigue level and improve driving safety.

[0886] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0887] In this invention, the server includes means for receiving facial image data and storing it in a database, means for inputting the facial image data into a generative AI model and extracting features such as eye congestion, eyelid movement, and facial color to evaluate fatigue level, and means for generating an alert if the fatigue level exceeds a reference value based on the evaluation result and sending the alert to the terminals of the driver and manager. This enables quick and accurate fatigue level evaluation based on the facial image of the driver, thereby improving driving safety.

[0888] "Driver" means a person who drives a vehicle professionally.

[0889] "Facial Image" refers to a photograph or digital image of a driving personnel's face.

[0890] "Terminal" refers to an electronic device used by a driver that has camera and communication capabilities.

[0891] "Network" refers to a communications infrastructure for transmitting data, including the Internet and mobile communications.

[0892] "Server" refers to a computer system for receiving, processing, and storing data over a network.

[0893] A "database" refers to a system for efficiently storing, retrieving, and managing data.

[0894] A "generative AI model" refers to a mathematical model that is trained to perform a specific task using artificial intelligence techniques.

[0895] "Features" are attributes and information extracted from facial images, including eye congestion, eyelid movement, and facial color.

[0896] "Fatigue level" refers to the degree of fatigue of the driver expressed numerically.

[0897] "Alert" refers to a warning or caution message sent to an operator or manager.

[0898] "Manager" refers to the person responsible for supervising and managing the health of drivers and the safety of operations.

[0899] This invention relates to a system for assessing the fatigue level of drivers and ensuring driving safety. Specifically, it includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the fatigue level using a generation AI on the server, and issuing an alert based on the results.

[0900] System Configuration

[0901] Device Role

[0902] The terminal is an electronic device used by the driver and has the following functions:

[0903] Camera function: Taking a facial image of the driver, for example, using the camera on a smartphone or tablet.

[0904] Communication function: Sends captured facial image data to a server via a network (Internet or mobile communications).

[0905] Display function: Displays alerts received from the server to the driver.

[0906] Server Roles

[0907] The server is responsible for the core processing and has the following functions:

[0908] Data reception and storage function: The server receives the facial image data sent from the device, temporarily stores it in memory, and then saves it in a database. The database used could be, for example, MySQL or PostgreSQL.

[0909] Generative AI model: The server inputs facial image data into a generative AI model (e.g., a model using TensorFlow or PyTorch) to extract features such as eye redness, eyelid movement, and facial color. Based on these features, the server then evaluates the level of fatigue on a scale of 0 to 100.

[0910] Alert generation and notification: If the evaluation result exceeds the reference value, the server generates alert information and sends it to the terminals of the driver and manager.

[0911] User Roles

[0912] Drivers use a device to take a picture of their face and send it to the server before and after work, providing the server with the necessary data.

[0913] Administrators monitor the alert information sent from the server, and when an alert occurs, they recheck the health status of the drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[0914] Specific examples

[0915] For example, driver A takes a facial image on his device before work. The device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into a generative AI model to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the manager's device, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[0916] Prompt Sentence Examples

[0917] Below is an example of a prompt for configuring the generative AI model used in this system.

[0918] We would like to develop an AI system that uses facial images of drivers to assess their fatigue level. We would like you to create an AI model that extracts eye congestion, eyelid movement, and facial color from the images and evaluates fatigue level on a scale of 0 to 100.

[0919] This system and process will enable efficient and objective evaluation of driver fatigue levels and improve operational safety, contributing to reducing accidents and improving the working environment.

[0920] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0921] Step 1: Take an image (device processing)

[0922] Specific explanation: The driver activates the device's camera to take a facial image. The device uses the camera function to capture the image and temporarily stores it in its internal memory. At this time, the device checks that the image resolution, brightness, etc. are appropriate.

[0923] Input and output: The input is the driver's face, and the output is the captured facial image data.

[0924] Specific operation: The driver launches the device's camera app, positions their face on the screen, and presses the shutter button.

[0925] Step 2: Data transmission (terminal processing)

[0926] Specifically, the device sends the captured facial image data to the server via the network. This transmission is performed as encrypted data to ensure the security of the communication.

[0927] Input and output: The input is the captured facial image data, and the output is the data sent to the server.

[0928] Specific operation: After taking a photo, the device automatically encrypts the image data and sends a POST request to the server via the Internet.

[0929] Step 3: Data reception and storage (server processing)

[0930] Specific explanation: The server receives the facial image data sent from the device and temporarily stores it in memory. This image data is then saved in a database. The saved data is used for fatigue evaluation and as learning data for the generation AI.

[0931] Input and output: The input is the facial image data sent from the terminal, and the output is the data stored in the database.

[0932] Specific operation: The server decodes the received image data, stores it in temporary memory, and then stores it in the "face image" table of the database.

[0933] Step 4: Feature extraction and fatigue evaluation (server processing)

[0934] Specifically, the server inputs facial image data obtained from a database into a generative AI model. The generative AI model extracts features from the image, such as bloodshot eyes, eyelid movement, and facial color. Then, based on the extracted features, the AI ​​evaluates the fatigue level on a scale of 0 to 100.

[0935] Input and output: The input is face image data obtained from the database, and the output is the fatigue score.

[0936] Specific operation: The server inputs image data into the TensorFlow model, and evaluates and integrates the eye bloodshot score, eyelid open / close score, and facial color score.

[0937] Step 5: Alert generation and notification (server processing)

[0938] Specific explanation: If the evaluation result exceeds the reference value, the server generates an alert information, which is sent to the driver's terminal and the manager's terminal.

[0939] Inputs and Outputs: The input is the fatigue score, and the output is the alert sent to the driver and manager.

[0940] Specific operation: When the evaluation score exceeds a threshold value (e.g., 80), the server generates an alert message and sends a push notification to the devices of the driver and manager.

[0941] Step 6: Receiving and displaying alerts (terminal processing)

[0942] Specific explanation: The device will receive the alert message as soon as it is sent and display an alert to the driver, which will include specific instructions such as "take a break."

[0943] Input and Output: The input is the alert message sent from the server, and the output is the alert displayed to the driver.

[0944] Specific operation: The driver's device receives a push notification and displays the message, "You are highly fatigued. Please take a break."

[0945] Step 7: Check the alert and respond (user action)

[0946] Specific explanation: Users (drivers and managers) ensure that drivers check the alerts on their devices and take breaks if necessary. Managers recheck the health status of drivers, and if necessary, change the operation plan and arrange for a substitute driver.

[0947] Input and output: Input is an alert message sent from a terminal or server, and output is a response such as a change in the operation plan or a driver's break.

[0948] Specific behavior:

[0949] The driver checks the alert, stops the vehicle in a safe place, and takes a break.

[0950] The administrator checks the driver's status on the system's management screen and, if necessary, contacts other drivers to arrange a replacement.

[0951] (Application example 1)

[0952] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0953] Conventional driver fatigue assessment systems have difficulty assessing driver fatigue in real time, making it difficult to issue alerts in a timely manner. Furthermore, fatigue assessments cannot be performed while driving, which in turn makes it difficult to ensure safety. Therefore, there is a need for a system that can assess driver fatigue in real time while driving and immediately notify drivers of alerts.

[0954] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0955] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for periodically capturing a facial image of the driver using smart glasses and transmitting the captured facial image data to the server in real time, and means for displaying an audio or visual alert on the smart glasses based on the evaluation result. This makes it possible to evaluate the fatigue level in real time even while driving and to immediately notify the driver of the alert.

[0956] "Operator" means an individual responsible for operating or driving a vehicle.

[0957] "Facial image" refers to video data showing the driver's face.

[0958] "Server" refers to the computer system that receives and stores facial image data and analyzes the data using a generative AI model.

[0959] "Fatigue level" is an index that indicates the degree of physical and mental fatigue of drivers.

[0960] "Generative AI" is an artificial intelligence model that learns from large amounts of data and uses the results to analyze and predict new data.

[0961] An "alert" is a warning message that is issued to call attention when a set reference value is exceeded.

[0962] "Smart glasses" are eyeglass-type devices that display digital information and connect to the Internet and applications.

[0963] "Real-time" is a term that refers to processing or transmission of information immediately, without delay.

[0964] "Voice alert" refers to a function that issues warnings or notifications by voice.

[0965] "Visual alert" refers to a function that provides warnings or notifications through visual means such as a display or LED.

[0966] The present invention relates to a system that acquires facial images of drivers through smart glasses, transmits the images to a server in real time, evaluates their fatigue level using a generative AI model, and issues audio and visual alerts.

[0967] System configuration

[0968] Server Roles and Operations

[0969] The server receives, stores, and generates AI to evaluate the level of fatigue and generates an alert if the level exceeds the threshold. The main hardware and software used are as follows:

[0970] Hardware: High-performance server machine

[0971] Software: Python, generative AI models, database systems

[0972] The server first receives the facial image data sent from the smart glasses and temporarily stores it in a database. It then inputs the image data into a generative AI model to extract features and evaluate the level of fatigue based on information such as bloodshot eyes, eyelid movement, and facial color. If the evaluation result exceeds a threshold, an alert is generated and sent to the driver's smart glasses and the operations manager.

[0973] The role and processing of smart glasses

[0974] The smart glasses used by drivers will:

[0975] Photographing: Take regular photographs of your face using a camera

[0976] Send: Send the captured facial image data to the server in real time

[0977] Alert Indication: Audio and visual indication of alerts sent from the server

[0978] The main hardware and software used are:

[0979] Hardware: Smart glasses (camera, microphone, display)

[0980] Software: Photo and transmission application, alert display application

[0981] User Roles and Actions

[0982] Users have two main roles: Driver and Operations Manager:

[0983] Driver: Wears smart glasses, takes a facial image and sends it to the server. This process is automatic while driving.

[0984] Operation manager: Monitors alert information sent from the server and makes changes to operation plans or arranges for substitute drivers as necessary.

[0985] Specific examples

[0986] For example, driver A puts on his smart glasses and begins driving. The smart glasses periodically take images of A's face and send the data to a server in real time. The server uses a generative AI model to evaluate A's fatigue level, and if the evaluation result exceeds the standard value, it issues a voice alert to A's smart glasses saying, "Your fatigue level is high. Please take a break." A similar alert is also sent to the operations manager, who checks the situation and takes action as necessary.

[0987] Examples of prompt statements

[0988] "Rate the fatigue level of the person in this image on a scale of 0-100."

[0989] The above is the details of each part in the embodiment of the present invention.

[0990] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0991] Step 1:

[0992] The driver wears smart glasses, and the camera in the smart glasses periodically captures facial images.

[0993] Input: Camera image

[0994] Data processing: Smart glasses camera captures driver's face

[0995] Output: Face image data

[0996] Step 2:

[0997] The captured facial image data is temporarily stored in the internal memory of the smart glasses.

[0998] Input: Facial image data

[0999] Data processing: Temporary storage of data

[1000] Output: Facial image data stored in internal memory

[1001] Step 3:

[1002] Facial image data is transmitted to a server in real time.

[1003] Input: Facial image data stored in the internal memory

[1004] Data processing: Sending image data

[1005] Output: Facial image data arriving at the server

[1006] Step 4:

[1007] The server receives the facial image data and temporarily stores it in memory.

[1008] Input: Facial image data that arrives at the server

[1009] Data processing: storing in memory

[1010] Output: Temporarily stored face image data

[1011] Step 5:

[1012] The server stores the facial image data in a database and inputs the stored data into a generative AI model.

[1013] Input: Temporarily stored facial image data

[1014] Data processing: storing in a database and inputting into a generative AI model

[1015] Output: Facial image data input to the generative AI model

[1016] Step 6:

[1017] The generative AI model extracts features from facial image data and assesses fatigue levels based on information such as eye congestion, eyelid movement, and facial color.

[1018] Input: Facial image data

[1019] Data calculation: feature extraction and fatigue evaluation

[1020] Output: Fatigue evaluation results

[1021] Step 7:

[1022] If the evaluation result exceeds the reference value, the server generates alert information and sends the alert to the driver's smart glasses and the operation manager.

[1023] Input: Fatigue evaluation result

[1024] Data Calculation: Alert Generation

[1025] Output: Alert information

[1026] Step 8:

[1027] The driver's smart glasses receive the alert information and display audio and visual alerts.

[1028] Input: Alert information

[1029] Data processing: Alert display

[1030] Output: Audio and visual alerts

[1031] Specific examples

[1032] For example, driver A puts on his smart glasses and steps 1 to 8 are carried out. An example of a prompt sentence at this time is "Please rate the fatigue level of the person in this image on a scale of 0 to 100." This series of steps makes it possible to assess fatigue levels in real time while driving, ensuring safety.

[1033] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1034] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure driving safety. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generation AI on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[1035] Server Roles and Operations

[1036] The server is responsible for the core processing. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion evaluations, but also for training the generative AI.

[1037] The server then inputs the image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the image, such as eye redness, eyelid movement, and facial color, and evaluates fatigue levels on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear.

[1038] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the device and the operation manager.

[1039] Terminal roles and processing

[1040] The terminal is a device used by the driver, and performs the following process: First, the driver uses the terminal to take a picture of his or her face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. The terminal then transmits the captured face image data to a server. Transmission is usually performed via a network.

[1041] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[1042] User Roles and Actions

[1043] Users mainly play two roles: drivers and operation managers. Drivers always use a device to take a facial image before and after work, which provides the necessary data to the server.

[1044] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[1045] Specific example explanation

[1046] For example, driver A takes a facial image on his device before starting work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion. If the evaluation reveals that fatigue level exceeds the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[1047] With the above system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[1048] The processing flow will be explained below.

[1049] Server Processing

[1050] Step 1:

[1051] The server receives the facial image data sent from the terminal. Specifically, it receives the image data sent as an HTTP request and temporarily stores it in memory.

[1052] Step 2:

[1053] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (such as transmission time and driver ID) in the database.

[1054] Step 3:

[1055] The server inputs the saved image data into the generative AI and emotion engine. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI and emotion engine.

[1056] Step 4:

[1057] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[1058] Step 5:

[1059] The server uses an emotion engine to analyze emotions from facial image data, specifically identifying emotions such as joy, anger, sadness, surprise, and fear, and converting them into numerical data.

[1060] Step 6:

[1061] The server then integrates the extracted fatigue level and emotion data to make a comprehensive assessment. For example, if fatigue is high and strong stress or anger is also detected, the server will assess the risk as high.

[1062] Step 7:

[1063] If the evaluation result exceeds the reference value, the server generates an alert message that includes the driver ID and the evaluation result.

[1064] Step 8:

[1065] The server sends the generated alert information to the terminal and the operation manager. Specifically, it sends the alert notification as an HTTP request to the terminal and the manager's device.

[1066] Terminal handling

[1067] Step 1:

[1068] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[1069] Step 2:

[1070] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[1071] Step 3:

[1072] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[1073] Step 4:

[1074] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[1075] Step 5:

[1076] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[1077] User Action

[1078] Step 1:

[1079] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[1080] Step 2:

[1081] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[1082] Step 3:

[1083] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[1084] Step 4:

[1085] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[1086] Example 2

[1087] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1088] Conventional driver health management systems mainly rely on self-reporting or some biometric sensors, making it difficult to accurately assess fatigue levels and emotions. Furthermore, it is difficult to issue warnings at the appropriate time, making it difficult to ensure operational safety. To address this issue, the present invention aims to provide a system that uses facial images of drivers to assess their fatigue levels and emotions and take appropriate countermeasures accurately and quickly.

[1089] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1090] In this invention, the server includes a means for capturing a facial image of the driver, a means for transmitting the captured facial image data to a remote computing device, a means for using artificial intelligence on the computing device to evaluate the fatigue level from the facial image data, and a means for issuing an alarm based on the evaluation result, thereby enabling accurate evaluation of the fatigue level and emotions of the driver based on the facial image and improving driving safety.

[1091] "Operator" means a person responsible for operating a vehicle.

[1092] The "face image" is image data of the driver's face.

[1093] "Capturing means" refers to a camera or other recording device used to capture a facial image.

[1094] The "transmitting means" refers to a communication device or software for transferring captured facial image data to a remote computing device.

[1095] A "server" is a type of computing device that is a data processing device that receives, stores, and evaluates facial image data.

[1096] "Computing device" refers to a device that contains hardware and software for processing data.

[1097] "Fatigue level" is an indicator of the driver's fatigue state and is evaluated using a generative AI model.

[1098] "Generative AI" refers to an artificial intelligence model that has been pre-trained to assess fatigue levels from facial image data.

[1099] "Evaluation results" are the analysis results of fatigue levels and emotions obtained by the generative AI and emotion engine.

[1100] An "alert" is a warning or notification issued to a driver or operations manager based on the evaluation results.

[1101] "Means for issuing an alert" refers to a method or device for generating an alert based on the evaluation results and notifying the driver or operations manager.

[1102] The "storage device" refers to a data storage device for storing photographed face image data and evaluation results.

[1103] "Means for assessing emotions" refers to software or systems for identifying emotions of drivers from facial image data.

[1104] "Comprehensive assessment" is a process of integrating the results of fatigue and emotional assessment to comprehensively judge the driver's condition.

[1105] "Terminal" refers to an electronic device used by a driver, and is equipped with a camera and communication functions.

[1106] "Operations manager" means a person responsible for managing the driving staff and the operational status of the vehicle.

[1107] "Means of notification" refers to the method or device for communicating an alert to an operator or dispatcher.

[1108] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure safe driving. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generative AI model on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[1109] The server is the core device responsible for the main processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. Database and storage systems such as MySQL and Amazon S3 are used for this purpose. The image data is also used for subsequent fatigue and emotion assessments, as well as for training the generative AI model.

[1110] The server then inputs the received image data into the generation AI and emotion engine. The generation AI uses pre-trained PyTorch and TensorFlow to extract fatigue characteristics from the image (such as bloodshot eyes, eyelid movement, and facial color). At the same time, the emotion engine uses OpenFace and the Microsoft Emotion API to analyze facial expressions from the facial image and identify emotions such as joy, anger, sadness, surprise, and fear. The evaluation results are displayed on a scale of 0 to 100.

[1111] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert using services such as "Twilio" and sends it to the device and the operation manager. This enables a prompt response according to the driver's condition.

[1112] The terminal is a device that is directly used by the driver. First, the terminal uses its camera function to capture a facial image of the driver and temporarily stores it in its internal memory. Then, it sends this captured facial image data to a server via the "HTTP" or "HTTPS" protocol. When an alert is sent from the server, the terminal immediately receives the information and displays the alert to the driver. This display is done through audio or a pop-up.

[1113] Users are mainly divided into drivers and operation managers. Drivers are required to take facial images using a terminal before and after work. This allows the system to always obtain the latest necessary data. Operation managers monitor alert information sent from the server. If an alert occurs, the operation manager reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[1114] As a specific example, driver A takes a facial image on a device before work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion.

[1115] Here are some example prompts to input to a generative AI model:

[1116] "Please rate the level of fatigue of driver A based on the facial image."

[1117] "Please rate the emotions of driver A based on the facial image."

[1118] If the evaluation reveals that fatigue levels exceed the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to Person A's device. At the same time, an alert is also sent to the driver's device, allowing the driver to urge Person A to take a break and arrange for a substitute driver if necessary.

[1119] With this system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[1120] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1121] Step 1:

[1122] The terminal captures a facial image of the driver.

[1123] Specific operation: The device activates the camera and plays a voice prompt saying, "Please face forward." When the driver follows the instructions and takes a facial image, the image data is saved in the device's internal memory.

[1124] Input: Face image captured from the camera

[1125] Output: Facial image data stored in internal memory

[1126] Step 2:

[1127] The terminal transmits the facial image data to the server.

[1128] Specific operation: The device makes an HTTP request and sends the captured face image data to the server's API endpoint. If the request is successfully received, the device displays a notification saying "Image sent successfully."

[1129] Input: Facial image data stored in the internal memory

[1130] Output: Sending face image data via HTTP request to the server

[1131] Step 3:

[1132] The server receives and stores the facial image data.

[1133] Specific operation: When the server receives the facial image data, it records a "reception completed" log and saves the data to storage such as Amazon S3. If the save is successful, it updates the status to "save completed."

[1134] Input: Facial image data sent from the device

[1135] Output: Facial image data stored in storage (e.g. Amazon S3)

[1136] Step 4:

[1137] The server inputs the facial image data into the generation AI and emotion engine.

[1138] How it works: The server reads the stored facial image data and inputs it into a generative AI model using PyTorch or TensorFlow, and also into an emotion engine using OpenFace or the Microsoft Emotion API.

[1139] Input: Facial image data stored in storage (e.g. Amazon S3)

[1140] Output: Input data to the generative AI model and emotion engine

[1141] Step 5:

[1142] The server evaluates fatigue and emotion.

[1143] How it works: The server uses a generative AI model to assess fatigue levels and an emotion engine to identify the driver's emotions. The assessment results are displayed on a scale of 0 to 100.

[1144] Input: Input data to the generative AI model and emotion engine

[1145] Output: Fatigue rating score and emotion rating score

[1146] Step 6:

[1147] The server aggregates the evaluation results and generates alerts.

[1148] Specific operation: The server integrates the fatigue and emotion evaluation scores, and if the threshold is exceeded, generates an alert with the message "Caution: High fatigue, unstable emotions." After generating the alert, the status is updated to "Alert generation completed."

[1149] Input: Fatigue rating score and emotion rating score

[1150] Output: Alert information for drivers and operation managers

[1151] Step 7:

[1152] The server sends the alert information to the terminal and the operation manager.

[1153] Specific operation: The server uses "Twilio" to send the alert information to the driver's device and the operation manager. If the transmission is successful, the status is updated to "Successful transmission."

[1154] Input: Generated alert information

[1155] Output: Alert information sent to drivers and dispatchers

[1156] Step 8:

[1157] The device displays an alert.

[1158] Specific operation: The device notifies the driver of the received alert information with a voice notification and a pop-up. When the notification is displayed, the device updates the status to "Alert display completed."

[1159] Input: Alert information sent from the server

[1160] Output: Alert display to the driver

[1161] Step 9:

[1162] The user (operation manager) implements countermeasures.

[1163] Specific actions: The operations manager checks the alert, promptly urges the driver to take a break, and arranges for a substitute driver if necessary. After completing the countermeasures, the status is updated to "Response Completed."

[1164] Input: Alert information sent to drivers and dispatchers

[1165] Output: Implementing countermeasures such as changing the operation plan or arranging for a substitute driver

[1166] (Application example 2)

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

[1168] The challenge is to provide a system that can reliably monitor the fatigue level and emotional state of drivers and improve safety. In particular, previous systems only issued alerts to the driver, and cooperation with the operation manager and the autonomous driving system was insufficient. Furthermore, unless the alert was displayed to the driver quickly and visually, there were limitations to improving safety.

[1169] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1170] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server via a communication device, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for displaying the alert information on the driver's visualization display device, and means for notifying the driving system based on the evaluation result and adjusting the control of automated driving. This enables real-time monitoring of the driver's fatigue level and emotional state, enabling quick responses to improve safety.

[1171] "Driver" means a person engaged in the business of driving a vehicle.

[1172] A "face image" is image data that captures facial features.

[1173] "Communication device" is a general term for hardware or software for transmitting and receiving data.

[1174] A "server" is a computer system that stores, processes, and distributes data over a network.

[1175] "Generative AI" is a form of artificial intelligence that uses pre-trained models to extract, generate, or classify specific information from data.

[1176] "Fatigue level" is an index that quantitatively indicates the degree of fatigue of a person.

[1177] An "alert" is a notification that warns of an abnormal condition or alerts you to something.

[1178] A "visual display device" is a device for visually displaying information to a user.

[1179] "Driving system" is a general term for hardware and software used to manage and control the operation of a vehicle.

[1180] "Evaluation results" are judgments made based on data analysis.

[1181] A "database" is a system for systematically storing, managing, and searching data.

[1182] The present invention relates to a system for evaluating the fatigue level and emotions of a driver and supporting safe driving. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS ... will be specifically described below.

[1183] Server Roles and Operations

[1184] The server plays a central role in processing and analyzing the main data. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion assessments, but also for training the generative AI.

[1185] The server inputs image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the facial image, such as eye redness, eyelid movement, and facial color, and evaluates the fatigue level on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear. Specifically, it is recommended to use machine learning frameworks such as TensorFlow and PyTorch.

[1186] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the driver's visualization display device and the operation manager. Furthermore, the server notifies the driving system based on the evaluation result and adjusts the autonomous driving control.

[1187] Terminal roles and processing

[1188] The terminal is a device used by the driver and performs the following process: First, the driver takes a picture of his / her face using the camera on the terminal. The terminal temporarily stores the captured facial image data in its internal memory and then transmits it to a server via the network.

[1189] When an alert is sent from the server, the terminal immediately receives it and displays it on the driver's visualization display device, prompting the driver to reconsider safety checks.

[1190] User Roles and Actions

[1191] Users mainly play the roles of drivers and operation managers. Drivers always use the device to take facial images before and after work, which provides the necessary data to the server. Operation managers monitor alert information sent from the server, and when an alert occurs, they reconfirm the health status of drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[1192] Specific example explanation

[1193] For example, when a driver wears smart glasses, the built-in camera automatically captures facial images at regular intervals. The image data is sent to a server in real time, and the server uses a generative AI model to evaluate the driver's fatigue level and emotions. If the driver's fatigue level exceeds the threshold, an alert will appear on the smart glasses' display, prompting the driver to take a break. At the same time, the autonomous driving system will be notified, improving driving safety.

[1194] An example prompt might be "input_image: captured_image.jpg, evaluate: [fatigue_level, emotional_state]", which inputs a facial image into a generative AI model and evaluates fatigue level and emotion.

[1195] As described above, the system of the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby realizing safe driving. This system contributes to reducing accidents and improving the working environment.

[1196] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1197] Step 1:

[1198] The terminal takes a facial image of the driver. Specifically, it activates the built-in camera of the terminal and captures the driver's face. The camera is operated as input, and facial image data (image file) is obtained as output.

[1199] Step 2:

[1200] The device temporarily stores the captured facial image data in its internal memory. It receives facial image data as input and stores the data in its memory as output.

[1201] Step 3:

[1202] The device sends the facial image data to the server. It sends an HTTP POST request to the specified URL on the server via the network and uploads the facial image data. The input is the facial image data, and the output is a success response to the server that the data was sent.

[1203] Step 4:

[1204] The server inputs the received facial image data into a generative AI model. Specifically, the server uses machine learning frameworks such as TensorFlow and PyTorch to input data into a pre-trained model. The input is facial image data, and the output is the extracted features (eye redness, eyelid movement, facial color, etc.).

[1205] Step 5:

[1206] The server evaluates the driver's fatigue level based on the extracted features. The evaluation scale is 0 to 100, and the generation AI performs the calculations. The input is the extracted features, and the output is an evaluation score of the fatigue level.

[1207] Step 6:

[1208] The server inputs facial image data into the emotion engine, which identifies emotions such as joy, anger, sadness, surprise, and fear from the image. The input is the facial image data, and the output is the identified emotion data.

[1209] Step 7:

[1210] The server integrates the fatigue evaluation score and emotional data to perform a comprehensive evaluation. Based on this, it generates an alert if the threshold is exceeded. The input is the fatigue evaluation score and emotional data, and the output is alert information.

[1211] Step 8:

[1212] The server sends the generated alert information to the terminal. The alert is displayed on the visualization display device of the driver. The input is the alert information, and the output is the alert displayed on the terminal.

[1213] Step 9:

[1214] The server also sends an alert to the dispatcher based on the evaluation results. The notification from the server is sent to the dispatcher's terminal in a remote location, and the manager monitors it. The input is the alert information, and the output is the alert display on the dispatcher's terminal.

[1215] Step 10:

[1216] The server notifies the driving system based on the evaluation results and adjusts the autonomous driving control. Specifically, it changes the parameters of the autonomous driving system based on the evaluation results to ensure safe driving. The input is the evaluation results, and the output is the adjusted autonomous driving control parameters.

[1217] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1218] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1219] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1220] [Fourth embodiment]

[1221] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1222] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1223] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1224] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1225] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1226] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1227] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1228] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1229] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1230] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1231] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1232] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1233] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1234] The present invention relates to a system for assessing driver fatigue levels and ensuring operational safety. This system includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the driver's fatigue level using a generation AI on the server, and issuing an alert based on the results.

[1235] Server Roles and Operations

[1236] The server is responsible for the core processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Second, it saves the image data in a database. This saved data is used not only for subsequent fatigue assessments but also for training the generation AI.

[1237] The server then inputs the image data into the generation AI, which uses a pre-trained model to extract features from the image, such as bloodshot eyes, eyelid movement, and facial color. Based on the extracted features, the AI ​​evaluates the driver's fatigue level on a scale of 0 to 100. If the result exceeds the threshold, the server generates an alert and sends it to the device and the operation manager.

[1238] Terminal roles and processing

[1239] The terminal is a device used by the driver, and performs the following process. First, the driver uses the terminal to take a picture of his or her own face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. Next, the terminal sends the captured face image data to a server. This transmission is usually performed via a network.

[1240] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[1241] User Roles and Actions

[1242] Users mainly play two roles: drivers and operation managers. Drivers must take facial images using a terminal before and after work, which provides the necessary data to the server.

[1243] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[1244] Specific example explanation

[1245] For example, driver A takes a picture of his face on his device before starting work. At this time, the device's camera activates and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[1246] By using the above system and process, the present invention can efficiently and objectively evaluate the fatigue level of drivers and improve operational safety. This system contributes to reducing accidents and improving the working environment.

[1247] The processing flow will be explained below.

[1248] Specific processing steps of the program

[1249] Server Processing

[1250] Step 1:

[1251] The server receives the facial image data sent from the terminal. Specifically, the image data sent as an HTTP request is processed by the server's reception process and temporarily stored in memory.

[1252] Step 2:

[1253] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (e.g., transmission time, driver ID) in the database.

[1254] Step 3:

[1255] The server inputs the saved image data into the generative AI. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI model.

[1256] Step 4:

[1257] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[1258] Step 5:

[1259] The server evaluates the fatigue level and checks the results. If the evaluation result exceeds the reference value, an alert is generated.

[1260] Step 6:

[1261] The server sends the generated alert information to the terminal and the operation manager. Specifically, it executes a process to send an alert notification to the terminal and the manager's device.

[1262] Terminal handling

[1263] Step 1:

[1264] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[1265] Step 2:

[1266] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[1267] Step 3:

[1268] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[1269] Step 4:

[1270] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[1271] Step 5:

[1272] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[1273] User Action

[1274] Step 1:

[1275] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[1276] Step 2:

[1277] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[1278] Step 3:

[1279] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[1280] Step 4:

[1281] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[1282] Example 1

[1283] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1284] Driver fatigue is one of the major causes of traffic accidents, and its prevention and management are important. Conventional methods often rely on subjective self-reporting or simple checklists, making it difficult to provide an immediate, objective assessment. This creates a risk that highly fatigued drivers may continue driving. This invention aims to provide a system that uses facial images of drivers to quickly and accurately assess their fatigue level and improve driving safety.

[1285] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1286] In this invention, the server includes means for receiving facial image data and storing it in a database, means for inputting the facial image data into a generative AI model and extracting features such as eye congestion, eyelid movement, and facial color to evaluate fatigue level, and means for generating an alert if the fatigue level exceeds a reference value based on the evaluation result and sending the alert to the terminals of the driver and manager. This enables quick and accurate fatigue level evaluation based on the facial image of the driver, thereby improving driving safety.

[1287] "Driver" means a person who drives a vehicle professionally.

[1288] "Facial Image" refers to a photograph or digital image of a driving personnel's face.

[1289] "Terminal" refers to an electronic device used by a driver that has camera and communication capabilities.

[1290] "Network" refers to a communications infrastructure for transmitting data, including the Internet and mobile communications.

[1291] "Server" refers to a computer system for receiving, processing, and storing data over a network.

[1292] A "database" refers to a system for efficiently storing, retrieving, and managing data.

[1293] A "generative AI model" refers to a mathematical model that is trained to perform a specific task using artificial intelligence techniques.

[1294] "Features" are attributes and information extracted from facial images, including eye congestion, eyelid movement, and facial color.

[1295] "Fatigue level" refers to the degree of fatigue of the driver expressed numerically.

[1296] "Alert" refers to a warning or caution message sent to an operator or manager.

[1297] "Manager" refers to the person responsible for supervising and managing the health of drivers and the safety of operations.

[1298] This invention relates to a system for assessing the fatigue level of drivers and ensuring driving safety. Specifically, it includes a method for capturing a facial image of a driver, transmitting the image data to a server, evaluating the fatigue level using a generation AI on the server, and issuing an alert based on the results.

[1299] System Configuration

[1300] Device Role

[1301] The terminal is an electronic device used by the driver and has the following functions:

[1302] Camera function: Taking a facial image of the driver, for example, using the camera on a smartphone or tablet.

[1303] Communication function: Sends captured facial image data to a server via a network (Internet or mobile communications).

[1304] Display function: Displays alerts received from the server to the driver.

[1305] Server Roles

[1306] The server is responsible for the core processing and has the following functions:

[1307] Data reception and storage function: The server receives the facial image data sent from the device, temporarily stores it in memory, and then saves it in a database. The database used could be, for example, MySQL or PostgreSQL.

[1308] Generative AI model: The server inputs facial image data into a generative AI model (e.g., a model using TensorFlow or PyTorch) to extract features such as eye redness, eyelid movement, and facial color. Based on these features, the server then evaluates the level of fatigue on a scale of 0 to 100.

[1309] Alert generation and notification: If the evaluation result exceeds the reference value, the server generates alert information and sends it to the terminals of the driver and manager.

[1310] User Roles

[1311] Drivers use a device to take a picture of their face and send it to the server before and after work, providing the server with the necessary data.

[1312] Administrators monitor the alert information sent from the server, and when an alert occurs, they recheck the health status of the drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[1313] Specific examples

[1314] For example, driver A takes a facial image on his device before work. The device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into a generative AI model to evaluate the level of fatigue. If the evaluation determines that the level of fatigue exceeds the standard value, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the manager's device, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[1315] Prompt Sentence Examples

[1316] Below is an example of a prompt for configuring the generative AI model used in this system.

[1317] We would like to develop an AI system that uses facial images of drivers to assess their fatigue level. We would like you to create an AI model that extracts eye congestion, eyelid movement, and facial color from the images and evaluates fatigue level on a scale of 0 to 100.

[1318] This system and process will enable efficient and objective evaluation of driver fatigue levels and improve operational safety, contributing to reducing accidents and improving the working environment.

[1319] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1320] Step 1: Take an image (device processing)

[1321] Specific explanation: The driver activates the device's camera to take a facial image. The device uses the camera function to capture the image and temporarily stores it in its internal memory. At this time, the device checks that the image resolution, brightness, etc. are appropriate.

[1322] Input and output: The input is the driver's face, and the output is the captured facial image data.

[1323] Specific operation: The driver launches the device's camera app, positions their face on the screen, and presses the shutter button.

[1324] Step 2: Data transmission (terminal processing)

[1325] Specifically, the device sends the captured facial image data to the server via the network. This transmission is performed as encrypted data to ensure the security of the communication.

[1326] Input and output: The input is the captured facial image data, and the output is the data sent to the server.

[1327] Specific operation: After taking a photo, the device automatically encrypts the image data and sends a POST request to the server via the Internet.

[1328] Step 3: Data reception and storage (server processing)

[1329] Specific explanation: The server receives the facial image data sent from the device and temporarily stores it in memory. This image data is then saved in a database. The saved data is used for fatigue evaluation and as learning data for the generation AI.

[1330] Input and output: The input is the facial image data sent from the terminal, and the output is the data stored in the database.

[1331] Specific operation: The server decodes the received image data, stores it in temporary memory, and then stores it in the "face image" table of the database.

[1332] Step 4: Feature extraction and fatigue evaluation (server processing)

[1333] Specifically, the server inputs facial image data obtained from a database into a generative AI model. The generative AI model extracts features from the image, such as bloodshot eyes, eyelid movement, and facial color. Then, based on the extracted features, the AI ​​evaluates the fatigue level on a scale of 0 to 100.

[1334] Input and output: The input is face image data obtained from the database, and the output is the fatigue score.

[1335] Specific operation: The server inputs image data into the TensorFlow model, and evaluates and integrates the eye bloodshot score, eyelid open / close score, and facial color score.

[1336] Step 5: Alert generation and notification (server processing)

[1337] Specific explanation: If the evaluation result exceeds the reference value, the server generates an alert information, which is sent to the driver's terminal and the manager's terminal.

[1338] Inputs and Outputs: The input is the fatigue score, and the output is the alert sent to the driver and manager.

[1339] Specific operation: When the evaluation score exceeds a threshold value (e.g., 80), the server generates an alert message and sends a push notification to the devices of the driver and manager.

[1340] Step 6: Receiving and displaying alerts (terminal processing)

[1341] Specific explanation: The device will receive the alert message as soon as it is sent and display an alert to the driver, which will include specific instructions such as "take a break."

[1342] Input and Output: The input is the alert message sent from the server, and the output is the alert displayed to the driver.

[1343] Specific operation: The driver's device receives a push notification and displays the message, "You are highly fatigued. Please take a break."

[1344] Step 7: Check the alert and respond (user action)

[1345] Specific explanation: Users (drivers and managers) ensure that drivers check the alerts on their devices and take breaks if necessary. Managers recheck the health status of drivers, and if necessary, change the operation plan and arrange for a substitute driver.

[1346] Input and output: Input is an alert message sent from a terminal or server, and output is a response such as a change in the operation plan or a driver's break.

[1347] Specific behavior:

[1348] The driver checks the alert, stops the vehicle in a safe place, and takes a break.

[1349] The administrator checks the driver's status on the system's management screen and, if necessary, contacts other drivers to arrange a replacement.

[1350] (Application example 1)

[1351] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1352] Conventional driver fatigue assessment systems have difficulty assessing driver fatigue in real time, making it difficult to issue alerts in a timely manner. Furthermore, fatigue assessments cannot be performed while driving, which in turn makes it difficult to ensure safety. Therefore, there is a need for a system that can assess driver fatigue in real time while driving and immediately notify drivers of alerts.

[1353] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1354] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for periodically capturing a facial image of the driver using smart glasses and transmitting the captured facial image data to the server in real time, and means for displaying an audio or visual alert on the smart glasses based on the evaluation result. This makes it possible to evaluate the fatigue level in real time even while driving and to immediately notify the driver of the alert.

[1355] "Operator" means an individual responsible for operating or driving a vehicle.

[1356] "Facial image" refers to video data showing the driver's face.

[1357] "Server" refers to the computer system that receives and stores facial image data and analyzes the data using a generative AI model.

[1358] "Fatigue level" is an index that indicates the degree of physical and mental fatigue of drivers.

[1359] "Generative AI" is an artificial intelligence model that learns from large amounts of data and uses the results to analyze and predict new data.

[1360] An "alert" is a warning message that is issued to call attention when a set reference value is exceeded.

[1361] "Smart glasses" are eyeglass-type devices that display digital information and connect to the Internet and applications.

[1362] "Real-time" is a term that refers to processing or transmission of information immediately, without delay.

[1363] "Voice alert" refers to a function that issues warnings or notifications by voice.

[1364] "Visual alert" refers to a function that provides warnings or notifications through visual means such as a display or LED.

[1365] The present invention relates to a system that acquires facial images of drivers through smart glasses, transmits the images to a server in real time, evaluates their fatigue level using a generative AI model, and issues audio and visual alerts.

[1366] System configuration

[1367] Server Roles and Operations

[1368] The server receives, stores, and generates AI to evaluate the level of fatigue and generates an alert if the level exceeds the threshold. The main hardware and software used are as follows:

[1369] Hardware: High-performance server machine

[1370] Software: Python, generative AI models, database systems

[1371] The server first receives the facial image data sent from the smart glasses and temporarily stores it in a database. It then inputs the image data into a generative AI model to extract features and evaluate the level of fatigue based on information such as bloodshot eyes, eyelid movement, and facial color. If the evaluation result exceeds a threshold, an alert is generated and sent to the driver's smart glasses and the operations manager.

[1372] The role and processing of smart glasses

[1373] The smart glasses used by drivers will:

[1374] Photographing: Take regular photographs of your face using a camera

[1375] Send: Send the captured facial image data to the server in real time

[1376] Alert Indication: Audio and visual indication of alerts sent from the server

[1377] The main hardware and software used are:

[1378] Hardware: Smart glasses (camera, microphone, display)

[1379] Software: Photo and transmission application, alert display application

[1380] User Roles and Actions

[1381] Users have two main roles: Driver and Operations Manager:

[1382] Driver: Wears smart glasses, takes a facial image and sends it to the server. This process is automatic while driving.

[1383] Operation manager: Monitors alert information sent from the server and makes changes to operation plans or arranges for substitute drivers as necessary.

[1384] Specific examples

[1385] For example, driver A puts on his smart glasses and begins driving. The smart glasses periodically take images of A's face and send the data to a server in real time. The server uses a generative AI model to evaluate A's fatigue level, and if the evaluation result exceeds the standard value, it issues a voice alert to A's smart glasses saying, "Your fatigue level is high. Please take a break." A similar alert is also sent to the operations manager, who checks the situation and takes action as necessary.

[1386] Examples of prompt statements

[1387] "Rate the fatigue level of the person in this image on a scale of 0-100."

[1388] The above is the details of each part in the embodiment of the present invention.

[1389] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1390] Step 1:

[1391] The driver wears smart glasses, and the camera in the smart glasses periodically captures facial images.

[1392] Input: Camera image

[1393] Data processing: Smart glasses camera captures driver's face

[1394] Output: Face image data

[1395] Step 2:

[1396] The captured facial image data is temporarily stored in the internal memory of the smart glasses.

[1397] Input: Facial image data

[1398] Data processing: Temporary storage of data

[1399] Output: Facial image data stored in internal memory

[1400] Step 3:

[1401] Facial image data is transmitted to a server in real time.

[1402] Input: Facial image data stored in the internal memory

[1403] Data processing: Sending image data

[1404] Output: Facial image data arriving at the server

[1405] Step 4:

[1406] The server receives the facial image data and temporarily stores it in memory.

[1407] Input: Facial image data that arrives at the server

[1408] Data processing: storing in memory

[1409] Output: Temporarily stored face image data

[1410] Step 5:

[1411] The server stores the facial image data in a database and inputs the stored data into a generative AI model.

[1412] Input: Temporarily stored facial image data

[1413] Data processing: storing in a database and inputting into a generative AI model

[1414] Output: Facial image data input to the generative AI model

[1415] Step 6:

[1416] The generative AI model extracts features from facial image data and assesses fatigue levels based on information such as eye congestion, eyelid movement, and facial color.

[1417] Input: Facial image data

[1418] Data calculation: feature extraction and fatigue evaluation

[1419] Output: Fatigue evaluation results

[1420] Step 7:

[1421] If the evaluation result exceeds the reference value, the server generates alert information and sends the alert to the driver's smart glasses and the operation manager.

[1422] Input: Fatigue evaluation result

[1423] Data Calculation: Alert Generation

[1424] Output: Alert information

[1425] Step 8:

[1426] The driver's smart glasses receive the alert information and display audio and visual alerts.

[1427] Input: Alert information

[1428] Data processing: Alert display

[1429] Output: Audio and visual alerts

[1430] Specific examples

[1431] For example, driver A puts on his smart glasses and steps 1 to 8 are carried out. An example of a prompt sentence at this time is "Please rate the fatigue level of the person in this image on a scale of 0 to 100." This series of steps makes it possible to assess fatigue levels in real time while driving, ensuring safety.

[1432] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1433] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure driving safety. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generation AI on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[1434] Server Roles and Operations

[1435] The server is responsible for the core processing. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion evaluations, but also for training the generative AI.

[1436] The server then inputs the image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the image, such as eye redness, eyelid movement, and facial color, and evaluates fatigue levels on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear.

[1437] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the device and the operation manager.

[1438] Terminal roles and processing

[1439] The terminal is a device used by the driver, and performs the following process: First, the driver uses the terminal to take a picture of his or her face. The terminal uses its camera function to acquire the face image and temporarily stores it in its internal memory. The terminal then transmits the captured face image data to a server. Transmission is usually performed via a network.

[1440] When an alert is sent from the server, the device immediately receives it and displays it to the driver, prompting them to reconsider the safety of their driving.

[1441] User Roles and Actions

[1442] Users mainly play two roles: drivers and operation managers. Drivers always use a device to take a facial image before and after work, which provides the necessary data to the server.

[1443] The dispatcher monitors the alert information sent from the server. When an alert occurs, the dispatcher reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[1444] Specific example explanation

[1445] For example, driver A takes a facial image on his device before starting work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion. If the evaluation reveals that fatigue level exceeds the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to A's device. At the same time, an alert is also sent to the device of the operations manager, so the manager can urge A to take a break and arrange for a substitute driver if necessary.

[1446] With the above system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[1447] The processing flow will be explained below.

[1448] Server Processing

[1449] Step 1:

[1450] The server receives the facial image data sent from the terminal. Specifically, it receives the image data sent as an HTTP request and temporarily stores it in memory.

[1451] Step 2:

[1452] The server stores the received image data in a database by executing an SQL query to store the image data and associated metadata (such as transmission time and driver ID) in the database.

[1453] Step 3:

[1454] The server inputs the saved image data into the generative AI and emotion engine. Specifically, it reads the saved data from the disk, rearranges it in memory, and converts it into the input format for the generative AI and emotion engine.

[1455] Step 4:

[1456] The server uses generative AI to extract features from facial image data, such as bloodshot eyes, eyelid movement, and facial color, and calculates the fatigue level on a scale of 0 to 100.

[1457] Step 5:

[1458] The server uses an emotion engine to analyze emotions from facial image data, specifically identifying emotions such as joy, anger, sadness, surprise, and fear, and converting them into numerical data.

[1459] Step 6:

[1460] The server then integrates the extracted fatigue level and emotion data to make a comprehensive assessment. For example, if fatigue is high and strong stress or anger is also detected, the server will assess the risk as high.

[1461] Step 7:

[1462] If the evaluation result exceeds the reference value, the server generates an alert message that includes the driver ID and the evaluation result.

[1463] Step 8:

[1464] The server sends the generated alert information to the terminal and the operation manager. Specifically, it sends the alert notification as an HTTP request to the terminal and the manager's device.

[1465] Terminal handling

[1466] Step 1:

[1467] The user acts as a driver and points their face towards the camera. The driver then starts up the device and opens a dedicated application.

[1468] Step 2:

[1469] The device activates the camera and captures a facial image. Specifically, the application controls the camera function, captures a facial image, and temporarily stores it in the internal memory.

[1470] Step 3:

[1471] The device sends the captured facial image data to the server. Specifically, the device attaches the image data to an HTTP request and executes a process to send it to the specified endpoint on the server.

[1472] Step 4:

[1473] Receives alert information from the server. Specifically, it periodically communicates with the server to check for alert information, and immediately receives any new alerts.

[1474] Step 5:

[1475] The device displays the received alert information to the user. Specifically, it notifies the user by using a pop-up message in the application or an alert sound.

[1476] User Action

[1477] Step 1:

[1478] Drivers must take facial images on the device before and after work. Specifically, they operate the device's application to perform the process of taking facial images.

[1479] Step 2:

[1480] Drivers check the alert information through their terminals, specifically by viewing and listening to the warning messages and audio notifications displayed on the terminals.

[1481] Step 3:

[1482] The operations manager monitors the alert information sent from the server, and specifically, checks the alert information in real time through a dedicated management application and reconfirms the health status of the drivers.

[1483] Step 4:

[1484] The dispatcher will change the operation plan or arrange for a substitute driver as necessary. Specifically, they will adjust the operation schedule based on the alert information and execute the process of arranging breaks and alternative drivers.

[1485] Example 2

[1486] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1487] Conventional driver health management systems mainly rely on self-reporting or some biometric sensors, making it difficult to accurately assess fatigue levels and emotions. Furthermore, it is difficult to issue warnings at the appropriate time, making it difficult to ensure operational safety. To address this issue, the present invention aims to provide a system that uses facial images of drivers to assess their fatigue levels and emotions and take appropriate countermeasures accurately and quickly.

[1488] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1489] In this invention, the server includes a means for capturing a facial image of the driver, a means for transmitting the captured facial image data to a remote computing device, a means for using artificial intelligence on the computing device to evaluate the fatigue level from the facial image data, and a means for issuing an alarm based on the evaluation result, thereby enabling accurate evaluation of the fatigue level and emotions of the driver based on the facial image and improving driving safety.

[1490] "Operator" means a person responsible for operating a vehicle.

[1491] The "face image" is image data of the driver's face.

[1492] "Capturing means" refers to a camera or other recording device used to capture a facial image.

[1493] The "transmitting means" refers to a communication device or software for transferring captured facial image data to a remote computing device.

[1494] A "server" is a type of computing device that is a data processing device that receives, stores, and evaluates facial image data.

[1495] "Computing device" refers to a device that contains hardware and software for processing data.

[1496] "Fatigue level" is an indicator of the driver's fatigue state and is evaluated using a generative AI model.

[1497] "Generative AI" refers to an artificial intelligence model that has been pre-trained to assess fatigue levels from facial image data.

[1498] "Evaluation results" are the analysis results of fatigue levels and emotions obtained by the generative AI and emotion engine.

[1499] An "alert" is a warning or notification issued to a driver or operations manager based on the evaluation results.

[1500] "Means for issuing an alert" refers to a method or device for generating an alert based on the evaluation results and notifying the driver or operations manager.

[1501] The "storage device" refers to a data storage device for storing photographed face image data and evaluation results.

[1502] "Means for assessing emotions" refers to software or systems for identifying emotions of drivers from facial image data.

[1503] "Comprehensive assessment" is a process of integrating the results of fatigue and emotional assessment to comprehensively judge the driver's condition.

[1504] "Terminal" refers to an electronic device used by a driver, and is equipped with a camera and communication functions.

[1505] "Operations manager" means a person responsible for managing the driving staff and the operational status of the vehicle.

[1506] "Means of notification" refers to the method or device for communicating an alert to an operator or dispatcher.

[1507] This invention relates to a system for assessing the fatigue level and emotions of drivers to ensure safe driving. This system involves taking a facial image of the driver, sending the image data to a server, evaluating the fatigue level and emotions using a generative AI model on the server, and issuing an alert based on the results. In addition, by combining it with an emotion engine, a more comprehensive evaluation is possible.

[1508] The server is the core device responsible for the main processing. First, it receives the facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. Database and storage systems such as MySQL and Amazon S3 are used for this purpose. The image data is also used for subsequent fatigue and emotion assessments, as well as for training the generative AI model.

[1509] The server then inputs the received image data into the generation AI and emotion engine. The generation AI uses pre-trained PyTorch and TensorFlow to extract fatigue characteristics from the image (such as bloodshot eyes, eyelid movement, and facial color). At the same time, the emotion engine uses OpenFace and the Microsoft Emotion API to analyze facial expressions from the facial image and identify emotions such as joy, anger, sadness, surprise, and fear. The evaluation results are displayed on a scale of 0 to 100.

[1510] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert using services such as "Twilio" and sends it to the device and the operation manager. This enables a prompt response according to the driver's condition.

[1511] The terminal is a device that is directly used by the driver. First, the terminal uses its camera function to capture a facial image of the driver and temporarily stores it in its internal memory. Then, it sends this captured facial image data to a server via the "HTTP" or "HTTPS" protocol. When an alert is sent from the server, the terminal immediately receives the information and displays the alert to the driver. This display is done through audio or a pop-up.

[1512] Users are mainly divided into drivers and operation managers. Drivers are required to take facial images using a terminal before and after work. This allows the system to always obtain the latest necessary data. Operation managers monitor alert information sent from the server. If an alert occurs, the operation manager reconfirms the health status of the drivers and, if necessary, changes the operation plan or arranges for a substitute driver.

[1513] As a specific example, driver A takes a facial image on a device before work. At this time, the device activates the camera and sends the image data of A's face to the server. The server receives this data and inputs it into the generation AI and emotion engine to evaluate fatigue level and emotion.

[1514] Here are some example prompts to input to a generative AI model:

[1515] "Please rate the level of fatigue of driver A based on the facial image."

[1516] "Please rate the emotions of driver A based on the facial image."

[1517] If the evaluation reveals that fatigue levels exceed the standard value and the emotion engine detects strong stress or anger, the server immediately generates an alert and sends it to Person A's device. At the same time, an alert is also sent to the driver's device, allowing the driver to urge Person A to take a break and arrange for a substitute driver if necessary.

[1518] With this system and process, the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby improving driving safety. This system contributes to reducing accidents and improving the working environment.

[1519] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1520] Step 1:

[1521] The terminal captures a facial image of the driver.

[1522] Specific operation: The device activates the camera and plays a voice prompt saying, "Please face forward." When the driver follows the instructions and takes a facial image, the image data is saved in the device's internal memory.

[1523] Input: Face image captured from the camera

[1524] Output: Facial image data stored in internal memory

[1525] Step 2:

[1526] The terminal transmits the facial image data to the server.

[1527] Specific operation: The device makes an HTTP request and sends the captured face image data to the server's API endpoint. If the request is successfully received, the device displays a notification saying "Image sent successfully."

[1528] Input: Facial image data stored in the internal memory

[1529] Output: Sending face image data via HTTP request to the server

[1530] Step 3:

[1531] The server receives and stores the facial image data.

[1532] Specific operation: When the server receives the facial image data, it records a "reception completed" log and saves the data to storage such as Amazon S3. If the save is successful, it updates the status to "save completed."

[1533] Input: Facial image data sent from the device

[1534] Output: Facial image data stored in storage (e.g. Amazon S3)

[1535] Step 4:

[1536] The server inputs the facial image data into the generation AI and emotion engine.

[1537] How it works: The server reads the stored facial image data and inputs it into a generative AI model using PyTorch or TensorFlow, and also into an emotion engine using OpenFace or the Microsoft Emotion API.

[1538] Input: Facial image data stored in storage (e.g. Amazon S3)

[1539] Output: Input data to the generative AI model and emotion engine

[1540] Step 5:

[1541] The server evaluates fatigue and emotion.

[1542] How it works: The server uses a generative AI model to assess fatigue levels and an emotion engine to identify the driver's emotions. The assessment results are displayed on a scale of 0 to 100.

[1543] Input: Input data to the generative AI model and emotion engine

[1544] Output: Fatigue rating score and emotion rating score

[1545] Step 6:

[1546] The server aggregates the evaluation results and generates alerts.

[1547] Specific operation: The server integrates the fatigue and emotion evaluation scores, and if the threshold is exceeded, generates an alert with the message "Caution: High fatigue, unstable emotions." After generating the alert, the status is updated to "Alert generation completed."

[1548] Input: Fatigue rating score and emotion rating score

[1549] Output: Alert information for drivers and operation managers

[1550] Step 7:

[1551] The server sends the alert information to the terminal and the operation manager.

[1552] Specific operation: The server uses "Twilio" to send the alert information to the driver's device and the operation manager. If the transmission is successful, the status is updated to "Successful transmission."

[1553] Input: Generated alert information

[1554] Output: Alert information sent to drivers and dispatchers

[1555] Step 8:

[1556] The device displays an alert.

[1557] Specific operation: The device notifies the driver of the received alert information with a voice notification and a pop-up. When the notification is displayed, the device updates the status to "Alert display completed."

[1558] Input: Alert information sent from the server

[1559] Output: Alert display to the driver

[1560] Step 9:

[1561] The user (operation manager) implements countermeasures.

[1562] Specific actions: The operations manager checks the alert, promptly urges the driver to take a break, and arranges for a substitute driver if necessary. After completing the countermeasures, the status is updated to "Response Completed."

[1563] Input: Alert information sent to drivers and dispatchers

[1564] Output: Implementing countermeasures such as changing the operation plan or arranging for a substitute driver

[1565] (Application example 2)

[1566] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1567] The challenge is to provide a system that can reliably monitor the fatigue level and emotional state of drivers and improve safety. In particular, previous systems only issued alerts to the driver, and cooperation with the operation manager and the autonomous driving system was insufficient. Furthermore, unless the alert was displayed to the driver quickly and visually, there were limitations to improving safety.

[1568] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1569] In this invention, the server includes means for capturing a facial image of the driver, means for transmitting the captured facial image data to the server via a communication device, means for using a generation AI on the server to evaluate the fatigue level from the facial image data, means for issuing an alert based on the evaluation result, means for displaying the alert information on the driver's visualization display device, and means for notifying the driving system based on the evaluation result and adjusting the control of automated driving. This enables real-time monitoring of the driver's fatigue level and emotional state, enabling quick responses to improve safety.

[1570] "Driver" means a person engaged in the business of driving a vehicle.

[1571] A "face image" is image data that captures facial features.

[1572] "Communication device" is a general term for hardware or software for transmitting and receiving data.

[1573] A "server" is a computer system that stores, processes, and distributes data over a network.

[1574] "Generative AI" is a form of artificial intelligence that uses pre-trained models to extract, generate, or classify specific information from data.

[1575] "Fatigue level" is an index that quantitatively indicates the degree of fatigue of a person.

[1576] An "alert" is a notification that warns of an abnormal condition or alerts you to something.

[1577] A "visual display device" is a device for visually displaying information to a user.

[1578] "Driving system" is a general term for hardware and software used to manage and control the operation of a vehicle.

[1579] "Evaluation results" are judgments made based on data analysis.

[1580] A "database" is a system for systematically storing, managing, and searching data.

[1581] The present invention relates to a system for evaluating the fatigue level and emotions of a driver and supporting safe driving. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS ... will be specifically described below.

[1582] Server Roles and Operations

[1583] The server plays a central role in processing and analyzing the main data. First, it receives facial image data sent by the driver and temporarily stores it in memory. Then, it saves this image data in a database. The saved data is used not only for subsequent fatigue and emotion assessments, but also for training the generative AI.

[1584] The server inputs image data into the generation AI and emotion engine. The generation AI uses a pre-trained model to extract features from the facial image, such as eye redness, eyelid movement, and facial color, and evaluates the fatigue level on a scale of 0 to 100. At the same time, the emotion engine analyzes facial expressions from the facial image and identifies emotions such as joy, anger, sadness, surprise, and fear. Specifically, it is recommended to use machine learning frameworks such as TensorFlow and PyTorch.

[1585] The server integrates the extracted fatigue level and emotion data to perform a comprehensive evaluation. If the evaluation result exceeds the reference value, the server generates an alert and sends it to the driver's visualization display device and the operation manager. Furthermore, the server notifies the driving system based on the evaluation result and adjusts the autonomous driving control.

[1586] Terminal roles and processing

[1587] The terminal is a device used by the driver and performs the following process: First, the driver takes a picture of his / her face using the camera on the terminal. The terminal temporarily stores the captured facial image data in its internal memory and then transmits it to a server via the network.

[1588] When an alert is sent from the server, the terminal immediately receives it and displays it on the driver's visualization display device, prompting the driver to reconsider safety checks.

[1589] User Roles and Actions

[1590] Users mainly play the roles of drivers and operation managers. Drivers always use the device to take facial images before and after work, which provides the necessary data to the server. Operation managers monitor alert information sent from the server, and when an alert occurs, they reconfirm the health status of drivers and, if necessary, change the operation plan or arrange for a substitute driver.

[1591] Specific example explanation

[1592] For example, when a driver wears smart glasses, the built-in camera automatically captures facial images at regular intervals. The image data is sent to a server in real time, and the server uses a generative AI model to evaluate the driver's fatigue level and emotions. If the driver's fatigue level exceeds the threshold, an alert will appear on the smart glasses' display, prompting the driver to take a break. At the same time, the autonomous driving system will be notified, improving driving safety.

[1593] An example prompt might be "input_image: captured_image.jpg, evaluate: [fatigue_level, emotional_state]", which inputs a facial image into a generative AI model and evaluates fatigue level and emotion.

[1594] As described above, the system of the present invention can efficiently and objectively evaluate the fatigue level and emotions of drivers, thereby realizing safe driving. This system contributes to reducing accidents and improving the working environment.

[1595] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1596] Step 1:

[1597] The terminal takes a facial image of the driver. Specifically, it activates the built-in camera of the terminal and captures the driver's face. The camera is operated as input, and facial image data (image file) is obtained as output.

[1598] Step 2:

[1599] The device temporarily stores the captured facial image data in its internal memory. It receives facial image data as input and stores the data in its memory as output.

[1600] Step 3:

[1601] The device sends the facial image data to the server. It sends an HTTP POST request to the specified URL on the server via the network and uploads the facial image data. The input is the facial image data, and the output is a success response to the server that the data was sent.

[1602] Step 4:

[1603] The server inputs the received facial image data into a generative AI model. Specifically, the server uses machine learning frameworks such as TensorFlow and PyTorch to input data into a pre-trained model. The input is facial image data, and the output is the extracted features (eye redness, eyelid movement, facial color, etc.).

[1604] Step 5:

[1605] The server evaluates the driver's fatigue level based on the extracted features. The evaluation scale is 0 to 100, and the generation AI performs the calculations. The input is the extracted features, and the output is an evaluation score of the fatigue level.

[1606] Step 6:

[1607] The server inputs facial image data into the emotion engine, which identifies emotions such as joy, anger, sadness, surprise, and fear from the image. The input is the facial image data, and the output is the identified emotion data.

[1608] Step 7:

[1609] The server integrates the fatigue evaluation score and emotional data to perform a comprehensive evaluation. Based on this, it generates an alert if the threshold is exceeded. The input is the fatigue evaluation score and emotional data, and the output is alert information.

[1610] Step 8:

[1611] The server sends the generated alert information to the terminal. The alert is displayed on the visualization display device of the driver. The input is the alert information, and the output is the alert displayed on the terminal.

[1612] Step 9:

[1613] The server also sends an alert to the dispatcher based on the evaluation results. The notification from the server is sent to the dispatcher's terminal in a remote location, and the manager monitors it. The input is the alert information, and the output is the alert display on the dispatcher's terminal.

[1614] Step 10:

[1615] The server notifies the driving system based on the evaluation results and adjusts the autonomous driving control. Specifically, it changes the parameters of the autonomous driving system based on the evaluation results to ensure safe driving. The input is the evaluation results, and the output is the adjusted autonomous driving control parameters.

[1616] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1617] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1618] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1619] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1620] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1621] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1622] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1623] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1624] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1625] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1626] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1627] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1628] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1629] 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.

[1630] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1631] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1632] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1633] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1634] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1635] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1636] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1637] The following is further disclosed regarding the above embodiment.

[1638] (Claim 1)

[1639] A means for capturing a facial image of a driver;

[1640] means for transmitting the captured facial image data to a server;

[1641] A means for using a generation AI that evaluates fatigue level from facial image data on a server;

[1642] means for issuing an alert based on the evaluation results;

[1643] A system including:

[1644] (Claim 2)

[1645] 10. The system according to claim 1, further comprising means for sending an alert to the driver and the operation manager when the fatigue level exceeds a reference value based on the evaluation result.

[1646] (Claim 3)

[1647] The system according to claim 1, further comprising means for storing the captured facial image data and the evaluation results in a database and training the generation AI based on the accumulated data.

[1648] "Example 1"

[1649] (Claim 1)

[1650] A device for capturing a facial image of a driver;

[1651] a device for transmitting the captured facial image data to a server via a network;

[1652] A device for receiving face image data at a server and storing the data in a database;

[1653] A device that inputs facial image data into a generative AI model and extracts features such as eye congestion, eyelid movement, and facial color to evaluate fatigue levels.

[1654] a device that generates an alert when the fatigue level exceeds a reference value based on the evaluation result and transmits the alert to the terminals of the driver and the manager;

[1655] a device for displaying the alert on a terminal that receives the alert;

[1656] A system including:

[1657] (Claim 2)

[1658] The system according to claim 1, further comprising a device that sends an alert to a terminal of an operator and a manager when the evaluation result exceeds a reference value, and displays the alert on the terminal.

[1659] (Claim 3)

[1660] The system according to claim 1, further comprising a device that stores the captured facial image data and evaluation results in a database and trains a generative AI model based on the accumulated data.

[1661] "Application Example 1"

[1662] (Claim 1)

[1663] A means for capturing a facial image of a driver;

[1664] means for transmitting the captured facial image data to a server;

[1665] A means for using a generation AI that evaluates fatigue level from facial image data on a server;

[1666] means for issuing an alert based on the evaluation results;

[1667] A means for periodically capturing facial images of a driver using smart glasses and transmitting the images to a server in real time;

[1668] a means for displaying audio and visual alerts on the smart glasses based on the evaluation results;

[1669] A system including:

[1670] (Claim 2)

[1671] 10. The system according to claim 1, further comprising means for sending an alert to the driver and the operation manager when the fatigue level exceeds a reference value based on the evaluation result.

[1672] (Claim 3)

[1673] The system according to claim 1, further comprising means for storing the captured facial image data and the evaluation results in a database and training the generation AI based on the accumulated data.

[1674] "Example 2: Combining Emotion Engines"

[1675] (Claim 1)

[1676] A means for capturing a facial image of a driver;

[1677] means for transmitting the captured facial image data to a remote computing device;

[1678] means for using artificial intelligence on a computing device to assess fatigue level from facial image data;

[1679] means for issuing an alert based on the evaluation result;

[1680] A system including:

[1681] (Claim 2)

[1682] 10. The system according to claim 1, further comprising means for sending an alert to the driver and the operation manager when the fatigue level exceeds a reference value based on the evaluation result.

[1683] (Claim 3)

[1684] 2. The system according to claim 1, further comprising means for storing the captured facial image data and the evaluation results in a storage device, and for performing artificial intelligence learning based on the stored data.

[1685] (Claim 4)

[1686] means for assessing emotion from facial image data on a computing device;

[1687] A means for comprehensively evaluating the driver's condition by integrating the results of the fatigue level and emotion evaluation;

[1688] means for issuing an alarm based on the comprehensive evaluation result;

[1689] 10. The system of claim 1, further comprising:

[1690] (Claim 5)

[1691] 10. The system according to claim 1, further comprising means for notifying a driver's terminal and an operation manager's terminal of the warning.

[1692] (Claim 6)

[1693] 3. The system of claim 2, further comprising means for displaying an alert on the terminal to prompt the driver to take a break.

[1694] "Application example 2 when combining emotion engines"

[1695] (Claim 1)

[1696] A means for capturing a facial image of a driver;

[1697] means for transmitting the captured face image data to a server via a communication device;

[1698] A means for using a generation AI that evaluates fatigue level from facial image data on a server;

[1699] means for issuing an alert based on the evaluation results;

[1700] means for displaying the alert information on a driver's visual display;

[1701] a means for notifying the driving system based on the evaluation results and adjusting the control of the automated driving;

[1702] A system including:

[1703] (Claim 2)

[1704] 10. The system according to claim 1, further comprising means for sending an alert to the driver and the operation manager when the fatigue level exceeds a reference value based on the evaluation result.

[1705] (Claim 3)

[1706] The system according to claim 1, further comprising means for storing the captured facial image data and the evaluation results in a database and training the generation AI based on the accumulated data. [Explanation of symbols]

[1707] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for capturing a facial image of a driver; means for transmitting the captured facial image data to a server; A means for using a generation AI that evaluates fatigue level from facial image data on a server; means for issuing an alert based on the evaluation results; A system including:

2. The system according to claim 1 , further comprising a means for sending an alert to the driver and the operation manager when the fatigue level exceeds a reference value based on the evaluation result.

3. The system according to claim 1, further comprising means for storing the captured facial image data and the evaluation results in a database and for training the generating AI based on the stored data.

Images