system

Abstract

We provide the system. [Solution] A means for acquiring video data from multiple imaging devices arranged in the work area, A means for analyzing the aforementioned video data in real time and recognizing the worker's movements and the operating status of the machinery, A method for evaluating the degree of risk in the work environment based on past cases, A notification system for issuing warnings to workers when a hazard is detected, A means of sending a warning to a mobile device based on the analysis results and issuing a warning by voice or light, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In work areas such as construction sites, there are many operations with high risks. Conventional safety management methods have problems in instantly detecting dangers and quickly responding. Also, operator carelessness and machine operation mistakes may lead to serious accidents, and safety education alone is insufficient. Therefore, there is an increasing need for a system that can detect dangerous situations in real time and immediately send instructions to operators. The burden on on-site supervisors to monitor all situations is also large, and means to reduce this burden are required.

Means for Solving the Problems

[0005] This invention provides a system that acquires video data from multiple imaging devices placed in a work area and performs real-time analysis. This makes it possible to recognize the movements of workers and the operating status of machinery and equipment, and to evaluate the degree of danger in the work environment based on past cases. If a danger is detected, instructions can be immediately sent to workers using a notification means for issuing warnings. Furthermore, in the event of an emergency, the system notifies the site supervisor and sends evacuation instructions to workers, providing an efficient and accurate means of avoiding danger. This improves safety on site and reduces the burden on supervisors.

[0006] An "imaging device" is a piece of equipment that includes cameras and sensors for acquiring video data within a work area.

[0007] "Video data" refers to image and video information collected by imaging devices, and its analysis serves as information material for understanding the actions of workers and the conditions of the environment.

[0008] "Real-time analysis" is a technical process that instantly processes acquired video data and immediately evaluates the situation within the work area.

[0009] A "worker" refers to a person who is active within a specific work area and is subject to monitoring and instructions by the system.

[0010] "Mechanical equipment" refers to heavy machinery, tools, and mechanical devices used within the work area.

[0011] "Risk assessment" is the process of determining safety risks in a work area using numerical values ​​and indicators, based on data obtained through real-time analysis.

[0012] "Notification means" refers to equipment or software functions that notify workers of information via sound, light, or vibration when a hazard is detected.

[0013] A "site supervisor" refers to a person responsible for monitoring safety management and the progress of work within the work area, and giving instructions.

[0014] An "emergency situation" refers to an incident that occurs at a site, such as an accident or an unexpected dangerous situation, requiring immediate action.

[0015] An "evacuation order" is a directional instruction issued by a system to safely evacuate workers from a dangerous situation. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12]It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the language used in the following description will be explained.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] One embodiment of this invention includes a safety management system that uses video data captured by multiple camera devices placed in a work area. A server receives this data in real time and analyzes it using an AI algorithm. The analysis monitors the movements of workers and machinery, and compares it with a database of past hazard cases to evaluate the level of risk in the current work environment.

[0038] When a hazard is detected, the server immediately sends a warning to the terminal via a notification system. This terminal is a portable device with voice output and vibration alarm functions carried by the worker, and it communicates the nature of the hazard and evasive actions through voice and light. For example, if a dangerous heavy machine approaches, a voice command such as "Do not approach the heavy machine" is delivered directly to the worker from the terminal.

[0039] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation route. Notifications are also sent to the site supervisor, who is the user, allowing them to coordinate emergency responses. The server guides workers through terminals with specific evacuation procedures and safe assembly points, supporting a swift and orderly evacuation.

[0040] For example, if a worker continues working at height without their safety harness, the server will detect this situation in real time and issue a warning message saying, "Please put on your safety harness." Furthermore, if a sudden gust of wind occurs and the risk of the crane tipping over increases, the server can notify the worker to temporarily stop work, thereby ensuring the safety of the entire site. This embodiment makes it possible to enhance site safety and further protect the lives and health of workers.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The server receives video data in real time from multiple camera devices placed in the workspace. The received data is converted to a format and resolution specified by the program for analysis.

[0044] Step 2:

[0045] The server uses AI algorithms to analyze received video data and detect worker movements, machine status, and environmental changes. This process utilizes computer vision technology to recognize the position, movement, and status of people and machines.

[0046] Step 3:

[0047] Based on the analysis results, the server compares them with past hazard cases stored in the database to evaluate the level of risk in the current work environment. The level of risk is determined using numerical values ​​and indicators, and if it exceeds the standard value, it is judged to be dangerous.

[0048] Step 4:

[0049] When a hazard is detected, the server immediately uses notification methods to send warning data to the terminal. The terminal then alerts the worker with sound, vibration, or light. For example, it might issue a specific instruction such as, "You are working at height; be sure to wear your safety harness."

[0050] Step 5:

[0051] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation order. It also sends a notification to the site supervisor (the user) and provides information to support the emergency response across the entire site.

[0052] Step 6:

[0053] The server transmits specific evacuation routes and instructions to workers via their terminals. These instructions include emergency action guidelines such as, "Follow the evacuation route and evacuate immediately to the designated safe location."

[0054] Step 7:

[0055] The server records all warnings and response histories in a database, which is used for future security improvements and analysis. Users can then use this data to provide feedback for further security enhancements.

[0056] (Example 1)

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

[0058] Safety management in the work area requires constantly monitoring worker movements and the status of machinery and equipment, and immediately detecting potential hazards to take countermeasures. However, existing safety management systems often lack real-time capabilities and accuracy, posing a challenge in providing appropriate evacuation instructions quickly, especially in emergencies.

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

[0060] In this invention, the server includes means for acquiring information from multiple imaging devices placed in the work area, means for analyzing the information in real time to recognize the actions of workers and the operation of machinery and equipment, and means for evaluating the degree of risk in the work environment based on past cases. This enables real-time risk detection and effective emergency response.

[0061] "Imaging device" is a general term for devices used to monitor a work area and acquire information in real time.

[0062] "Information" refers to all data acquired from imaging devices and other sensors, which the system uses for analysis and evaluation.

[0063] "Real-time analysis" refers to the process of immediately processing acquired information and quickly evaluating the current situation.

[0064] "Worker behavior" refers to all human movements and actions within the work area and is subject to safety management.

[0065] "Operation of machinery and equipment" refers to the state and operating conditions of devices operating within a work area.

[0066] "Past incidents" refer to data on accidents and dangerous situations recorded previously, and are used for risk assessment.

[0067] "Assessing the level of risk" is the process of expressing the risks to the current work environment using numerical values ​​and indicators.

[0068] "Communication means" refers to a device or function used to transmit information to workers or managers.

[0069] An "emergency situation" refers to a situation that indicates the occurrence of a danger or accident requiring immediate action.

[0070] An "evasive route" refers to a safe path for movement during an emergency.

[0071] "Support measures" refer to functions that provide workers with supplementary information and instructions to help them avoid danger and ensure safety.

[0072] In this embodiment of the invention, the server acquires information from multiple imaging devices installed in the work area. The imaging devices cover a wide area and monitor the actions of workers and the operation of machinery from multiple angles. This allows the server to receive information in real time. Next, the server preprocesses this information, removes noise, and then processes it to the required resolution.

[0073] The analysis utilizes AI algorithms such as TENSORFLOW® and PyTorch. The server uses these tools to evaluate the work environment in real time and detect hazards by comparing them with past case data. If a hazard is detected, the server sends a warning to the terminal via communication.

[0074] The terminal is a device carried or worn by the worker, and its role is to convey warnings through voice output or vibration alarms. For example, if a worker is approaching dangerous machinery, it will notify them by voice, "Heavy machinery is approaching. Please maintain a safe distance."

[0075] In the event of an emergency, the server reassesss the risks and determines the optimal evacuation route. This information is communicated to the site supervisor, who is the user, enabling adjustments to the comprehensive evacuation plan. The server also guides workers through terminals with specific evacuation procedures and assembly points, supporting an orderly evacuation.

[0076] For example, if a worker removes a safety device while working at height, the server will immediately recognize this and issue a warning such as, "Please put on the safety device." Furthermore, if weather conditions make crane operation unsafe, the system will ensure safety by ordering a temporary halt to operations.

[0077] Examples of prompt statements to input into a generative AI model include the following:

[0078] "Please explain the hazard detection and real-time warning functions of the safety management system used at the work site."

[0079] "Please explain how to ensure necessary safety measures during work at heights using AI-based analysis."

[0080] Through these elements, this invention aims to efficiently improve safety in the workplace and protect the lives and health of workers.

[0081] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0082] Step 1:

[0083] The server acquires video data in real time from multiple imaging devices installed in the work area. This video data is image information based on settings such as resolution and field of view. The input image data is subjected to noise reduction filtering and conversion to the required resolution, and output as video data suitable for analysis.

[0084] Step 2:

[0085] The server analyzes pre-processed video data using AI algorithms. Specifically, it uses TensorFlow to detect and recognize worker movements and machinery movements. The input is pre-processed video data frame by frame, and the output is identified motion patterns and data on the state of machine operation. Based on this data, it determines whether the operation is normal or if there is a potential hazard.

[0086] Step 3:

[0087] The server compares the analysis results with a database of past hazard cases to assess the risk level of the current work environment. After performing a comparison process using the past case database as input, it scores the current situation based on the output hazard assessment data to determine whether it is a high-risk situation.

[0088] Step 4:

[0089] If a high level of risk is detected, the server sends a warning message to the terminal. The input is risk assessment data, and based on this data, the server generates a warning and sends an appropriate message to the terminal as output. The terminal notifies the worker of this message via voice or vibration, for example, "Heavy machinery is approaching. Please maintain a safe distance."

[0090] Step 5:

[0091] In the event of an emergency, the server will examine the current situation, reassess the risks, and then calculate the optimal evacuation route. The inputs are risk assessment data and real-time environmental data, which are used to formulate an evacuation plan, and the output is a specific evacuation order.

[0092] Step 6:

[0093] The server also sends emergency notifications to site supervisors, who are the users of the system, providing information to coordinate overall evacuation orders. The input is evacuation plan data, and based on this, the server outputs instructions such as operating procedures and personnel assignments to the users. This support enables orderly and rapid evacuations.

[0094] (Application Example 1)

[0095] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0096] Reducing the risk of accidents involving workers and machinery is crucial at work sites. Conventional management methods make it difficult to predict hazards and respond quickly, requiring prompt and accurate responses to ensure safety.

[0097] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0098] In this invention, the server includes means for acquiring video data from a plurality of imaging devices arranged in the work area, means for analyzing the video data in real time and recognizing the movements of workers and the operating status of machinery, and means for sending a warning to a portable information terminal and issuing a warning by sound or light when a hazard is detected. This makes it possible to immediately assess hazards in the work environment and provide workers with prompt warnings and specific countermeasures.

[0099] "Work area" refers to the area where a worker performs a specific task, and is the space in which machinery and equipment are located.

[0100] An "imaging device" is a device used to acquire video data, and includes image acquisition devices such as cameras.

[0101] "Video data" refers to visual information data acquired by an imaging device, in a format that allows for real-time analysis.

[0102] "Real-time analysis" is a method of processing video data instantly and evaluating the current situation almost instantly.

[0103] A "worker" refers to a person who is in a work area to perform a specific task.

[0104] "Mechanical equipment" refers to automatic or semi-automatic devices designed to perform specific tasks industrially or functionally.

[0105] "Hazard assessment" is a method of analyzing potential hazards in the work environment and evaluating their degree quantitatively or qualitatively.

[0106] "Notification means" refers to methods for transmitting warnings or information to workers, and includes feedback such as sound, light, and vibration.

[0107] A "portable information terminal" is an information processing device intended to be carried by workers, and includes smartphones and portable computers.

[0108] "Means of issuing warnings by sound or light" refers to methods of visually or audibly alerting workers when a hazard is detected.

[0109] The system that embodies an application of this invention is an advanced monitoring system aimed at safety management in a work area. The server acquires video data in real time from multiple imaging devices installed within the work area. The server analyzes this data using an AI algorithm to recognize the movements of workers and the operating status of machinery. For the analysis, Python and the TensorFlow library are used, and the degree of risk is evaluated by comparing it with past hazard cases stored in a database.

[0110] When a hazard is detected, the server immediately sends a warning to the worker's portable information terminal. The terminal has the function of alerting the worker using sound and light. This allows the worker to recognize the hazard in real time and take appropriate avoidance actions. In addition, in the event of an emergency, the server generates information on safe evacuation routes and assembly points and sends instructions to the worker and site supervisor via the terminal.

[0111] For example, if a forklift makes an unexpected movement in a factory, the server will determine this to be dangerous and send a warning to the terminal saying, "Be careful of the forklift." This system allows workers to receive safety information immediately, making it possible to prevent potential accidents.

[0112] When utilizing a generative AI model, an example of a prompt would be, "Please tell me how to quickly send an audio warning to a worker when a hazard is detected in the factory." In this way, the system can support the creation of a safe working environment and significantly reduce the risk of accidents.

[0113] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0114] Step 1:

[0115] The server receives video data in real time from multiple imaging devices located in the workspace. The input is a video stream from the camera devices, and the output is raw data stored on the server. In this process, video from the camera devices is captured and transferred to the server as a data stream.

[0116] Step 2:

[0117] The server analyzes the received video data using an AI algorithm. The input is the raw data acquired in step 1, and the output is the analysis result showing the worker's movements and the operating status of the machinery. The TensorFlow library is used for data processing, and image recognition technology is used to extract movements and anomalies.

[0118] Step 3:

[0119] The server compares the analysis results with a database of past hazard cases to assess the degree of risk. The input is the analysis results from step 2 and the past database, and the output is a risk assessment score. Data calculations are performed to calculate the probability and likelihood of occurrence of the hazard and generate the assessment score.

[0120] Step 4:

[0121] When a hazard is detected, the server sends a warning to the worker's portable device. The input is the hazard assessment score generated in step 3, and the output is the display of the warning message on the device. If a certain hazard level is exceeded, the server generates an audio or light warning and sends it to the device.

[0122] Step 5:

[0123] In the event of an emergency, the server determines evacuation routes and safe assembly points and sends instructions to terminals. Inputs are real-time work environment data and standard evacuation procedures, while output is specific evacuation instructions for the terminals. The server utilizes a generative AI model to calculate the optimal evacuation route based on prompt messages and provides information in real time.

[0124] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0125] This invention is a safety management system that combines an imaging device and an emotion recognition engine placed in the work area, aiming to comprehensively improve safety by focusing not only on the worker's actions but also on their emotional state. The server uses video data acquired from the imaging device and employs AI technology to analyze the worker's actions and the operating status of machinery in real time. Simultaneously, the emotion recognition engine analyzes the worker's facial expressions and tone of voice to evaluate stress levels and fatigue levels.

[0126] Based on the analysis results, the server assesses the level of risk in the work environment and, if necessary, immediately issues warnings to workers via terminals. For example, if a worker is determined to be under excessive stress, a voice message such as "Let's take a short break" is sent to suggest reducing the burden on the worker. In addition, site supervisors, who are the users, are notified of suggestions for reallocating workloads according to the emotional state of the workers, and, if necessary, suggestions for psychological support.

[0127] For example, if the emotion engine determines that a worker is fatigued due to prolonged work at height, the server will use this information to suggest a break and recommend that the user adjust their workload. Furthermore, in the event of an emergency, the server can utilize data from the emotion engine to provide workers with more appropriate evacuation instructions, preventing panic and supporting a swift and safe response. This embodiment enables comprehensive safety management that considers the mental health of workers, going beyond mere risk assessment.

[0128] The following describes the processing flow.

[0129] Step 1:

[0130] The server receives video data in real time from imaging devices placed in the work area. This includes video of each worker's facial expressions and voice, preparing the data for analysis.

[0131] Step 2:

[0132] The server uses AI algorithms to analyze video data in real time, recognizing the worker's movements and the operating status of machinery. Based on the analysis results, it evaluates whether the movements comply with safety standards.

[0133] Step 3:

[0134] Simultaneously, the emotion recognition engine analyzes the video data and evaluates the emotional state of the worker based on their facial expressions and tone of voice. This allows for the detection of stress levels, fatigue, and signs of distraction.

[0135] Step 4:

[0136] The server integrates the results of behavioral analysis and sentiment assessment, and calculates the overall risk level of the work environment by referring to a database of past cases. If the risk level exceeds a certain threshold, it is determined that immediate action is required.

[0137] Step 5:

[0138] When a hazard is detected, the server sends a warning to the terminal and provides necessary instructions to the worker. Specifically, if the work is being performed in a dangerous condition or the worker is emotionally unstable, it will issue instructions such as "Be careful" or "Take a break."

[0139] Step 6:

[0140] The server notifies the site supervisor (the user) based on their emotional state, suggesting workload reallocation and psychological support. This helps to support efficient management of the entire site.

[0141] Step 7:

[0142] In emergencies, the server provides evacuation instructions via terminals, taking into account the workers' emotional state. This promotes a swift and safe evacuation while minimizing stress and confusion.

[0143] Step 8:

[0144] The server records all warnings, sentiment evaluations, and response history in a database, which is used for future analysis and improvement of safety measures. Users use this information to develop safety improvement measures at the site.

[0145] (Example 2)

[0146] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0147] The challenge lies in comprehensively managing worker safety at the worksite and improving safety by taking into account factors such as workers' emotional state and fatigue levels, which were not adequately considered in conventional technologies. In particular, the accumulation of worker stress and fatigue increases the risk of overlooking dangerous situations, so it is necessary to detect these early and take appropriate measures.

[0148] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0149] In this invention, the server includes an information processing device that collects video information obtained from an imaging device, an information processing device that analyzes the video information to identify the actions of workers and the operation of work machines, and an information processing device that analyzes the emotional state of workers and evaluates their stress and fatigue levels. This enables comprehensive safety management that takes into account not only the actions of workers but also their emotional state.

[0150] An "imaging device" is a device used to acquire video information from a work site.

[0151] "Visual information" refers to the visual data of the work site acquired by the imaging device.

[0152] An "information processing device" is hardware or software that analyzes acquired video information and performs data processing according to a specific purpose.

[0153] A "worker" refers to a person who performs work within a specific work area.

[0154] "Action" refers to a series of physical movements performed by a worker.

[0155] "Working machinery" refers to devices and equipment used by workers or operating within a work area.

[0156] "Operation" is a comprehensive term that refers to the operating state of a work machine or the activities of a worker.

[0157] "Identification" is the process of recognizing a specific thing as distinct from others.

[0158] "Emotional state" refers to the mental and emotional state of a worker, and is primarily evaluated based on facial expressions and tone of voice.

[0159] "Stress" refers to a state of mental and physical tension that workers experience in response to external circumstances or stimuli.

[0160] "Fatigue level" is an indicator that represents the degree of physical and mental exhaustion experienced by a worker.

[0161] "Comprehensive safety management" is a management method aimed at maintaining and improving safety by taking into account both the actions and emotional state of workers.

[0162] This invention involves the coordinated operation of multiple information processing devices to construct a safety management system for work sites. Specifically, an imaging device installed in the work area collects video information of the worker and the surrounding environment and transmits it to a server. The server utilizes a generative AI model to analyze the video information.

[0163] The server first receives video information collected from a high-resolution imaging device and performs analysis using AI technology. This AI technology includes, for example, an "image analysis engine" and a "voice analysis engine." This allows the server to identify the worker's movements and the operating status of machinery. Furthermore, it uses an emotion recognition engine to analyze the worker's facial expressions and voice tone to assess stress and fatigue levels. For this, common software solutions such as a "voice analysis API" are used.

[0164] Furthermore, the terminal receives warning information sent from the server and immediately alerts the worker. This terminal has a built-in voice output function and can send appropriate messages based on the situation according to the manager's instructions. For example, if the terminal determines that the worker is fatigued, it will output a message such as "Let's take a short break."

[0165] Furthermore, the server makes a comprehensive decision based on these analysis results to ensure worker safety and notifies the site supervisor, who is the user, of the information. This allows the site supervisor to readjust the workload according to the workers' conditions and provide appropriate psychological support.

[0166] Examples of prompts for generative AI models:

[0167] "Please explain an emotion recognition system aimed at worker safety management. Specifically, please describe what kind of data is analyzed and how, and how the analysis results are utilized, including examples of specific hardware and software used."

[0168] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0169] Step 1:

[0170] The server collects video information from imaging devices installed in the work area. This information includes real-time status data of workers and machinery. It receives video data transmitted from the imaging devices as input and creates digital video information as output. This data serves as the raw material for subsequent analysis.

[0171] Step 2:

[0172] The server uses collected video information to perform motion analysis with a generated AI model. Specifically, it uses an image analysis engine to identify the worker's posture and movements, as well as the machine's operation. Video information is supplied to the AI ​​model as input, and motion analysis results are obtained as output. Based on these results, the safety status of the worker and the normal operation of the machine are checked.

[0173] Step 3:

[0174] The server uses an emotion recognition engine to assess the worker's emotional state. It analyzes facial expressions and voice tone to calculate stress levels and fatigue levels. It receives voice information and facial video as input and generates an emotional state assessment result as output. This allows for an understanding of the degree of mental burden the worker is experiencing.

[0175] Step 4:

[0176] The server integrates the results of behavioral analysis and emotional evaluation to assess the risk level of the work environment. It uses various analysis results as input and performs a comprehensive risk assessment based on these. The output provides a risk assessment, and data is prepared for issuing warnings as needed.

[0177] Step 5:

[0178] The terminal issues warnings to workers as needed, based on the risk assessment results sent from the server. For example, it may use voice messages or visual alerts to give instructions such as "take a break" or "be careful." It is programmed to receive risk assessment results as input and take appropriate actions based on their content. Specific warning messages are sent as output.

[0179] Step 6:

[0180] The site supervisor, acting as the user, receives information from the server and readjusts the work plan according to the workers' situations. For example, if a particular worker is experiencing stress, arrangements are made to alleviate it. The system receives a situation assessment from the server as input and sends instructions to the site, such as work assignments or additional breaks, as output.

[0181] (Application Example 2)

[0182] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0183] Ensuring worker safety is becoming increasingly important in modern industrial settings. However, many conventional technologies only monitor worker movements and do not anticipate or respond to emotionally-based risks. As a result, potential hazards caused by stress and fatigue may be overlooked. Furthermore, a challenge remains in providing specific support and adjustments tailored to the individual worker's condition.

[0184] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0185] In this invention, the server includes means for acquiring video data from multiple imaging devices arranged in the work area, means for analyzing the video data in real time to recognize the worker's movements and the operating status of machinery, and means for analyzing the worker's emotional state using an emotion recognition engine to evaluate stress levels and fatigue levels. This enables comprehensive safety management that takes into account not only the worker's movements but also their emotional state. Furthermore, by adjusting the operation of machinery based on the emotional state, it becomes possible to realize a safer and more efficient work environment tailored to each individual worker.

[0186] A "work area" is the physical or virtual area where work is performed, and it is the space in which workers and machinery are located.

[0187] An "imaging device" is a device used to acquire video data, and includes surveillance cameras and webcams.

[0188] "Video data" refers to visual data acquired by an imaging device, which records the actions of workers and their environment.

[0189] "Real-time analysis" refers to a method of instantly understanding the work situation by immediately processing the acquired video data.

[0190] "Worker actions" refer to the movements of a person's hands, feet, and body during work, and are factors that affect the work situation and efficiency.

[0191] "Operating status of machinery and equipment" refers to information indicating how the machinery in the factory is currently operating, including whether it is operating normally or stopped.

[0192] An "emotion recognition engine" is a system that analyzes the emotional state of a worker, using technology to infer emotions from facial expressions, tone of voice, and other factors.

[0193] "Stress levels and fatigue levels" indicate the degree of psychological and physical burden on workers and are indicators that can affect safety and efficiency.

[0194] "Assessing the level of risk" is the process of analyzing potential risks in the work environment and determining their severity.

[0195] "Notification means for issuing warnings" refers to communication methods used to warn workers when a hazard is detected, and includes voice and visual messages.

[0196] "Adjusting the operation of machinery" means modifying the movement of machinery based on analysis results to improve worker safety and efficiency.

[0197] To implement this invention, a high-precision imaging device installed in the work area and a server equipped with an emotion recognition engine are used. The server acquires video data in real time and uses AI technology to analyze the worker's movements and the operating status of machinery in detail. Image processing libraries such as OpenCV and Dlib are used for this analysis. Furthermore, the emotion recognition engine analyzes facial expressions and voice tone to evaluate stress levels and fatigue levels. This element requires an emotion recognition model, and for example, Deep Learning technology can be used for emotion analysis.

[0198] The server uses these analysis results to assess the level of risk in the work environment in real time. If a hazard is detected, it provides voice and visual warnings to the worker via a terminal. Mobile terminals and wearable devices may be used for this communication. Furthermore, if the server determines that a worker is under stress, it adjusts the operation of machinery and prompts the worker to take a break.

[0199] For example, if the emotion recognition engine determines that a worker performing long hours of work in a factory is experiencing high stress levels and accumulating fatigue, the server can temporarily suspend the robot's operation based on a pre-configured plan and send a voice message to the worker instructing them to take a break. This not only enhances the safety of the work environment but also supports the health management of workers.

[0200] Examples of prompts for generative AI models include the following:

[0201] "Design an AI system to enhance safety within a factory. Specifically, describe in detail how a factory robot can analyze workers' emotions and prompt them to take breaks when they are fatigued. Explain this using code examples, including the real-time processing flow."

[0202] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0203] Step 1:

[0204] The server acquires video data from an imaging device located in the workspace. It receives a real-time video stream as input and generates frame-by-frame image data as output. This image data is used for subsequent processing. Specifically, the server decodes the signal from the camera and prepares it for image processing.

[0205] Step 2:

[0206] The server uses acquired image data to analyze the worker's movements and the operating status of machinery in real time. The input is the image data generated in the previous step, and the output is the motion pattern and machine operating status as a result of the analysis. Image recognition technology using OpenCV and Dlib is applied to this data calculation. Specifically, the server identifies the worker's position and posture and compares them with a predetermined motion pattern.

[0207] Step 3:

[0208] The server analyzes the emotional state of workers using an emotion recognition engine. It receives real-time facial and voice data from workers as input and provides analysis data on stress levels and fatigue levels as output. A deep learning-based facial expression analysis model is used for data processing. Specifically, the server extracts facial feature points, matches them with the emotion model, and assigns emotion labels.

[0209] Step 4:

[0210] The server evaluates the risk level of the work environment based on these analysis results. It receives motion analysis results and emotion analysis data as input and calculates a risk evaluation score as output. This evaluation is compared with past case data. If the risk level exceeds a certain level, the server develops measures to mitigate the risk.

[0211] Step 5:

[0212] The server issues a warning to the worker via the terminal when a hazard is detected. The input is a hazard assessment score, and the output is a warning message. This message is sent to the terminal in either audio or visual format. Specifically, the server selects the content of the warning based on pre-configured criteria and communicates it quickly.

[0213] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0214] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0215] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0216] [Second Embodiment]

[0217] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0218] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0219] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0220] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0221] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0222] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0223] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0224] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0225] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0226] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0227] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0228] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0229] One embodiment of this invention includes a safety management system that uses video data captured by multiple camera devices placed in a work area. A server receives this data in real time and analyzes it using an AI algorithm. The analysis monitors the movements of workers and machinery, and compares it with a database of past hazard cases to evaluate the level of risk in the current work environment.

[0230] When a hazard is detected, the server immediately sends a warning to the terminal via a notification system. This terminal is a portable device with voice output and vibration alarm functions carried by the worker, and it communicates the nature of the hazard and evasive actions through voice and light. For example, if a dangerous heavy machine approaches, a voice command such as "Do not approach the heavy machine" is delivered directly to the worker from the terminal.

[0231] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation route. Notifications are also sent to the site supervisor, who is the user, allowing them to coordinate emergency responses. The server guides workers through terminals with specific evacuation procedures and safe assembly points, supporting a swift and orderly evacuation.

[0232] For example, if a worker continues working at height without their safety harness, the server will detect this situation in real time and issue a warning message saying, "Please put on your safety harness." Furthermore, if a sudden gust of wind occurs and the risk of the crane tipping over increases, the server can notify the worker to temporarily stop work, thereby ensuring the safety of the entire site. This embodiment makes it possible to enhance site safety and further protect the lives and health of workers.

[0233] The following describes the processing flow.

[0234] Step 1:

[0235] The server receives video data in real time from multiple camera devices placed in the workspace. The received data is converted to a format and resolution specified by the program for analysis.

[0236] Step 2:

[0237] The server uses AI algorithms to analyze received video data and detect worker movements, machine status, and environmental changes. This process utilizes computer vision technology to recognize the position, movement, and status of people and machines.

[0238] Step 3:

[0239] Based on the analysis results, the server compares them with past hazard cases stored in the database to evaluate the level of risk in the current work environment. The level of risk is determined using numerical values ​​and indicators, and if it exceeds the standard value, it is judged to be dangerous.

[0240] Step 4:

[0241] When a hazard is detected, the server immediately uses notification methods to send warning data to the terminal. The terminal then alerts the worker with sound, vibration, or light. For example, it might issue a specific instruction such as, "You are working at height; be sure to wear your safety harness."

[0242] Step 5:

[0243] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation order. It also sends a notification to the site supervisor (the user) and provides information to support the emergency response across the entire site.

[0244] Step 6:

[0245] The server transmits specific evacuation routes and instructions to workers via their terminals. These instructions include emergency action guidelines such as, "Follow the evacuation route and evacuate immediately to the designated safe location."

[0246] Step 7:

[0247] The server records all warnings and response histories in a database, which is used for future security improvements and analysis. Users can then use this data to provide feedback for further security enhancements.

[0248] (Example 1)

[0249] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0250] Safety management in the work area requires constantly monitoring worker movements and the status of machinery and equipment, and immediately detecting potential hazards to take countermeasures. However, existing safety management systems often lack real-time capabilities and accuracy, posing a challenge in providing appropriate evacuation instructions quickly, especially in emergencies.

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

[0252] In this invention, the server includes means for acquiring information from multiple imaging devices placed in the work area, means for analyzing the information in real time to recognize the actions of workers and the operation of machinery and equipment, and means for evaluating the degree of risk in the work environment based on past cases. This enables real-time risk detection and effective emergency response.

[0253] "Imaging device" is a general term for devices used to monitor a work area and acquire information in real time.

[0254] "Information" refers to all data acquired from imaging devices and other sensors, which the system uses for analysis and evaluation.

[0255] "Real-time analysis" refers to the process of immediately processing acquired information and quickly evaluating the current situation.

[0256] "Worker behavior" refers to all human movements and actions within the work area and is subject to safety management.

[0257] "Operation of machinery and equipment" refers to the state and operating conditions of devices operating within a work area.

[0258] "Past incidents" refer to data on accidents and dangerous situations recorded previously, and are used for risk assessment.

[0259] "Assessing the level of risk" is the process of expressing the risks to the current work environment using numerical values ​​and indicators.

[0260] "Communication means" refers to a device or function used to transmit information to workers or managers.

[0261] An "emergency situation" refers to a situation that indicates the occurrence of a danger or accident requiring immediate action.

[0262] An "evasive route" refers to a safe path for movement during an emergency.

[0263] "Support measures" refer to functions that provide workers with supplementary information and instructions to help them avoid danger and ensure safety.

[0264] In this embodiment of the invention, the server acquires information from multiple imaging devices installed in the work area. The imaging devices cover a wide area and monitor the actions of workers and the operation of machinery from multiple angles. This allows the server to receive information in real time. Next, the server preprocesses this information, removes noise, and then processes it to the required resolution.

[0265] The analysis uses AI algorithms such as TensorFlow and PyTorch. The server uses these tools to evaluate the work environment in real time and detect hazards by comparing it with past case data. If a hazard is detected, the server sends a warning to the terminal via communication.

[0266] The terminal is a device carried or worn by the worker, and its role is to convey warnings through voice output or vibration alarms. For example, if a worker is approaching dangerous machinery, it will notify them by voice, "Heavy machinery is approaching. Please maintain a safe distance."

[0267] In the event of an emergency, the server reassesss the risks and determines the optimal evacuation route. This information is communicated to the site supervisor, who is the user, enabling adjustments to the comprehensive evacuation plan. The server also guides workers through terminals with specific evacuation procedures and assembly points, supporting an orderly evacuation.

[0268] For example, if a worker removes a safety device while working at height, the server will immediately recognize this and issue a warning such as, "Please put on the safety device." Furthermore, if weather conditions make crane operation unsafe, the system will ensure safety by ordering a temporary halt to operations.

[0269] Examples of prompt statements to input into a generative AI model include the following:

[0270] "Please explain the hazard detection and real-time warning functions of the safety management system used at the work site."

[0271] "Please explain how to ensure necessary safety measures during work at heights using AI-based analysis."

[0272] Through these elements, this invention aims to efficiently improve safety in the workplace and protect the lives and health of workers.

[0273] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0274] Step 1:

[0275] The server acquires video data in real time from multiple imaging devices installed in the work area. This video data is image information based on settings such as resolution and field of view. The input image data is subjected to noise reduction filtering and conversion to the required resolution, and output as video data suitable for analysis.

[0276] Step 2:

[0277] The server analyzes the pre - processed video data using an AI algorithm. Specifically, it uses TensorFlow to detect and recognize the movements of workers and machinery. The input is the pre - processed video data for each frame, and the output is the identified movement patterns and status data of the machinery operations. Based on this data, it determines whether it is normal operation or there is a potential danger.

[0278] Step 3:

[0279] Based on the analysis results, the server compares with the past danger case database and evaluates the risk level of the current working environment. After performing the comparison process with the past case database as the input, it scores the current situation based on the output risk level evaluation data and determines whether it is a high - risk state.

[0280] Step 4:

[0281] If it is determined that the danger is high, the server sends a warning message to the terminal. The input is the risk level evaluation data. Based on this data, it generates the warning content and sends an appropriate message to the terminal as the output. The terminal notifies the operator of the message as sound or vibration, for example, conveys "Heavy machinery is approaching. Keep a distance."

[0282] Step 5:

[0283] In case of an emergency, the server examines the current situation, re - evaluates the risks, and then calculates the optimal evacuation route. The input is the risk level evaluation data and real - time environmental data. Based on this, it formulates an evacuation plan, and the output is a reasonable evacuation instruction.

[0284] Step 6:

[0285] The server also sends an emergency notice to the on - site supervisor, who is the user, and provides information for adjusting the overall evacuation instruction. The input is the evacuation plan data. Based on this, guidance such as operation procedures and personnel arrangements is output to the user. With this support, an orderly and rapid evacuation can be achieved.

[0286] (Application Example 1)

[0287] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[0288] At the work site, it is important to reduce the risk of accidents involving workers and machinery. With conventional management methods, it is difficult to predict dangers and respond quickly, and a prompt and accurate response to ensure safety is required.

[0289] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0290] In this invention, the server includes means for acquiring video data from a plurality of imaging devices arranged in a work area, means for analyzing the video data in real time to recognize the operations of workers and the operating states of machinery, and means for transmitting a warning to a portable information terminal and issuing a warning by sound or light when a danger is detected. Thereby, it becomes possible to immediately evaluate the danger in the working environment and provide workers with prompt warnings and specific countermeasures.

[0291] The "work area" is the range of the place where a worker performs a specific task and refers to the space where machinery is arranged.

[0292] The "imaging device" is a device for acquiring video data and includes an image acquisition device such as a camera.

[0293] The "video data" is the data of visual information acquired by an imaging device and is in a format that can be analyzed in real time.

[0294] "Real-time analysis" is a method of immediately processing video data and evaluating the current situation almost instantaneously.

[0295] A "worker" refers to a person who is in a work area to perform a specific task.

[0296] "Mechanical equipment" refers to automatic or semi-automatic devices designed to perform specific tasks industrially or functionally.

[0297] "Hazard assessment" is a method of analyzing potential hazards in the work environment and evaluating their degree quantitatively or qualitatively.

[0298] "Notification means" refers to methods for transmitting warnings or information to workers, and includes feedback such as sound, light, and vibration.

[0299] A "portable information terminal" is an information processing device intended to be carried by workers, and includes smartphones and portable computers.

[0300] "Means of issuing warnings by sound or light" refers to methods of visually or audibly alerting workers when a hazard is detected.

[0301] The system that embodies an application of this invention is an advanced monitoring system aimed at safety management in a work area. The server acquires video data in real time from multiple imaging devices installed within the work area. The server analyzes this data using an AI algorithm to recognize the movements of workers and the operating status of machinery. For the analysis, Python and the TensorFlow library are used, and the degree of risk is evaluated by comparing it with past hazard cases stored in a database.

[0302] When a hazard is detected, the server immediately sends a warning to the worker's portable information terminal. The terminal has the function of alerting the worker using sound and light. This allows the worker to recognize the hazard in real time and take appropriate avoidance actions. In addition, in the event of an emergency, the server generates information on safe evacuation routes and assembly points and sends instructions to the worker and site supervisor via the terminal.

[0303] As a specific example, when a forklift moves unexpectedly in a factory, the server determines this as dangerous and sends a warning "Please be careful of the forklift" to the terminal. With this system, workers can receive safety information immediately and it is possible to prevent potential accidents.

[0304] When utilizing a generative AI model, an example of a prompt sentence is "Please tell me how to quickly send an audio warning to workers when detecting danger in a factory." In this way, the system can support ensuring a safe working environment and significantly reduce the risk of accidents.

[0305] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0306] Step 1:

[0307] The server receives video data in real time from a plurality of imaging devices arranged in the work area. The input is a video stream from a camera device, and the output is raw data stored in the server. In this process, the video of the camera device is captured and transferred to the server as a data stream.

[0308] Step 2:

[0309] The server analyzes the received video data using an AI algorithm. The input is the raw data obtained in Step 1, and the output is an analysis result indicating the actions of the worker and the operating state of the mechanical device. The TensorFlow library is used for data processing, and operations are performed to extract actions and abnormalities using image recognition technology.

[0310] Step 3:

[0311] The server compares the analysis results with a database of past hazard cases to assess the degree of risk. The input is the analysis results from step 2 and the past database, and the output is a risk assessment score. Data calculations are performed to calculate the probability and likelihood of occurrence of the hazard and generate the assessment score.

[0312] Step 4:

[0313] When a hazard is detected, the server sends a warning to the worker's portable device. The input is the hazard assessment score generated in step 3, and the output is the display of the warning message on the device. If a certain hazard level is exceeded, the server generates an audio or light warning and sends it to the device.

[0314] Step 5:

[0315] In the event of an emergency, the server determines evacuation routes and safe assembly points and sends instructions to terminals. Inputs are real-time work environment data and standard evacuation procedures, while output is specific evacuation instructions for the terminals. The server utilizes a generative AI model to calculate the optimal evacuation route based on prompt messages and provides information in real time.

[0316] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0317] This invention is a safety management system that combines an imaging device and an emotion recognition engine placed in the work area, aiming to comprehensively improve safety by focusing not only on the worker's actions but also on their emotional state. The server uses video data acquired from the imaging device and employs AI technology to analyze the worker's actions and the operating status of machinery in real time. Simultaneously, the emotion recognition engine analyzes the worker's facial expressions and tone of voice to evaluate stress levels and fatigue levels.

[0318] Based on the analysis results, the server assesses the level of risk in the work environment and, if necessary, immediately issues warnings to workers via terminals. For example, if a worker is determined to be under excessive stress, a voice message such as "Let's take a short break" is sent to suggest reducing the burden on the worker. In addition, site supervisors, who are the users, are notified of suggestions for reallocating workloads according to the emotional state of the workers, and, if necessary, suggestions for psychological support.

[0319] For example, if the emotion engine determines that a worker is fatigued due to prolonged work at height, the server will use this information to suggest a break and recommend that the user adjust their workload. Furthermore, in the event of an emergency, the server can utilize data from the emotion engine to provide workers with more appropriate evacuation instructions, preventing panic and supporting a swift and safe response. This embodiment enables comprehensive safety management that considers the mental health of workers, going beyond mere risk assessment.

[0320] The following describes the processing flow.

[0321] Step 1:

[0322] The server receives video data in real time from imaging devices placed in the work area. This includes video of each worker's facial expressions and voice, preparing the data for analysis.

[0323] Step 2:

[0324] The server uses AI algorithms to analyze video data in real time, recognizing the worker's movements and the operating status of machinery. Based on the analysis results, it evaluates whether the movements comply with safety standards.

[0325] Step 3:

[0326] Simultaneously, the emotion recognition engine analyzes the video data and evaluates the emotional state of the worker based on their facial expressions and tone of voice. This allows for the detection of stress levels, fatigue, and signs of distraction.

[0327] Step 4:

[0328] The server integrates the results of behavioral analysis and sentiment assessment, and calculates the overall risk level of the work environment by referring to a database of past cases. If the risk level exceeds a certain threshold, it is determined that immediate action is required.

[0329] Step 5:

[0330] When a hazard is detected, the server sends a warning to the terminal and provides necessary instructions to the worker. Specifically, if the work is being performed in a dangerous condition or the worker is emotionally unstable, it will issue instructions such as "Be careful" or "Take a break."

[0331] Step 6:

[0332] The server notifies the site supervisor (the user) based on their emotional state, suggesting workload reallocation and psychological support. This helps to support efficient management of the entire site.

[0333] Step 7:

[0334] In emergencies, the server provides evacuation instructions via terminals, taking into account the workers' emotional state. This promotes a swift and safe evacuation while minimizing stress and confusion.

[0335] Step 8:

[0336] The server records all warnings, sentiment evaluations, and response history in a database, which is used for future analysis and improvement of safety measures. Users use this information to develop safety improvement measures at the site.

[0337] (Example 2)

[0338] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0339] The challenge lies in comprehensively managing worker safety at the worksite and improving safety by taking into account factors such as workers' emotional state and fatigue levels, which were not adequately considered in conventional technologies. In particular, the accumulation of worker stress and fatigue increases the risk of overlooking dangerous situations, so it is necessary to detect these early and take appropriate measures.

[0340] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0341] In this invention, the server includes an information processing device that collects video information obtained from an imaging device, an information processing device that analyzes the video information to identify the actions of workers and the operation of work machines, and an information processing device that analyzes the emotional state of workers and evaluates their stress and fatigue levels. This enables comprehensive safety management that takes into account not only the actions of workers but also their emotional state.

[0342] An "imaging device" is a device used to acquire video information from a work site.

[0343] "Visual information" refers to the visual data of the work site acquired by the imaging device.

[0344] An "information processing device" is hardware or software that analyzes acquired video information and performs data processing according to a specific purpose.

[0345] A "worker" refers to a person who performs work within a specific work area.

[0346] "Action" refers to a series of physical movements performed by a worker.

[0347] "Working machinery" refers to devices and equipment used by workers or operating within a work area.

[0348] "Operation" is a comprehensive term that refers to the operating state of a work machine or the activities of a worker.

[0349] "Identification" is the process of recognizing a specific thing as distinct from others.

[0350] "Emotional state" refers to the mental and emotional state of a worker, and is primarily evaluated based on facial expressions and tone of voice.

[0351] "Stress" refers to a state of mental and physical tension that workers experience in response to external circumstances or stimuli.

[0352] "Fatigue level" is an indicator that represents the degree of physical and mental exhaustion experienced by a worker.

[0353] "Comprehensive safety management" is a management method aimed at maintaining and improving safety by taking into account both the actions and emotional state of workers.

[0354] This invention involves the coordinated operation of multiple information processing devices to construct a safety management system for work sites. Specifically, an imaging device installed in the work area collects video information of the worker and the surrounding environment and transmits it to a server. The server utilizes a generative AI model to analyze the video information.

[0355] The server first receives video information collected from a high-resolution imaging device and performs analysis using AI technology. This AI technology includes, for example, an "image analysis engine" and a "voice analysis engine." This allows the server to identify the worker's movements and the operating status of machinery. Furthermore, it uses an emotion recognition engine to analyze the worker's facial expressions and voice tone to assess stress and fatigue levels. For this, common software solutions such as a "voice analysis API" are used.

[0356] Furthermore, the terminal receives warning information sent from the server and immediately alerts the worker. This terminal has a built-in voice output function and can send appropriate messages based on the situation according to the manager's instructions. For example, if the terminal determines that the worker is fatigued, it will output a message such as "Let's take a short break."

[0357] Furthermore, the server makes a comprehensive decision based on these analysis results to ensure worker safety and notifies the site supervisor, who is the user, of the information. This allows the site supervisor to readjust the workload according to the workers' conditions and provide appropriate psychological support.

[0358] Examples of prompts for generative AI models:

[0359] "Please explain an emotion recognition system aimed at worker safety management. Specifically, please describe what kind of data is analyzed and how, and how the analysis results are utilized, including examples of specific hardware and software used."

[0360] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0361] Step 1:

[0362] The server collects video information from imaging devices installed in the work area. This information includes real-time status data of workers and machinery. It receives video data transmitted from the imaging devices as input and creates digital video information as output. This data serves as the raw material for subsequent analysis.

[0363] Step 2:

[0364] The server uses collected video information to perform motion analysis with a generated AI model. Specifically, it uses an image analysis engine to identify the worker's posture and movements, as well as the machine's operation. Video information is supplied to the AI ​​model as input, and motion analysis results are obtained as output. Based on these results, the safety status of the worker and the normal operation of the machine are checked.

[0365] Step 3:

[0366] The server uses an emotion recognition engine to assess the worker's emotional state. It analyzes facial expressions and voice tone to calculate stress levels and fatigue levels. It receives voice information and facial video as input and generates an emotional state assessment result as output. This allows for an understanding of the degree of mental burden the worker is experiencing.

[0367] Step 4:

[0368] The server integrates the results of behavioral analysis and emotional evaluation to assess the risk level of the work environment. It uses various analysis results as input and performs a comprehensive risk assessment based on these. The output provides a risk assessment, and data is prepared for issuing warnings as needed.

[0369] Step 5:

[0370] The terminal issues warnings to workers as needed, based on the risk assessment results sent from the server. For example, it may use voice messages or visual alerts to give instructions such as "take a break" or "be careful." It is programmed to receive risk assessment results as input and take appropriate actions based on their content. Specific warning messages are sent as output.

[0371] Step 6:

[0372] The site supervisor, acting as the user, receives information from the server and readjusts the work plan according to the workers' situations. For example, if a particular worker is experiencing stress, arrangements are made to alleviate it. The system receives a situation assessment from the server as input and sends instructions to the site, such as work assignments or additional breaks, as output.

[0373] (Application Example 2)

[0374] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0375] Ensuring worker safety is becoming increasingly important in modern industrial settings. However, many conventional technologies only monitor worker movements and do not anticipate or respond to emotionally-based risks. As a result, potential hazards caused by stress and fatigue may be overlooked. Furthermore, a challenge remains in providing specific support and adjustments tailored to the individual worker's condition.

[0376] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0377] In this invention, the server includes means for acquiring video data from multiple imaging devices arranged in the work area, means for analyzing the video data in real time to recognize the worker's movements and the operating status of machinery, and means for analyzing the worker's emotional state using an emotion recognition engine to evaluate stress levels and fatigue levels. This enables comprehensive safety management that takes into account not only the worker's movements but also their emotional state. Furthermore, by adjusting the operation of machinery based on the emotional state, it becomes possible to realize a safer and more efficient work environment tailored to each individual worker.

[0378] A "work area" is the physical or virtual area where work is performed, and it is the space in which workers and machinery are located.

[0379] An "imaging device" is a device used to acquire video data, and includes surveillance cameras and webcams.

[0380] "Video data" refers to visual data acquired by an imaging device, which records the actions of workers and their environment.

[0381] "Real-time analysis" refers to a method of instantly understanding the work situation by immediately processing the acquired video data.

[0382] "Worker actions" refer to the movements of a person's hands, feet, and body during work, and are factors that affect the work situation and efficiency.

[0383] "Operating status of machinery and equipment" refers to information indicating how the machinery in the factory is currently operating, including whether it is operating normally or stopped.

[0384] An "emotion recognition engine" is a system that analyzes the emotional state of a worker, using technology to infer emotions from facial expressions, tone of voice, and other factors.

[0385] "Stress levels and fatigue levels" indicate the degree of psychological and physical burden on workers and are indicators that can affect safety and efficiency.

[0386] "Assessing the level of risk" is the process of analyzing potential risks in the work environment and determining their severity.

[0387] "Notification means for issuing warnings" refers to communication methods used to warn workers when a hazard is detected, and includes voice and visual messages.

[0388] "Adjusting the operation of machinery" means modifying the movement of machinery based on analysis results to improve worker safety and efficiency.

[0389] To implement this invention, a high-precision imaging device installed in the work area and a server equipped with an emotion recognition engine are used. The server acquires video data in real time and uses AI technology to analyze the worker's movements and the operating status of machinery in detail. Image processing libraries such as OpenCV and Dlib are used for this analysis. Furthermore, the emotion recognition engine analyzes facial expressions and voice tone to evaluate stress levels and fatigue levels. This element requires an emotion recognition model, and for example, Deep Learning technology can be used for emotion analysis.

[0390] The server uses these analysis results to assess the level of risk in the work environment in real time. If a hazard is detected, it provides voice and visual warnings to the worker via a terminal. Mobile terminals and wearable devices may be used for this communication. Furthermore, if the server determines that a worker is under stress, it adjusts the operation of machinery and prompts the worker to take a break.

[0391] For example, if the emotion recognition engine determines that a worker performing long hours of work in a factory is experiencing high stress levels and accumulating fatigue, the server can temporarily suspend the robot's operation based on a pre-configured plan and send a voice message to the worker instructing them to take a break. This not only enhances the safety of the work environment but also supports the health management of workers.

[0392] Examples of prompts for generative AI models include the following:

[0393] "Design an AI system to enhance safety within a factory. Specifically, describe in detail how a factory robot can analyze workers' emotions and prompt them to take breaks when they are fatigued. Explain this using code examples, including the real-time processing flow."

[0394] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0395] Step 1:

[0396] The server acquires video data from an imaging device located in the workspace. It receives a real-time video stream as input and generates frame-by-frame image data as output. This image data is used for subsequent processing. Specifically, the server decodes the signal from the camera and prepares it for image processing.

[0397] Step 2:

[0398] The server uses acquired image data to analyze the worker's movements and the operating status of machinery in real time. The input is the image data generated in the previous step, and the output is the motion pattern and machine operating status as a result of the analysis. Image recognition technology using OpenCV and Dlib is applied to this data calculation. Specifically, the server identifies the worker's position and posture and compares them with a predetermined motion pattern.

[0399] Step 3:

[0400] The server analyzes the emotional state of workers using an emotion recognition engine. It receives real-time facial and voice data from workers as input and provides analysis data on stress levels and fatigue levels as output. A deep learning-based facial expression analysis model is used for data processing. Specifically, the server extracts facial feature points, matches them with the emotion model, and assigns emotion labels.

[0401] Step 4:

[0402] The server evaluates the risk level of the work environment based on these analysis results. It receives motion analysis results and emotion analysis data as input and calculates a risk evaluation score as output. This evaluation is compared with past case data. If the risk level exceeds a certain level, the server develops measures to mitigate the risk.

[0403] Step 5:

[0404] The server issues a warning to the worker via the terminal when a hazard is detected. The input is a hazard assessment score, and the output is a warning message. This message is sent to the terminal in either audio or visual format. Specifically, the server selects the content of the warning based on pre-configured criteria and communicates it quickly.

[0405] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0406] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0407] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0408] [Third Embodiment]

[0409] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0410] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0411] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0412] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0413] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0414] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0415] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0416] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0417] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0418] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0419] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0420] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0421] One embodiment of this invention includes a safety management system that uses video data captured by multiple camera devices placed in a work area. A server receives this data in real time and analyzes it using an AI algorithm. The analysis monitors the movements of workers and machinery, and compares it with a database of past hazard cases to evaluate the level of risk in the current work environment.

[0422] When a hazard is detected, the server immediately sends a warning to the terminal via a notification system. This terminal is a portable device with voice output and vibration alarm functions carried by the worker, and it communicates the nature of the hazard and evasive actions through voice and light. For example, if a dangerous heavy machine approaches, a voice command such as "Do not approach the heavy machine" is delivered directly to the worker from the terminal.

[0423] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation route. Notifications are also sent to the site supervisor, who is the user, allowing them to coordinate emergency responses. The server guides workers through terminals with specific evacuation procedures and safe assembly points, supporting a swift and orderly evacuation.

[0424] For example, if a worker continues working at height without their safety harness, the server will detect this situation in real time and issue a warning message saying, "Please put on your safety harness." Furthermore, if a sudden gust of wind occurs and the risk of the crane tipping over increases, the server can notify the worker to temporarily stop work, thereby ensuring the safety of the entire site. This embodiment makes it possible to enhance site safety and further protect the lives and health of workers.

[0425] The following describes the processing flow.

[0426] Step 1:

[0427] The server receives video data in real time from multiple camera devices placed in the workspace. The received data is converted to a format and resolution specified by the program for analysis.

[0428] Step 2:

[0429] The server uses AI algorithms to analyze received video data and detect worker movements, machine status, and environmental changes. This process utilizes computer vision technology to recognize the position, movement, and status of people and machines.

[0430] Step 3:

[0431] Based on the analysis results, the server compares them with past hazard cases stored in the database to evaluate the level of risk in the current work environment. The level of risk is determined using numerical values ​​and indicators, and if it exceeds the standard value, it is judged to be dangerous.

[0432] Step 4:

[0433] When a hazard is detected, the server immediately uses notification methods to send warning data to the terminal. The terminal then alerts the worker with sound, vibration, or light. For example, it might issue a specific instruction such as, "You are working at height; be sure to wear your safety harness."

[0434] Step 5:

[0435] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation order. It also sends a notification to the site supervisor (the user) and provides information to support the emergency response across the entire site.

[0436] Step 6:

[0437] The server transmits specific evacuation routes and instructions to workers via their terminals. These instructions include emergency action guidelines such as, "Follow the evacuation route and evacuate immediately to the designated safe location."

[0438] Step 7:

[0439] The server records all warnings and response histories in a database, which is used for future security improvements and analysis. Users can then use this data to provide feedback for further security enhancements.

[0440] (Example 1)

[0441] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0442] Safety management in the work area requires constantly monitoring worker movements and the status of machinery and equipment, and immediately detecting potential hazards to take countermeasures. However, existing safety management systems often lack real-time capabilities and accuracy, posing a challenge in providing appropriate evacuation instructions quickly, especially in emergencies.

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

[0444] In this invention, the server includes means for acquiring information from multiple imaging devices placed in the work area, means for analyzing the information in real time to recognize the actions of workers and the operation of machinery and equipment, and means for evaluating the degree of risk in the work environment based on past cases. This enables real-time risk detection and effective emergency response.

[0445] "Imaging device" is a general term for devices used to monitor a work area and acquire information in real time.

[0446] "Information" refers to all data acquired from imaging devices and other sensors, which the system uses for analysis and evaluation.

[0447] "Real-time analysis" refers to the process of immediately processing acquired information and quickly evaluating the current situation.

[0448] "Worker behavior" refers to all human movements and actions within the work area and is subject to safety management.

[0449] "Operation of machinery and equipment" refers to the state and operating conditions of devices operating within a work area.

[0450] "Past incidents" refer to data on accidents and dangerous situations recorded previously, and are used for risk assessment.

[0451] "Assessing the level of risk" is the process of expressing the risks to the current work environment using numerical values ​​and indicators.

[0452] "Communication means" refers to a device or function used to transmit information to workers or managers.

[0453] An "emergency situation" refers to a situation that indicates the occurrence of a danger or accident requiring immediate action.

[0454] An "evasive route" refers to a safe path for movement during an emergency.

[0455] "Support measures" refer to functions that provide workers with supplementary information and instructions to help them avoid danger and ensure safety.

[0456] In this embodiment of the invention, the server acquires information from multiple imaging devices installed in the work area. The imaging devices cover a wide area and monitor the actions of workers and the operation of machinery from multiple angles. This allows the server to receive information in real time. Next, the server preprocesses this information, removes noise, and then processes it to the required resolution.

[0457] The analysis uses AI algorithms such as TensorFlow and PyTorch. The server uses these tools to evaluate the work environment in real time and detect hazards by comparing it with past case data. If a hazard is detected, the server sends a warning to the terminal via communication.

[0458] The terminal is a device carried or worn by the worker, and its role is to convey warnings through voice output or vibration alarms. For example, if a worker is approaching dangerous machinery, it will notify them by voice, "Heavy machinery is approaching. Please maintain a safe distance."

[0459] In the event of an emergency, the server reassesss the risks and determines the optimal evacuation route. This information is communicated to the site supervisor, who is the user, enabling adjustments to the comprehensive evacuation plan. The server also guides workers through terminals with specific evacuation procedures and assembly points, supporting an orderly evacuation.

[0460] For example, if a worker removes a safety device while working at height, the server will immediately recognize this and issue a warning such as, "Please put on the safety device." Furthermore, if weather conditions make crane operation unsafe, the system will ensure safety by ordering a temporary halt to operations.

[0461] Examples of prompt statements to input into a generative AI model include the following:

[0462] "Please explain the hazard detection and real-time warning functions of the safety management system used at the work site."

[0463] "Please explain how to ensure necessary safety measures during work at heights using AI-based analysis."

[0464] Through these elements, this invention aims to efficiently improve safety in the workplace and protect the lives and health of workers.

[0465] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0466] Step 1:

[0467] The server acquires video data in real time from multiple imaging devices installed in the work area. This video data is image information based on settings such as resolution and field of view. The input image data is subjected to noise reduction filtering and conversion to the required resolution, and output as video data suitable for analysis.

[0468] Step 2:

[0469] The server analyzes pre-processed video data using AI algorithms. Specifically, it uses TensorFlow to detect and recognize worker movements and machinery movements. The input is pre-processed video data frame by frame, and the output is identified motion patterns and data on the state of machine operation. Based on this data, it determines whether the operation is normal or if there is a potential hazard.

[0470] Step 3:

[0471] The server compares the analysis results with a database of past hazard cases to assess the risk level of the current work environment. After performing a comparison process using the past case database as input, it scores the current situation based on the output hazard assessment data to determine whether it is a high-risk situation.

[0472] Step 4:

[0473] If a high level of risk is detected, the server sends a warning message to the terminal. The input is risk assessment data, and based on this data, the server generates a warning and sends an appropriate message to the terminal as output. The terminal notifies the worker of this message via voice or vibration, for example, "Heavy machinery is approaching. Please maintain a safe distance."

[0474] Step 5:

[0475] In the event of an emergency, the server will examine the current situation, reassess the risks, and then calculate the optimal evacuation route. The inputs are risk assessment data and real-time environmental data, which are used to formulate an evacuation plan, and the output is a specific evacuation order.

[0476] Step 6:

[0477] The server also sends emergency notifications to site supervisors, who are the users of the system, providing information to coordinate overall evacuation orders. The input is evacuation plan data, and based on this, the server outputs instructions such as operating procedures and personnel assignments to the users. This support enables orderly and rapid evacuations.

[0478] (Application Example 1)

[0479] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0480] Reducing the risk of accidents involving workers and machinery is crucial at work sites. Conventional management methods make it difficult to predict hazards and respond quickly, requiring prompt and accurate responses to ensure safety.

[0481] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0482] In this invention, the server includes means for acquiring video data from a plurality of imaging devices arranged in the work area, means for analyzing the video data in real time and recognizing the movements of workers and the operating status of machinery, and means for sending a warning to a portable information terminal and issuing a warning by sound or light when a hazard is detected. This makes it possible to immediately assess hazards in the work environment and provide workers with prompt warnings and specific countermeasures.

[0483] "Work area" refers to the area where a worker performs a specific task, and is the space in which machinery and equipment are located.

[0484] An "imaging device" is a device used to acquire video data, and includes image acquisition devices such as cameras.

[0485] "Video data" refers to visual information data acquired by an imaging device, in a format that allows for real-time analysis.

[0486] "Real-time analysis" is a method of processing video data instantly and evaluating the current situation almost instantly.

[0487] A "worker" refers to a person who is in a work area to perform a specific task.

[0488] "Mechanical equipment" refers to automatic or semi-automatic devices designed to perform specific tasks industrially or functionally.

[0489] "Hazard assessment" is a method of analyzing potential hazards in the work environment and evaluating their degree quantitatively or qualitatively.

[0490] "Notification means" refers to methods for transmitting warnings or information to workers, and includes feedback such as sound, light, and vibration.

[0491] A "portable information terminal" is an information processing device intended to be carried by workers, and includes smartphones and portable computers.

[0492] "Means of issuing warnings by sound or light" refers to methods of visually or audibly alerting workers when a hazard is detected.

[0493] The system that embodies an application of this invention is an advanced monitoring system aimed at safety management in a work area. The server acquires video data in real time from multiple imaging devices installed within the work area. The server analyzes this data using an AI algorithm to recognize the movements of workers and the operating status of machinery. For the analysis, Python and the TensorFlow library are used, and the degree of risk is evaluated by comparing it with past hazard cases stored in a database.

[0494] When a hazard is detected, the server immediately sends a warning to the worker's portable information terminal. The terminal has the function of alerting the worker using sound and light. This allows the worker to recognize the hazard in real time and take appropriate avoidance actions. In addition, in the event of an emergency, the server generates information on safe evacuation routes and assembly points and sends instructions to the worker and site supervisor via the terminal.

[0495] For example, if a forklift makes an unexpected movement in a factory, the server will determine this to be dangerous and send a warning to the terminal saying, "Be careful of the forklift." This system allows workers to receive safety information immediately, making it possible to prevent potential accidents.

[0496] When utilizing a generative AI model, an example of a prompt would be, "Please tell me how to quickly send an audio warning to a worker when a hazard is detected in the factory." In this way, the system can support the creation of a safe working environment and significantly reduce the risk of accidents.

[0497] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0498] Step 1:

[0499] The server receives video data in real time from multiple imaging devices located in the workspace. The input is a video stream from the camera devices, and the output is raw data stored on the server. In this process, video from the camera devices is captured and transferred to the server as a data stream.

[0500] Step 2:

[0501] The server analyzes the received video data using an AI algorithm. The input is the raw data acquired in step 1, and the output is the analysis result showing the worker's movements and the operating status of the machinery. The TensorFlow library is used for data processing, and image recognition technology is used to extract movements and anomalies.

[0502] Step 3:

[0503] The server compares the analysis results with a database of past hazard cases to assess the degree of risk. The input is the analysis results from step 2 and the past database, and the output is a risk assessment score. Data calculations are performed to calculate the probability and likelihood of occurrence of the hazard and generate the assessment score.

[0504] Step 4:

[0505] When a hazard is detected, the server sends a warning to the worker's portable device. The input is the hazard assessment score generated in step 3, and the output is the display of the warning message on the device. If a certain hazard level is exceeded, the server generates an audio or light warning and sends it to the device.

[0506] Step 5:

[0507] In the event of an emergency, the server determines evacuation routes and safe assembly points and sends instructions to terminals. Inputs are real-time work environment data and standard evacuation procedures, while output is specific evacuation instructions for the terminals. The server utilizes a generative AI model to calculate the optimal evacuation route based on prompt messages and provides information in real time.

[0508] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0509] This invention is a safety management system that combines an imaging device and an emotion recognition engine placed in the work area, aiming to comprehensively improve safety by focusing not only on the worker's actions but also on their emotional state. The server uses video data acquired from the imaging device and employs AI technology to analyze the worker's actions and the operating status of machinery in real time. Simultaneously, the emotion recognition engine analyzes the worker's facial expressions and tone of voice to evaluate stress levels and fatigue levels.

[0510] Based on the analysis results, the server assesses the level of risk in the work environment and, if necessary, immediately issues warnings to workers via terminals. For example, if a worker is determined to be under excessive stress, a voice message such as "Let's take a short break" is sent to suggest reducing the burden on the worker. In addition, site supervisors, who are the users, are notified of suggestions for reallocating workloads according to the emotional state of the workers, and, if necessary, suggestions for psychological support.

[0511] For example, if the emotion engine determines that a worker is fatigued due to prolonged work at height, the server will use this information to suggest a break and recommend that the user adjust their workload. Furthermore, in the event of an emergency, the server can utilize data from the emotion engine to provide workers with more appropriate evacuation instructions, preventing panic and supporting a swift and safe response. This embodiment enables comprehensive safety management that considers the mental health of workers, going beyond mere risk assessment.

[0512] The following describes the processing flow.

[0513] Step 1:

[0514] The server receives video data in real time from imaging devices placed in the work area. This includes video of each worker's facial expressions and voice, preparing the data for analysis.

[0515] Step 2:

[0516] The server uses AI algorithms to analyze video data in real time, recognizing the worker's movements and the operating status of machinery. Based on the analysis results, it evaluates whether the movements comply with safety standards.

[0517] Step 3:

[0518] Simultaneously, the emotion recognition engine analyzes the video data and evaluates the emotional state of the worker based on their facial expressions and tone of voice. This allows for the detection of stress levels, fatigue, and signs of distraction.

[0519] Step 4:

[0520] The server integrates the results of behavioral analysis and sentiment assessment, and calculates the overall risk level of the work environment by referring to a database of past cases. If the risk level exceeds a certain threshold, it is determined that immediate action is required.

[0521] Step 5:

[0522] When a hazard is detected, the server sends a warning to the terminal and provides necessary instructions to the worker. Specifically, if the work is being performed in a dangerous condition or the worker is emotionally unstable, it will issue instructions such as "Be careful" or "Take a break."

[0523] Step 6:

[0524] The server notifies the site supervisor (the user) based on their emotional state, suggesting workload reallocation and psychological support. This helps to support efficient management of the entire site.

[0525] Step 7:

[0526] In emergencies, the server provides evacuation instructions via terminals, taking into account the workers' emotional state. This promotes a swift and safe evacuation while minimizing stress and confusion.

[0527] Step 8:

[0528] The server records all warnings, sentiment evaluations, and response history in a database, which is used for future analysis and improvement of safety measures. Users use this information to develop safety improvement measures at the site.

[0529] (Example 2)

[0530] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0531] The challenge lies in comprehensively managing worker safety at the worksite and improving safety by taking into account factors such as workers' emotional state and fatigue levels, which were not adequately considered in conventional technologies. In particular, the accumulation of worker stress and fatigue increases the risk of overlooking dangerous situations, so it is necessary to detect these early and take appropriate measures.

[0532] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0533] In this invention, the server includes an information processing device that collects video information obtained from an imaging device, an information processing device that analyzes the video information to identify the actions of workers and the operation of work machines, and an information processing device that analyzes the emotional state of workers and evaluates their stress and fatigue levels. This enables comprehensive safety management that takes into account not only the actions of workers but also their emotional state.

[0534] An "imaging device" is a device used to acquire video information from a work site.

[0535] "Visual information" refers to the visual data of the work site acquired by the imaging device.

[0536] An "information processing device" is hardware or software that analyzes acquired video information and performs data processing according to a specific purpose.

[0537] A "worker" refers to a person who performs work within a specific work area.

[0538] "Action" refers to a series of physical movements performed by a worker.

[0539] "Working machinery" refers to devices and equipment used by workers or operating within a work area.

[0540] "Operation" is a comprehensive term that refers to the operating state of a work machine or the activities of a worker.

[0541] "Identification" is the process of recognizing a specific thing as distinct from others.

[0542] "Emotional state" refers to the mental and emotional state of a worker, and is primarily evaluated based on facial expressions and tone of voice.

[0543] "Stress" refers to a state of mental and physical tension that workers experience in response to external circumstances or stimuli.

[0544] "Fatigue level" is an indicator that represents the degree of physical and mental exhaustion experienced by a worker.

[0545] "Comprehensive safety management" is a management method aimed at maintaining and improving safety by taking into account both the actions and emotional state of workers.

[0546] This invention involves the coordinated operation of multiple information processing devices to construct a safety management system for work sites. Specifically, an imaging device installed in the work area collects video information of the worker and the surrounding environment and transmits it to a server. The server utilizes a generative AI model to analyze the video information.

[0547] The server first receives video information collected from a high-resolution imaging device and performs analysis using AI technology. This AI technology includes, for example, an "image analysis engine" and a "voice analysis engine." This allows the server to identify the worker's movements and the operating status of machinery. Furthermore, it uses an emotion recognition engine to analyze the worker's facial expressions and voice tone to assess stress and fatigue levels. For this, common software solutions such as a "voice analysis API" are used.

[0548] Furthermore, the terminal receives warning information sent from the server and immediately alerts the worker. This terminal has a built-in voice output function and can send appropriate messages based on the situation according to the manager's instructions. For example, if the terminal determines that the worker is fatigued, it will output a message such as "Let's take a short break."

[0549] Furthermore, the server makes a comprehensive decision based on these analysis results to ensure worker safety and notifies the site supervisor, who is the user, of the information. This allows the site supervisor to readjust the workload according to the workers' conditions and provide appropriate psychological support.

[0550] Examples of prompts for generative AI models:

[0551] "Please explain an emotion recognition system aimed at worker safety management. Specifically, please describe what kind of data is analyzed and how, and how the analysis results are utilized, including examples of specific hardware and software used."

[0552] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0553] Step 1:

[0554] The server collects video information from imaging devices installed in the work area. This information includes real-time status data of workers and machinery. It receives video data transmitted from the imaging devices as input and creates digital video information as output. This data serves as the raw material for subsequent analysis.

[0555] Step 2:

[0556] The server uses collected video information to perform motion analysis with a generated AI model. Specifically, it uses an image analysis engine to identify the worker's posture and movements, as well as the machine's operation. Video information is supplied to the AI ​​model as input, and motion analysis results are obtained as output. Based on these results, the safety status of the worker and the normal operation of the machine are checked.

[0557] Step 3:

[0558] The server uses an emotion recognition engine to assess the worker's emotional state. It analyzes facial expressions and voice tone to calculate stress levels and fatigue levels. It receives voice information and facial video as input and generates an emotional state assessment result as output. This allows for an understanding of the degree of mental burden the worker is experiencing.

[0559] Step 4:

[0560] The server integrates the results of behavioral analysis and emotional evaluation to assess the risk level of the work environment. It uses various analysis results as input and performs a comprehensive risk assessment based on these. The output provides a risk assessment, and data is prepared for issuing warnings as needed.

[0561] Step 5:

[0562] The terminal issues warnings to workers as needed, based on the risk assessment results sent from the server. For example, it may use voice messages or visual alerts to give instructions such as "take a break" or "be careful." It is programmed to receive risk assessment results as input and take appropriate actions based on their content. Specific warning messages are sent as output.

[0563] Step 6:

[0564] The site supervisor, acting as the user, receives information from the server and readjusts the work plan according to the workers' situations. For example, if a particular worker is experiencing stress, arrangements are made to alleviate it. The system receives a situation assessment from the server as input and sends instructions to the site, such as work assignments or additional breaks, as output.

[0565] (Application Example 2)

[0566] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0567] Ensuring worker safety is becoming increasingly important in modern industrial settings. However, many conventional technologies only monitor worker movements and do not anticipate or respond to emotionally-based risks. As a result, potential hazards caused by stress and fatigue may be overlooked. Furthermore, a challenge remains in providing specific support and adjustments tailored to the individual worker's condition.

[0568] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0569] In this invention, the server includes means for acquiring video data from multiple imaging devices arranged in the work area, means for analyzing the video data in real time to recognize the worker's movements and the operating status of machinery, and means for analyzing the worker's emotional state using an emotion recognition engine to evaluate stress levels and fatigue levels. This enables comprehensive safety management that takes into account not only the worker's movements but also their emotional state. Furthermore, by adjusting the operation of machinery based on the emotional state, it becomes possible to realize a safer and more efficient work environment tailored to each individual worker.

[0570] A "work area" is the physical or virtual area where work is performed, and it is the space in which workers and machinery are located.

[0571] An "imaging device" is a device used to acquire video data, and includes surveillance cameras and webcams.

[0572] "Video data" refers to visual data acquired by an imaging device, which records the actions of workers and their environment.

[0573] "Real-time analysis" refers to a method of instantly understanding the work situation by immediately processing the acquired video data.

[0574] "Worker actions" refer to the movements of a person's hands, feet, and body during work, and are factors that affect the work situation and efficiency.

[0575] "Operating status of machinery and equipment" refers to information indicating how the machinery in the factory is currently operating, including whether it is operating normally or stopped.

[0576] An "emotion recognition engine" is a system that analyzes the emotional state of a worker, using technology to infer emotions from facial expressions, tone of voice, and other factors.

[0577] "Stress levels and fatigue levels" indicate the degree of psychological and physical burden on workers and are indicators that can affect safety and efficiency.

[0578] "Assessing the level of risk" is the process of analyzing potential risks in the work environment and determining their severity.

[0579] "Notification means for issuing warnings" refers to communication methods used to warn workers when a hazard is detected, and includes voice and visual messages.

[0580] "Adjusting the operation of machinery" means modifying the movement of machinery based on analysis results to improve worker safety and efficiency.

[0581] To implement this invention, a high-precision imaging device installed in the work area and a server equipped with an emotion recognition engine are used. The server acquires video data in real time and uses AI technology to analyze the worker's movements and the operating status of machinery in detail. Image processing libraries such as OpenCV and Dlib are used for this analysis. Furthermore, the emotion recognition engine analyzes facial expressions and voice tone to evaluate stress levels and fatigue levels. This element requires an emotion recognition model, and for example, Deep Learning technology can be used for emotion analysis.

[0582] The server uses these analysis results to assess the level of risk in the work environment in real time. If a hazard is detected, it provides voice and visual warnings to the worker via a terminal. Mobile terminals and wearable devices may be used for this communication. Furthermore, if the server determines that a worker is under stress, it adjusts the operation of machinery and prompts the worker to take a break.

[0583] For example, if the emotion recognition engine determines that a worker performing long hours of work in a factory is experiencing high stress levels and accumulating fatigue, the server can temporarily suspend the robot's operation based on a pre-configured plan and send a voice message to the worker instructing them to take a break. This not only enhances the safety of the work environment but also supports the health management of workers.

[0584] Examples of prompts for generative AI models include the following:

[0585] "Design an AI system to enhance safety within a factory. Specifically, describe in detail how a factory robot can analyze workers' emotions and prompt them to take breaks when they are fatigued. Explain this using code examples, including the real-time processing flow."

[0586] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0587] Step 1:

[0588] The server acquires video data from an imaging device located in the workspace. It receives a real-time video stream as input and generates frame-by-frame image data as output. This image data is used for subsequent processing. Specifically, the server decodes the signal from the camera and prepares it for image processing.

[0589] Step 2:

[0590] The server uses acquired image data to analyze the worker's movements and the operating status of machinery in real time. The input is the image data generated in the previous step, and the output is the motion pattern and machine operating status as a result of the analysis. Image recognition technology using OpenCV and Dlib is applied to this data calculation. Specifically, the server identifies the worker's position and posture and compares them with a predetermined motion pattern.

[0591] Step 3:

[0592] The server analyzes the emotional state of workers using an emotion recognition engine. It receives real-time facial and voice data from workers as input and provides analysis data on stress levels and fatigue levels as output. A deep learning-based facial expression analysis model is used for data processing. Specifically, the server extracts facial feature points, matches them with the emotion model, and assigns emotion labels.

[0593] Step 4:

[0594] The server evaluates the risk level of the work environment based on these analysis results. It receives motion analysis results and emotion analysis data as input and calculates a risk evaluation score as output. This evaluation is compared with past case data. If the risk level exceeds a certain level, the server develops measures to mitigate the risk.

[0595] Step 5:

[0596] The server issues a warning to the worker via the terminal when a hazard is detected. The input is a hazard assessment score, and the output is a warning message. This message is sent to the terminal in either audio or visual format. Specifically, the server selects the content of the warning based on pre-configured criteria and communicates it quickly.

[0597] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0598] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0599] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0600] [Fourth Embodiment]

[0601] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0602] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0603] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0604] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0605] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0606] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0607] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0608] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0609] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0610] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0611] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0612] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0613] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0614] One embodiment of this invention includes a safety management system that uses video data captured by multiple camera devices placed in a work area. A server receives this data in real time and analyzes it using an AI algorithm. The analysis monitors the movements of workers and machinery, and compares it with a database of past hazard cases to evaluate the level of risk in the current work environment.

[0615] When a hazard is detected, the server immediately sends a warning to the terminal via a notification system. This terminal is a portable device with voice output and vibration alarm functions carried by the worker, and it communicates the nature of the hazard and evasive actions through voice and light. For example, if a dangerous heavy machine approaches, a voice command such as "Do not approach the heavy machine" is delivered directly to the worker from the terminal.

[0616] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation route. Notifications are also sent to the site supervisor, who is the user, allowing them to coordinate emergency responses. The server guides workers through terminals with specific evacuation procedures and safe assembly points, supporting a swift and orderly evacuation.

[0617] For example, if a worker continues working at height without their safety harness, the server will detect this situation in real time and issue a warning message saying, "Please put on your safety harness." Furthermore, if a sudden gust of wind occurs and the risk of the crane tipping over increases, the server can notify the worker to temporarily stop work, thereby ensuring the safety of the entire site. This embodiment makes it possible to enhance site safety and further protect the lives and health of workers.

[0618] The following describes the processing flow.

[0619] Step 1:

[0620] The server receives video data in real time from multiple camera devices placed in the workspace. The received data is converted to a format and resolution specified by the program for analysis.

[0621] Step 2:

[0622] The server uses AI algorithms to analyze received video data and detect worker movements, machine status, and environmental changes. This process utilizes computer vision technology to recognize the position, movement, and status of people and machines.

[0623] Step 3:

[0624] Based on the analysis results, the server compares them with past hazard cases stored in the database to evaluate the level of risk in the current work environment. The level of risk is determined using numerical values ​​and indicators, and if it exceeds the standard value, it is judged to be dangerous.

[0625] Step 4:

[0626] When a hazard is detected, the server immediately uses notification methods to send warning data to the terminal. The terminal then alerts the worker with sound, vibration, or light. For example, it might issue a specific instruction such as, "You are working at height; be sure to wear your safety harness."

[0627] Step 5:

[0628] In the event of an emergency, the server automatically reassesss the risks and determines the optimal evacuation order. It also sends a notification to the site supervisor (the user) and provides information to support the emergency response across the entire site.

[0629] Step 6:

[0630] The server transmits specific evacuation routes and instructions to workers via their terminals. These instructions include emergency action guidelines such as, "Follow the evacuation route and evacuate immediately to the designated safe location."

[0631] Step 7:

[0632] The server records all warnings and response histories in a database, which is used for future security improvements and analysis. Users can then use this data to provide feedback for further security enhancements.

[0633] (Example 1)

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

[0635] Safety management in the work area requires constantly monitoring worker movements and the status of machinery and equipment, and immediately detecting potential hazards to take countermeasures. However, existing safety management systems often lack real-time capabilities and accuracy, posing a challenge in providing appropriate evacuation instructions quickly, especially in emergencies.

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

[0637] In this invention, the server includes means for acquiring information from multiple imaging devices placed in the work area, means for analyzing the information in real time to recognize the actions of workers and the operation of machinery and equipment, and means for evaluating the degree of risk in the work environment based on past cases. This enables real-time risk detection and effective emergency response.

[0638] "Imaging device" is a general term for devices used to monitor a work area and acquire information in real time.

[0639] "Information" refers to all data acquired from imaging devices and other sensors, which the system uses for analysis and evaluation.

[0640] "Real-time analysis" refers to the process of immediately processing acquired information and quickly evaluating the current situation.

[0641] "Worker behavior" refers to all human movements and actions within the work area and is subject to safety management.

[0642] "Operation of machinery and equipment" refers to the state and operating conditions of devices operating within a work area.

[0643] "Past incidents" refer to data on accidents and dangerous situations recorded previously, and are used for risk assessment.

[0644] "Assessing the level of risk" is the process of expressing the risks to the current work environment using numerical values ​​and indicators.

[0645] "Communication means" refers to a device or function used to transmit information to workers or managers.

[0646] An "emergency situation" refers to a situation that indicates the occurrence of a danger or accident requiring immediate action.

[0647] An "evasive route" refers to a safe path for movement during an emergency.

[0648] "Support measures" refer to functions that provide workers with supplementary information and instructions to help them avoid danger and ensure safety.

[0649] In this embodiment of the invention, the server acquires information from multiple imaging devices installed in the work area. The imaging devices cover a wide area and monitor the actions of workers and the operation of machinery from multiple angles. This allows the server to receive information in real time. Next, the server preprocesses this information, removes noise, and then processes it to the required resolution.

[0650] The analysis uses AI algorithms such as TensorFlow and PyTorch. The server uses these tools to evaluate the work environment in real time and detect hazards by comparing it with past case data. If a hazard is detected, the server sends a warning to the terminal via communication.

[0651] The terminal is a device carried or worn by the worker, and its role is to convey warnings through voice output or vibration alarms. For example, if a worker is approaching dangerous machinery, it will notify them by voice, "Heavy machinery is approaching. Please maintain a safe distance."

[0652] In the event of an emergency, the server reassesss the risks and determines the optimal evacuation route. This information is communicated to the site supervisor, who is the user, enabling adjustments to the comprehensive evacuation plan. The server also guides workers through terminals with specific evacuation procedures and assembly points, supporting an orderly evacuation.

[0653] For example, if a worker removes a safety device while working at height, the server will immediately recognize this and issue a warning such as, "Please put on the safety device." Furthermore, if weather conditions make crane operation unsafe, the system will ensure safety by ordering a temporary halt to operations.

[0654] Examples of prompt statements to input into a generative AI model include the following:

[0655] "Please explain the hazard detection and real-time warning functions of the safety management system used at the work site."

[0656] "Please explain how to ensure necessary safety measures during work at heights using AI-based analysis."

[0657] Through these elements, this invention aims to efficiently improve safety in the workplace and protect the lives and health of workers.

[0658] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0659] Step 1:

[0660] The server acquires video data in real time from multiple imaging devices installed in the work area. This video data is image information based on settings such as resolution and field of view. The input image data is subjected to noise reduction filtering and conversion to the required resolution, and output as video data suitable for analysis.

[0661] Step 2:

[0662] The server analyzes pre-processed video data using AI algorithms. Specifically, it uses TensorFlow to detect and recognize worker movements and machinery movements. The input is pre-processed video data frame by frame, and the output is identified motion patterns and data on the state of machine operation. Based on this data, it determines whether the operation is normal or if there is a potential hazard.

[0663] Step 3:

[0664] The server compares the analysis results with a database of past hazard cases to assess the risk level of the current work environment. After performing a comparison process using the past case database as input, it scores the current situation based on the output hazard assessment data to determine whether it is a high-risk situation.

[0665] Step 4:

[0666] If a high level of risk is detected, the server sends a warning message to the terminal. The input is risk assessment data, and based on this data, the server generates a warning and sends an appropriate message to the terminal as output. The terminal notifies the worker of this message via voice or vibration, for example, "Heavy machinery is approaching. Please maintain a safe distance."

[0667] Step 5:

[0668] In the event of an emergency, the server will examine the current situation, reassess the risks, and then calculate the optimal evacuation route. The inputs are risk assessment data and real-time environmental data, which are used to formulate an evacuation plan, and the output is a specific evacuation order.

[0669] Step 6:

[0670] The server also sends emergency notifications to site supervisors, who are the users of the system, providing information to coordinate overall evacuation orders. The input is evacuation plan data, and based on this, the server outputs instructions such as operating procedures and personnel assignments to the users. This support enables orderly and rapid evacuations.

[0671] (Application Example 1)

[0672] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0673] Reducing the risk of accidents involving workers and machinery is crucial at work sites. Conventional management methods make it difficult to predict hazards and respond quickly, requiring prompt and accurate responses to ensure safety.

[0674] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0675] In this invention, the server includes means for acquiring video data from a plurality of imaging devices arranged in the work area, means for analyzing the video data in real time and recognizing the movements of workers and the operating status of machinery, and means for sending a warning to a portable information terminal and issuing a warning by sound or light when a hazard is detected. This makes it possible to immediately assess hazards in the work environment and provide workers with prompt warnings and specific countermeasures.

[0676] "Work area" refers to the area where a worker performs a specific task, and is the space in which machinery and equipment are located.

[0677] An "imaging device" is a device used to acquire video data, and includes image acquisition devices such as cameras.

[0678] "Video data" refers to visual information data acquired by an imaging device, in a format that allows for real-time analysis.

[0679] "Real-time analysis" is a method of processing video data instantly and evaluating the current situation almost instantly.

[0680] A "worker" refers to a person who is in a work area to perform a specific task.

[0681] "Mechanical equipment" refers to automatic or semi-automatic devices designed to perform specific tasks industrially or functionally.

[0682] "Hazard assessment" is a method of analyzing potential hazards in the work environment and evaluating their degree quantitatively or qualitatively.

[0683] "Notification means" refers to methods for transmitting warnings or information to workers, and includes feedback such as sound, light, and vibration.

[0684] A "portable information terminal" is an information processing device intended to be carried by workers, and includes smartphones and portable computers.

[0685] "Means of issuing warnings by sound or light" refers to methods of visually or audibly alerting workers when a hazard is detected.

[0686] The system that embodies an application of this invention is an advanced monitoring system aimed at safety management in a work area. The server acquires video data in real time from multiple imaging devices installed within the work area. The server analyzes this data using an AI algorithm to recognize the movements of workers and the operating status of machinery. For the analysis, Python and the TensorFlow library are used, and the degree of risk is evaluated by comparing it with past hazard cases stored in a database.

[0687] When a hazard is detected, the server immediately sends a warning to the worker's portable information terminal. The terminal has the function of alerting the worker using sound and light. This allows the worker to recognize the hazard in real time and take appropriate avoidance actions. In addition, in the event of an emergency, the server generates information on safe evacuation routes and assembly points and sends instructions to the worker and site supervisor via the terminal.

[0688] For example, if a forklift makes an unexpected movement in a factory, the server will determine this to be dangerous and send a warning to the terminal saying, "Be careful of the forklift." This system allows workers to receive safety information immediately, making it possible to prevent potential accidents.

[0689] When utilizing a generative AI model, an example of a prompt would be, "Please tell me how to quickly send an audio warning to a worker when a hazard is detected in the factory." In this way, the system can support the creation of a safe working environment and significantly reduce the risk of accidents.

[0690] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0691] Step 1:

[0692] The server receives video data in real time from multiple imaging devices located in the workspace. The input is a video stream from the camera devices, and the output is raw data stored on the server. In this process, video from the camera devices is captured and transferred to the server as a data stream.

[0693] Step 2:

[0694] The server analyzes the received video data using an AI algorithm. The input is the raw data acquired in step 1, and the output is the analysis result showing the worker's movements and the operating status of the machinery. The TensorFlow library is used for data processing, and image recognition technology is used to extract movements and anomalies.

[0695] Step 3:

[0696] The server compares the analysis results with a database of past hazard cases to assess the degree of risk. The input is the analysis results from step 2 and the past database, and the output is a risk assessment score. Data calculations are performed to calculate the probability and likelihood of occurrence of the hazard and generate the assessment score.

[0697] Step 4:

[0698] When a hazard is detected, the server sends a warning to the worker's portable device. The input is the hazard assessment score generated in step 3, and the output is the display of the warning message on the device. If a certain hazard level is exceeded, the server generates an audio or light warning and sends it to the device.

[0699] Step 5:

[0700] In the event of an emergency, the server determines evacuation routes and safe assembly points and sends instructions to terminals. Inputs are real-time work environment data and standard evacuation procedures, while output is specific evacuation instructions for the terminals. The server utilizes a generative AI model to calculate the optimal evacuation route based on prompt messages and provides information in real time.

[0701] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0702] This invention is a safety management system that combines an imaging device and an emotion recognition engine placed in the work area, aiming to comprehensively improve safety by focusing not only on the worker's actions but also on their emotional state. The server uses video data acquired from the imaging device and employs AI technology to analyze the worker's actions and the operating status of machinery in real time. Simultaneously, the emotion recognition engine analyzes the worker's facial expressions and tone of voice to evaluate stress levels and fatigue levels.

[0703] Based on the analysis results, the server assesses the level of risk in the work environment and, if necessary, immediately issues warnings to workers via terminals. For example, if a worker is determined to be under excessive stress, a voice message such as "Let's take a short break" is sent to suggest reducing the burden on the worker. In addition, site supervisors, who are the users, are notified of suggestions for reallocating workloads according to the emotional state of the workers, and, if necessary, suggestions for psychological support.

[0704] For example, if the emotion engine determines that a worker is fatigued due to prolonged work at height, the server will use this information to suggest a break and recommend that the user adjust their workload. Furthermore, in the event of an emergency, the server can utilize data from the emotion engine to provide workers with more appropriate evacuation instructions, preventing panic and supporting a swift and safe response. This embodiment enables comprehensive safety management that considers the mental health of workers, going beyond mere risk assessment.

[0705] The following describes the processing flow.

[0706] Step 1:

[0707] The server receives video data in real time from imaging devices placed in the work area. This includes video of each worker's facial expressions and voice, preparing the data for analysis.

[0708] Step 2:

[0709] The server uses AI algorithms to analyze video data in real time, recognizing the worker's movements and the operating status of machinery. Based on the analysis results, it evaluates whether the movements comply with safety standards.

[0710] Step 3:

[0711] Simultaneously, the emotion recognition engine analyzes the video data and evaluates the emotional state of the worker based on their facial expressions and tone of voice. This allows for the detection of stress levels, fatigue, and signs of distraction.

[0712] Step 4:

[0713] The server integrates the results of behavioral analysis and sentiment assessment, and calculates the overall risk level of the work environment by referring to a database of past cases. If the risk level exceeds a certain threshold, it is determined that immediate action is required.

[0714] Step 5:

[0715] When a hazard is detected, the server sends a warning to the terminal and provides necessary instructions to the worker. Specifically, if the work is being performed in a dangerous condition or the worker is emotionally unstable, it will issue instructions such as "Be careful" or "Take a break."

[0716] Step 6:

[0717] The server notifies the site supervisor (the user) based on their emotional state, suggesting workload reallocation and psychological support. This helps to support efficient management of the entire site.

[0718] Step 7:

[0719] In emergencies, the server provides evacuation instructions via terminals, taking into account the workers' emotional state. This promotes a swift and safe evacuation while minimizing stress and confusion.

[0720] Step 8:

[0721] The server records all warnings, sentiment evaluations, and response history in a database, which is used for future analysis and improvement of safety measures. Users use this information to develop safety improvement measures at the site.

[0722] (Example 2)

[0723] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0724] The challenge lies in comprehensively managing worker safety at the worksite and improving safety by taking into account factors such as workers' emotional state and fatigue levels, which were not adequately considered in conventional technologies. In particular, the accumulation of worker stress and fatigue increases the risk of overlooking dangerous situations, so it is necessary to detect these early and take appropriate measures.

[0725] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0726] In this invention, the server includes an information processing device that collects video information obtained from an imaging device, an information processing device that analyzes the video information to identify the actions of workers and the operation of work machines, and an information processing device that analyzes the emotional state of workers and evaluates their stress and fatigue levels. This enables comprehensive safety management that takes into account not only the actions of workers but also their emotional state.

[0727] An "imaging device" is a device used to acquire video information from a work site.

[0728] "Visual information" refers to the visual data of the work site acquired by the imaging device.

[0729] An "information processing device" is hardware or software that analyzes acquired video information and performs data processing according to a specific purpose.

[0730] A "worker" refers to a person who performs work within a specific work area.

[0731] "Action" refers to a series of physical movements performed by a worker.

[0732] "Working machinery" refers to devices and equipment used by workers or operating within a work area.

[0733] "Operation" is a comprehensive term that refers to the operating state of a work machine or the activities of a worker.

[0734] "Identification" is the process of recognizing a specific thing as distinct from others.

[0735] "Emotional state" refers to the mental and emotional state of a worker, and is primarily evaluated based on facial expressions and tone of voice.

[0736] "Stress" refers to a state of mental and physical tension that workers experience in response to external circumstances or stimuli.

[0737] "Fatigue level" is an indicator that represents the degree of physical and mental exhaustion experienced by a worker.

[0738] "Comprehensive safety management" is a management method aimed at maintaining and improving safety by taking into account both the actions and emotional state of workers.

[0739] This invention involves the coordinated operation of multiple information processing devices to construct a safety management system for work sites. Specifically, an imaging device installed in the work area collects video information of the worker and the surrounding environment and transmits it to a server. The server utilizes a generative AI model to analyze the video information.

[0740] The server first receives video information collected from a high-resolution imaging device and performs analysis using AI technology. This AI technology includes, for example, an "image analysis engine" and a "voice analysis engine." This allows the server to identify the worker's movements and the operating status of machinery. Furthermore, it uses an emotion recognition engine to analyze the worker's facial expressions and voice tone to assess stress and fatigue levels. For this, common software solutions such as a "voice analysis API" are used.

[0741] Furthermore, the terminal receives warning information sent from the server and immediately alerts the worker. This terminal has a built-in voice output function and can send appropriate messages based on the situation according to the manager's instructions. For example, if the terminal determines that the worker is fatigued, it will output a message such as "Let's take a short break."

[0742] Furthermore, the server makes a comprehensive decision based on these analysis results to ensure worker safety and notifies the site supervisor, who is the user, of the information. This allows the site supervisor to readjust the workload according to the workers' conditions and provide appropriate psychological support.

[0743] Examples of prompts for generative AI models:

[0744] "Please explain an emotion recognition system aimed at worker safety management. Specifically, please describe what kind of data is analyzed and how, and how the analysis results are utilized, including examples of specific hardware and software used."

[0745] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0746] Step 1:

[0747] The server collects video information from imaging devices installed in the work area. This information includes real-time status data of workers and machinery. It receives video data transmitted from the imaging devices as input and creates digital video information as output. This data serves as the raw material for subsequent analysis.

[0748] Step 2:

[0749] The server uses collected video information to perform motion analysis with a generated AI model. Specifically, it uses an image analysis engine to identify the worker's posture and movements, as well as the machine's operation. Video information is supplied to the AI ​​model as input, and motion analysis results are obtained as output. Based on these results, the safety status of the worker and the normal operation of the machine are checked.

[0750] Step 3:

[0751] The server uses an emotion recognition engine to assess the worker's emotional state. It analyzes facial expressions and voice tone to calculate stress levels and fatigue levels. It receives voice information and facial video as input and generates an emotional state assessment result as output. This allows for an understanding of the degree of mental burden the worker is experiencing.

[0752] Step 4:

[0753] The server integrates the results of behavioral analysis and emotional evaluation to assess the risk level of the work environment. It uses various analysis results as input and performs a comprehensive risk assessment based on these. The output provides a risk assessment, and data is prepared for issuing warnings as needed.

[0754] Step 5:

[0755] The terminal issues warnings to workers as needed, based on the risk assessment results sent from the server. For example, it may use voice messages or visual alerts to give instructions such as "take a break" or "be careful." It is programmed to receive risk assessment results as input and take appropriate actions based on their content. Specific warning messages are sent as output.

[0756] Step 6:

[0757] The site supervisor, acting as the user, receives information from the server and readjusts the work plan according to the workers' situations. For example, if a particular worker is experiencing stress, arrangements are made to alleviate it. The system receives a situation assessment from the server as input and sends instructions to the site, such as work assignments or additional breaks, as output.

[0758] (Application Example 2)

[0759] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0760] Ensuring worker safety is becoming increasingly important in modern industrial settings. However, many conventional technologies only monitor worker movements and do not anticipate or respond to emotionally-based risks. As a result, potential hazards caused by stress and fatigue may be overlooked. Furthermore, a challenge remains in providing specific support and adjustments tailored to the individual worker's condition.

[0761] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0762] In this invention, the server includes means for acquiring video data from multiple imaging devices arranged in the work area, means for analyzing the video data in real time to recognize the worker's movements and the operating status of machinery, and means for analyzing the worker's emotional state using an emotion recognition engine to evaluate stress levels and fatigue levels. This enables comprehensive safety management that takes into account not only the worker's movements but also their emotional state. Furthermore, by adjusting the operation of machinery based on the emotional state, it becomes possible to realize a safer and more efficient work environment tailored to each individual worker.

[0763] A "work area" is the physical or virtual area where work is performed, and it is the space in which workers and machinery are located.

[0764] An "imaging device" is a device used to acquire video data, and includes surveillance cameras and webcams.

[0765] "Video data" refers to visual data acquired by an imaging device, which records the actions of workers and their environment.

[0766] "Real-time analysis" refers to a method of instantly understanding the work situation by immediately processing the acquired video data.

[0767] "Worker actions" refer to the movements of a person's hands, feet, and body during work, and are factors that affect the work situation and efficiency.

[0768] "Operating status of machinery and equipment" refers to information indicating how the machinery in the factory is currently operating, including whether it is operating normally or stopped.

[0769] An "emotion recognition engine" is a system that analyzes the emotional state of a worker, using technology to infer emotions from facial expressions, tone of voice, and other factors.

[0770] "Stress levels and fatigue levels" indicate the degree of psychological and physical burden on workers and are indicators that can affect safety and efficiency.

[0771] "Assessing the level of risk" is the process of analyzing potential risks in the work environment and determining their severity.

[0772] "Notification means for issuing warnings" refers to communication methods used to warn workers when a hazard is detected, and includes voice and visual messages.

[0773] "Adjusting the operation of machinery" means modifying the movement of machinery based on analysis results to improve worker safety and efficiency.

[0774] To implement this invention, a high-precision imaging device installed in the work area and a server equipped with an emotion recognition engine are used. The server acquires video data in real time and uses AI technology to analyze the worker's movements and the operating status of machinery in detail. Image processing libraries such as OpenCV and Dlib are used for this analysis. Furthermore, the emotion recognition engine analyzes facial expressions and voice tone to evaluate stress levels and fatigue levels. This element requires an emotion recognition model, and for example, Deep Learning technology can be used for emotion analysis.

[0775] The server uses these analysis results to assess the level of risk in the work environment in real time. If a hazard is detected, it provides voice and visual warnings to the worker via a terminal. Mobile terminals and wearable devices may be used for this communication. Furthermore, if the server determines that a worker is under stress, it adjusts the operation of machinery and prompts the worker to take a break.

[0776] For example, if the emotion recognition engine determines that a worker performing long hours of work in a factory is experiencing high stress levels and accumulating fatigue, the server can temporarily suspend the robot's operation based on a pre-configured plan and send a voice message to the worker instructing them to take a break. This not only enhances the safety of the work environment but also supports the health management of workers.

[0777] Examples of prompts for generative AI models include the following:

[0778] "Design an AI system to enhance safety within a factory. Specifically, describe in detail how a factory robot can analyze workers' emotions and prompt them to take breaks when they are fatigued. Explain this using code examples, including the real-time processing flow."

[0779] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0780] Step 1:

[0781] The server acquires video data from an imaging device located in the workspace. It receives a real-time video stream as input and generates frame-by-frame image data as output. This image data is used for subsequent processing. Specifically, the server decodes the signal from the camera and prepares it for image processing.

[0782] Step 2:

[0783] The server uses acquired image data to analyze the worker's movements and the operating status of machinery in real time. The input is the image data generated in the previous step, and the output is the motion pattern and machine operating status as a result of the analysis. Image recognition technology using OpenCV and Dlib is applied to this data calculation. Specifically, the server identifies the worker's position and posture and compares them with a predetermined motion pattern.

[0784] Step 3:

[0785] The server analyzes the emotional state of workers using an emotion recognition engine. It receives real-time facial and voice data from workers as input and provides analysis data on stress levels and fatigue levels as output. A deep learning-based facial expression analysis model is used for data processing. Specifically, the server extracts facial feature points, matches them with the emotion model, and assigns emotion labels.

[0786] Step 4:

[0787] The server evaluates the risk level of the work environment based on these analysis results. It receives motion analysis results and emotion analysis data as input and calculates a risk evaluation score as output. This evaluation is compared with past case data. If the risk level exceeds a certain level, the server develops measures to mitigate the risk.

[0788] Step 5:

[0789] The server issues a warning to the worker via the terminal when a hazard is detected. The input is a hazard assessment score, and the output is a warning message. This message is sent to the terminal in either audio or visual format. Specifically, the server selects the content of the warning based on pre-configured criteria and communicates it quickly.

[0790] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0791] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0792] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0793] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0794] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0795] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0796] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0797] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0798] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0799] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0800] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0801] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0802] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0803] 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.

[0804] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0805] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0806] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0807] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0808] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0809] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0810] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0811] The following is further disclosed regarding the embodiments described above.

[0812] (Claim 1)

[0813] A means for acquiring video data from multiple imaging devices arranged in the work area,

[0814] A means for analyzing the aforementioned video data in real time and recognizing the worker's movements and the operating status of the machinery,

[0815] A method for evaluating the degree of risk in the work environment based on past cases,

[0816] A notification system for issuing warnings to workers when a hazard is detected,

[0817] A system that includes this.

[0818] (Claim 2)

[0819] The system according to claim 1, which provides specific work instructions to workers based on a risk assessment.

[0820] (Claim 3)

[0821] The system according to claim 1, which notifies the site supervisor and sends evacuation instructions to workers when an emergency occurs.

[0822] "Example 1"

[0823] (Claim 1)

[0824] Means for acquiring information from multiple imaging devices arranged in the work area,

[0825] A means for analyzing the aforementioned information in real time and recognizing the actions of workers and the operation of machinery and equipment,

[0826] A method for evaluating the degree of risk in the work environment based on past cases,

[0827] A means of communication to warn workers when a hazard is detected,

[0828] Means for identifying emergencies and determining the optimal avoidance route,

[0829] Support means to guide workers on safe evacuation methods,

[0830] A system that includes this.

[0831] (Claim 2)

[0832] The system according to claim 1, which provides specific action instructions to an operator based on a risk assessment.

[0833] (Claim 3)

[0834] The system according to claim 1, which notifies the administrator and sends evacuation instructions to workers when an emergency occurs.

[0835] "Application Example 1"

[0836] (Claim 1)

[0837] A means for acquiring video data from multiple imaging devices arranged in the work area,

[0838] A means for analyzing the aforementioned video data in real time and recognizing the worker's movements and the operating status of the machinery,

[0839] A method for evaluating the degree of risk in the work environment based on past cases,

[0840] A notification system for issuing warnings to workers when a hazard is detected,

[0841] A means of sending a warning to a mobile device based on the analysis results and issuing a warning by voice or light,

[0842] A system that includes this.

[0843] (Claim 2)

[0844] The system according to claim 1, which provides workers with specific work instructions and guides them along evacuation routes based on a risk assessment.

[0845] (Claim 3)

[0846] The system according to claim 1, which notifies the site supervisor and transmits evacuation instructions and a safe assembly point to workers in the event of an emergency.

[0847] "Example 2 of combining an emotion engine"

[0848] (Claim 1)

[0849] An information processing device that collects video information obtained from an imaging device,

[0850] An information processing device that analyzes the aforementioned video information to identify the actions of the worker and the operation of the work machine,

[0851] An information processing device that analyzes the emotional state of workers and evaluates their stress and fatigue levels,

[0852] An information processing device that evaluates and records the degree of risk in the work environment based on the analysis results,

[0853] A terminal for issuing warnings to workers in response to detected hazards,

[0854] A means of notifying the manager of information according to the safety status of the workers,

[0855] A system that includes this.

[0856] (Claim 2)

[0857] The system according to claim 1, which proposes work adjustments based on the emotional state of the worker.

[0858] (Claim 3)

[0859] The system according to claim 1, which issues a notification including safe evacuation instructions in the event of an emergency.

[0860] "Application example 2 when combining with an emotional engine"

[0861] (Claim 1)

[0862] A means for acquiring video data from multiple imaging devices arranged in the work area,

[0863] A means for analyzing the aforementioned video data in real time and recognizing the worker's movements and the operating status of the machinery,

[0864] A means of analyzing the emotional state of workers using an emotion recognition engine and evaluating their stress levels and fatigue levels,

[0865] A method for evaluating the degree of risk in the work environment based on past cases,

[0866] A notification system for issuing warnings to workers when a hazard is detected,

[0867] A means for adjusting the operation of a machine based on the emotional state of the worker,

[0868] A system that includes this.

[0869] (Claim 2)

[0870] The system according to claim 1, which provides specific work instructions to workers based on a risk assessment and encourages breaks based on emotional recognition.

[0871] (Claim 3)

[0872] The system according to claim 1, which notifies the site supervisor when an emergency occurs, sends evacuation instructions to workers, and provides appropriate support based on their emotional state. [Explanation of Symbols]

[0873] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for acquiring video data from multiple imaging devices arranged in the work area, A means for analyzing the aforementioned video data in real time and recognizing the worker's movements and the operating status of the machinery, A method for evaluating the degree of risk in the work environment based on past cases, A notification system for issuing warnings to workers when a hazard is detected, A means of sending a warning to a mobile device based on the analysis results and issuing a warning by voice or light, A system that includes this.

2. The system according to claim 1, which provides workers with specific work instructions and guides them along evacuation routes based on a risk assessment.

3. The system according to claim 1, which notifies the site supervisor and transmits evacuation instructions and a safe assembly point to workers in the event of an emergency.

Images