system

Abstract

We provide the system. [Solution] A means of receiving video information acquired from a monitoring device in real time, Action recognition means for analyzing the movements and position of a person from the aforementioned video information, A determination means for determining whether the action analyzed by the action recognition means is abnormal, An alert issuing means generates and transmits warning information when the aforementioned discrimination means determines that an abnormality has occurred, 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 as a 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 a facility that monitors a wide range of multiple individuals simultaneously, it is physically difficult for the monitoring staff to constantly visually confirm all movements, and immediate response is required especially when dangerous operations are performed. However, there is a risk that serious accidents cannot be prevented due to false detection or delayed response. To solve this problem, a system utilizing highly accurate action recognition technology is necessary.

Means for Solving the Problems

[0005] This invention features a high-precision behavior recognition means that analyzes a person's movements and location in real time based on video information acquired from a monitoring device. This means quickly identifies abnormal movements, and in such cases, generates warning information using an alert sending means and immediately transmits it to the person in charge. Furthermore, it has an automatic notification function to external rescue personnel, enabling a rapid response. In addition, because it has a function to identify individual people based on video information, it supports the rapid identification of risks to specific individuals.

[0006] A "surveillance device" is a device consisting of cameras and sensors that acquire and transmit video information of a target area.

[0007] "Video information" refers to real-time or recorded image data acquired by surveillance equipment.

[0008] "Real-time" refers to processing and communication that takes place simultaneously with the time an event or process is currently underway.

[0009] "Human movement" refers to physical actions, including changes in a person's position, posture, and speed, as observed within video footage.

[0010] "Position" refers to the specific location or spatial coordinates of a person or object captured within the video information.

[0011] "Action recognition means" refers to an algorithm or software for identifying and analyzing a person's movements and location from video information.

[0012] A "discrimination means" is a process that compares the results of an action recognition means with pre-set criteria to determine whether it meets specific conditions.

[0013] "Warning information" refers to a notification or message generated to inform of a detected anomaly.

[0014] "The 'alert sending means' is a mechanism or process that notifies the monitoring person or an external agency of the detected abnormal information."

[0015] "The 'external rescue personnel' are specialized agencies or individuals such as medical, fire, and police that are called to respond to the scene in case of an emergency."

[0016] "The 'identification of individual persons' is a process of identifying each person in video information using behavior recognition means and distinguishing them from others."

Brief Explanation of Drawings

[0017] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment." [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment." [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment." [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment." [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment." [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment." [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment." [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment." [Figure 9] It shows an emotion map to which a plurality of emotions are mapped." [Figure 10] It shows an emotion map to which a plurality of emotions are mapped." [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1." [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0018] 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.

[0019] First, the language used in the following description will be explained.

[0020] 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 CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), and the like.

[0021] 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.

[0022] 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.

[0023] 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).

[0024] 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."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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".

[0038] This invention relates to a system for monitoring multiple individuals over a wide area, and consists of a monitoring device, a server, a terminal, and a user. The monitoring device transmits video information acquired within the target area to the server in real time. The server applies a high-precision behavior recognition algorithm and analyzes the movements and locations of individuals from the video information to quickly detect abnormal behavior.

[0039] As a concrete example, if a camera installed in the pool area of ​​a childcare facility is used as a monitoring device, the server will determine that a child suddenly running off the poolside is an abnormal action. When this action is detected, the server will promptly generate an alert and notify the terminal of the user responsible for monitoring.

[0040] The alerts displayed on the device include detailed information such as the time, location, and video clips of the incident, allowing users to immediately understand the specific situation. By checking the alerts, users can take appropriate and prompt action.

[0041] Furthermore, the system has a function that automatically contacts external rescue personnel in the event of a serious anomaly. For example, it can immediately contact the nearest ambulance, preventing delays in emergency response. This enhances safety within the monitored area and also reduces the burden on monitoring personnel.

[0042] The introduction of such a system will enable highly accurate and efficient monitoring of a large number of individuals across a wide area.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] The server receives video information transmitted from the monitoring device in real time. The video information is streamed and buffered for processing within the server.

[0046] Step 2:

[0047] The server divides the received video information into frames and executes an action recognition algorithm. This allows the server to identify the actions and positions of people within each frame.

[0048] Step 3:

[0049] The server uses facial recognition technology to identify individual people. This step allows the server to track each person's activity history and easily detect anomalies related to specific individuals.

[0050] Step 4:

[0051] The server analyzes the behavioral patterns and compares them to pre-configured dangerous behavioral patterns. If abnormal behavior is detected, the server registers it as a flag.

[0052] Step 5:

[0053] When the server detects an anomaly, it generates an alert. This alert includes the time and location of the anomaly, as well as identification information of the parties involved.

[0054] Step 6:

[0055] The server sends the generated alert to the assigned person's terminal. The alert is immediately notified and displayed as a pop-up on the terminal.

[0056] Step 7:

[0057] The device receives alerts in real time and displays notifications to the user. The user can then review the alert details and take prompt action on-site.

[0058] Step 8:

[0059] If a user detects a serious anomaly while responding to an alert, the server automatically sends a notification to external rescue personnel. This automated notification enables a rapid response to the emergency.

[0060] Step 9:

[0061] The server saves all alert information and response history to a database. This information is used for later analysis and report generation.

[0062] (Example 1)

[0063] 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."

[0064] In monitoring large numbers of people, there is a need to detect abnormal behavior with high accuracy and speed, reduce the workload on personnel, and respond quickly. Current technologies are insufficient to meet these requirements, and further efficiency improvements and automation are particularly important for systems that operate in real time.

[0065] 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.

[0066] In this invention, the server includes means for receiving video data in real time, means for analyzing motion and location, and means for determining anomalies. This enables highly accurate motion recognition and anomaly detection, and improves safety within the monitored area through rapid response.

[0067] A "monitoring device" is a device that captures images of the situation within a target area in real time, generates video data, and transmits it.

[0068] "Video data" refers to data containing visual information acquired by devices such as cameras.

[0069] "Receiving in real time" means that video data is transmitted directly to a server or other device with almost no delay after it is captured.

[0070] "Behavioral analysis means" refers to technologies and algorithms for processing received video data and analyzing the actions and positions of a subject.

[0071] An "anomaly detection method" is a means of identifying unusual behavior from analyzed operational data and identifying events that trigger warnings.

[0072] A "warning transmission means" is a means that has the function of generating warning information when an abnormality is detected and transmitting it to an external party.

[0073] "User's device" refers to electronic devices such as smartphones and tablets used to receive and notify warning information.

[0074] "External rescue personnel" refers to external organizations or individuals that have been pre-registered to deal with emergencies.

[0075] "Identification target" refers to a person or object identified through the analysis of video data.

[0076] This invention provides a system for efficiently monitoring a wide-area target, and includes monitoring equipment, a server, a terminal, and a user. The monitoring equipment acquires video data within the target area in real time and provides a foundation for analyzing people's movements and locations. This equipment typically uses network cameras capable of processing 1080p high-resolution video.

[0077] The server receives video data transmitted from monitoring devices and performs high-precision behavioral analysis on that data. This behavioral analysis is carried out using deep learning frameworks such as TENSORFLOW® or PyTorch. Through this analysis, the server analyzes the target's movements and position frame by frame and detects anomalies. For example, a sudden movement of running can be identified as an anomaly.

[0078] When an anomaly is detected, the server immediately generates warning information and sends a push notification to the device. This device is a smartphone or tablet carried by the user, with a dedicated application installed. The warning information includes the time the anomaly occurred, location information, and a link to a relevant video clip, allowing the user to quickly understand the situation and take immediate action.

[0079] In the event of a serious anomaly, the server has a function to automatically contact external rescue personnel. By making an emergency call using VoIP, it is possible to request assistance immediately.

[0080] For example, when used in a swimming pool at a childcare facility, the server would immediately detect any dangerous behavior by children at the poolside and transmit appropriate warnings to the user in charge of management.

[0081] An example of a prompt used in a generative AI model is, "Please explain the specific method of anomaly detection using a highly accurate behavior recognition algorithm." This system improves security within the monitoring area and streamlines monitoring operations.

[0082] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0083] Step 1:

[0084] The server receives video data transmitted in real time from the monitoring device. The input is digital video data, and the output is stored internally as frames of this data. The server receives this as streaming data and prepares it for the next analysis process.

[0085] Step 2:

[0086] The server performs behavioral analysis on the received video data. The input is the previously received video frames, and the output is the analyzed motion and location information. The server uses a deep learning framework to analyze the person's movements in real time and identify individual movements. In this process, a model using TensorFlow analyzes the movement patterns and classifies each movement in preparation for anomaly detection.

[0087] Step 3:

[0088] The server identifies anomalies based on the analysis results. The input is the result of the behavioral analysis, and the output is a determination of whether or not an anomaly has occurred. The server uses an anomaly detection algorithm to identify behavior that deviates from normal behavioral patterns. For example, suddenly starting to run is generally considered an anomaly, and this determination is passed on to the next warning generation process.

[0089] Step 4:

[0090] The server immediately generates warning information and sends a push notification to the user's device upon detecting an anomaly. The input is the result of the anomaly detection, and the output is the warning notification sent to the user's device. The warning information is generated as an alert message and includes the time and location information of the anomaly, contributing to a rapid response.

[0091] Step 5:

[0092] The user checks the alert notification received on their device. The input is the warning message received on the device, and the output is the user's action based on the alert. Based on the information received, the user assesses the situation on site and, if necessary, rushes to the site or takes other countermeasures.

[0093] Step 6:

[0094] The server automatically contacts external rescue personnel when a critical anomaly is detected. The input is the determination of a critical anomaly, and the output is the notification to the external party. By using the VoIP protocol to make emergency contacts quickly, problems can be reported with minimal time lag, enabling immediate response.

[0095] (Application Example 1)

[0096] 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."

[0097] There is a need to detect abnormal behavior in real time within a wide-ranging monitoring area and to improve security quickly and efficiently. However, current systems have a lack of responsiveness due to insufficient rapid information dissemination after anomaly detection and inadequate notification to responders such as security guards.

[0098] 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.

[0099] In this invention, the server includes means for processing video information acquired from a monitoring device, behavior recognition means for analyzing the movement and position of a target from the video information, and discrimination means for determining whether the movement analyzed by the behavior recognition means is abnormal. This makes it possible to automatically detect abnormal behavior and quickly send warning information to the responder. Furthermore, notifications can be visually presented via a wearable device to enhance responsiveness.

[0100] A "surveillance device" is a device used to acquire video information within a target area.

[0101] "Visual information" refers to visual data acquired by surveillance equipment, including the movements and positions of individuals.

[0102] "Action recognition means" refers to a means for analyzing the movement and position of an object from video information and recognizing a specific action.

[0103] "Discrimination means" refers to means for determining whether or not the behavior analyzed by the behavior recognition means is abnormal.

[0104] An "alert generation means" is a means for generating warning information when the discrimination means determines that an abnormality has occurred, and for sending it to a predetermined recipient.

[0105] A "wearable device" is a device worn on the body to display information, and has the function of visually displaying warning information.

[0106] The system used to realize this application consists of a monitoring device, a server, and a wearable device. The server receives video information acquired from the monitoring device and analyzes the video information using a high-precision behavior recognition algorithm (e.g., a TensorFlow model) on a data processing platform (e.g., AWS® Lambda) that runs on the cloud.

[0107] The server recognizes the target's movements and location from video information and determines whether abnormal behavior is present. During this process, it uses computer vision technologies such as OpenCV to process the data and analyze behavioral patterns. If an anomaly is detected, the server generates an alert and sends warning information to the responder's wearable device via a messaging platform such as Firebase.

[0108] As a concrete example, in a large shopping mall, security guards could wear smart glasses and utilize a surveillance system to enhance security within the commercial facility. If an individual exhibiting suspicious behavior is detected within the commercial facility, an alert would immediately appear on the security guard's smart glasses, allowing the guard to quickly verify and respond, thereby preventing crime.

[0109] An example of a prompt message for a generated AI model could be, "Detect any suspicious activity within the commercial facility and promptly notify security personnel." This would help improve security.

[0110] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0111] Step 1:

[0112] The server receives video information from the monitoring device in real time. The video information becomes the input, and at this stage it is still raw video data. The server temporarily stores this data and prepares it for the next processing step.

[0113] Step 2:

[0114] The server preprocesses the received video information using OpenCV, a computer vision technology. This adjusts the image resolution and removes noise, converting it into a format suitable as input data for the action recognition model.

[0115] Step 3:

[0116] Based on pre-processed video information, the server performs motion analysis using an action recognition model trained with TensorFlow. The input for the analysis is pre-processed image data, and the output is motion and position information for each frame.

[0117] Step 4:

[0118] Based on the analysis results, the server performs anomaly detection using a generated AI model. In this step, current behavior data against past normal behavior patterns is used as input, and the presence or absence of abnormal behavior is determined as output.

[0119] Step 5:

[0120] If the server detects abnormal operation, it uses Firebase to generate alert information and sends it to the notification device. The input is the result of the abnormality detection, and the output generates a warning message to be sent to the wearable device.

[0121] Step 6:

[0122] The terminal displays received alerts on a wearable device. The input is alert information, and the output is a visual notification on the device. The user reviews this and takes prompt action on-site as needed.

[0123] 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.

[0124] This invention begins with transmitting video information acquired from a target area in real time to a server using a monitoring device. The server receives this video information and applies an action recognition means to analyze the movements and positions of people in each frame. Based on the analysis, a discrimination means determines whether there are any abnormalities in the movements. This process also includes a function to perform facial recognition of people and enable individual identification.

[0125] Furthermore, in this invention, by combining an emotion engine, the server also analyzes the user's emotional state from the video information. This process makes it possible to recognize emotions such as excitement, stress, and anxiety from subtle changes in the user's facial expressions and behavior. The server integrates this emotional information and adjusts the alert content when an anomaly is detected. For example, if a highly urgent emotional state is recognized, a more attention-grabbing alert is generated.

[0126] Meanwhile, the device receives alerts sent from the server and displays them to the user as pop-up notifications. Specifically, considering a situation by a swimming pool, if a child makes a movement that suggests they might drown, the device analyzes the child's emotional state along with the movement. If there are signs of stress or fear, a high-priority alert is immediately displayed on the device.

[0127] Users can check alerts on their devices and take quick action based on the details, including emotional information. The server also has an option to automatically notify external rescue personnel of the detected information, triggering a rapid external response if necessary.

[0128] Thus, the system of the present invention achieves a more precise and responsive monitoring system by integrating behavioral and emotional recognition.

[0129] The following describes the processing flow.

[0130] Step 1:

[0131] The server receives video information in real time from the monitoring device. This information is converted to an appropriate format within the server for analysis.

[0132] Step 2:

[0133] The server breaks down the received video information frame by frame and applies an action recognition algorithm. This allows for a detailed analysis of the movements and positions of the people appearing in each frame.

[0134] Step 3:

[0135] The server uses an emotion engine to analyze a person's facial expressions and body language from video information to infer their emotional state. This analysis determines whether the person is experiencing excitement, stress, anxiety, or other emotions.

[0136] Step 4:

[0137] The server integrates the results of behavior recognition and emotion analysis, and uses a discrimination mechanism to evaluate whether or not there is abnormal behavior. If an abnormality is detected, it is flagged as an escalated abnormality.

[0138] Step 5:

[0139] When a server is detected as abnormal, an alert system generates warning information. The alert includes the time of detection, location, and urgency level based on the emotional state of the person affected.

[0140] Step 6:

[0141] The server generates alert information and sends it to the assigned person's terminal. The terminal receives this information and displays it as a notification to the user.

[0142] Step 7:

[0143] The user checks the alert from their device. The alert provides a description of the situation and details of their emotional state, which the user uses to decide how to respond.

[0144] Step 8:

[0145] If a user determines that a serious anomaly or emergency has occurred, an external rescue team will be immediately contacted, either manually or through the server's automated functions.

[0146] Step 9:

[0147] The server saves logs of these processes to a database, which are then used for later analysis and report generation. The saved data helps improve the system and enhance future anomaly detection capabilities.

[0148] (Example 2)

[0149] 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".

[0150] Conventional surveillance systems focus on detecting anomalies based on human movement, making it difficult to grasp the overall situation. Furthermore, because they do not consider emotional states, the appropriateness and speed of responses to emergencies may be insufficient. Therefore, there is a need for systems that can meticulously assess anomalies from video information and issue optimal warnings that also take emotional states into account.

[0151] 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.

[0152] In this invention, the server includes means for receiving visual information acquired from a video recording device in real time, motion recognition means for analyzing human movement and position from the visual information, and evaluation means. This makes it possible to determine with high accuracy whether there is an abnormality in a person's movement, and to adjust warning information more appropriately using emotion estimation means.

[0153] A "video recording device" is a device that acquires visual information from a target area and has the function of collecting data in real time and transmitting it to a server.

[0154] "Visual information" refers to digital images and video data acquired by video recording equipment, and is the subject of analysis.

[0155] "Motion recognition means" refers to a method or device that analyzes human movement and position from visual information to detect specific movement patterns.

[0156] "Evaluation means" refers to a method or device for determining whether the motion analyzed by the motion recognition means is abnormal.

[0157] "Instruction transmission means" refers to a method or device for generating warning information when the evaluation means detects an abnormality and transmitting it to other devices or users.

[0158] An "emotion inference method" is a technique or device that analyzes a person's emotional state from visual information and uses the results to make situational judgments or adjust warning information.

[0159] Modes for carrying out the invention

[0160] This system consists of a server, terminals, and users working together. The server plays a central role in receiving visual information acquired from video recording equipment in real time and performing various analyses. Specifically, the server receives visual information and analyzes human movement and position using motion recognition software. In this process, a server equipped with a high-performance GPU is used as hardware, while general motion recognition tools and AI models are used as software. For example, custom models based on TensorFlow or PyTorch are common.

[0161] Furthermore, the server analyzes the user's emotional state using emotion estimation tools such as Microsoft Cognitive Services and open-source facial expression analysis tools. This allows the system to estimate the user's emotions not only from their actions but also from their facial expressions, and adjust the urgency of warning information as needed. Other hardware components include large-capacity storage and network equipment that enables high-speed data communication.

[0162] The terminal receives instructions from the server and conveys them intuitively to the user. For example, mobile devices such as smartphones and tablets are often used as terminals. These terminals send warnings to the user through pop-up notifications and audio alerts.

[0163] Users can check warnings from their devices and take appropriate action as needed. For example, if an anomaly is detected in a monitored area, users are required to take action in accordance with the warning in order to respond quickly.

[0164] For example, when this system is used for poolside monitoring, the server detects children exhibiting drowning-like behavior based on visual information obtained from video recording equipment. Furthermore, emotion inference measures identify fear and stress from the child's facial expressions, and a highly urgent warning generated based on this information is immediately displayed on the terminal. This allows the user to take swift action.

[0165] An example of a prompt for a generating AI model is the instruction, "Build a system that analyzes video data from a monitoring device, identifies specific actions and emotions, and generates warnings."

[0166] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0167] Step 1:

[0168] The server receives visual information in real time from video recording equipment. The input is raw data captured by the camera. The server converts this data into a format that is easy to analyze and prepares it for analysis. Specifically, this involves decompressing the data and splitting it into frames.

[0169] Step 2:

[0170] The server applies a motion recognition model to extract the person's movements and position from the received visual information. The input used is the video frame formatted in step 1. The server performs image recognition frame by frame through the AI ​​model, analyzes the motion patterns, and outputs the results as motion data. Specifically, actions such as where a person walks or raises their hand are recognized.

[0171] Step 3:

[0172] The server performs an evaluation to detect anomalies based on the operational data. The input is the operational data extracted in step 2. The server compares this data with a pre-configured operational pattern to determine whether it is normal or abnormal. If an abnormality is detected, alert data is generated. In this process, the number of abnormal operations that occurred within a specific time period is also taken into consideration.

[0173] Step 4:

[0174] The server analyzes a person's emotional state from visual information. To do this, it applies an emotion prediction model to evaluate the person's facial expressions. Visual information is used as input. The server performs emotion prediction and outputs emotion labels such as "joy," "fear," and "surprise." This process, performed simultaneously with motion detection, results in more comprehensive analysis.

[0175] Step 5:

[0176] The server integrates the generated behavioral data and emotional information to prepare for generating alert information. The input is the output data from steps 3 and 4. The server determines the urgency and priority of the anomaly and outputs it as an alert message. For example, if a child is drowning in a pool while also showing signs of fear, a high-urgency alert will be generated.

[0177] Step 6:

[0178] The device receives alerts sent from the server and displays a warning to the user. The input is the alert data from the server. Based on the notification settings, the device outputs a pop-up notification or an audio alert. This makes it possible to immediately attract the user's attention.

[0179] Step 7:

[0180] The user checks the alert displayed on the device and takes the necessary action. The input is the alert displayed on the device. The user is required to understand the content of the alert and its details, and take appropriate action, for example, in the case of an emergency. Specific actions may include reporting an anomaly or rushing to the scene.

[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] Modern surveillance systems focus on detecting anomalies based on human movement, but methods for detecting abnormal behavior accompanied by emotional changes are underdeveloped. Therefore, from a security and safety perspective, there is a need for an integrated surveillance system that can quickly identify anomalies not only in movement but also in emotion, and prompt immediate action.

[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 receiving video information acquired from a monitoring device in real time, means for recognizing a person's movements and position from the video information, and means for recognizing a person's emotional state from the video information. This enables immediate detection of abnormalities in movements and emotions, and allows for immediate response via smart glasses.

[0186] A "monitoring device" is a device that acquires video information from a target area and has the function of transmitting this information to a server in real time.

[0187] "Video information" refers to image and video data acquired by surveillance devices, and is the information that is subject to analysis.

[0188] "Action recognition means" refers to means that have the function of analyzing a person's movements and position from video information.

[0189] A "discrimination means" is a means of determining whether the behavior analyzed by the behavior recognition means is abnormal.

[0190] An "alert transmission means" is a means that has the function of generating and transmitting warning information when the discrimination means determines that an abnormality has occurred.

[0191] An "emotion recognition means" is a means that has the function of analyzing a person's emotional state from video information.

[0192] "Warning information" refers to information that includes the content of alerts generated when an anomaly is detected.

[0193] "Smart glasses" are glasses-type information terminals worn by security guards that function as display devices for showing warning information.

[0194] This invention relates to a surveillance system applied to security services. The server receives video information acquired from a target area by a monitoring device and analyzes the movements and positions of people in the video using an action recognition means. At this time, it uses an image processing library such as OpenCV to detect movements frame by frame and uses a discrimination means to determine whether there is an anomaly. Furthermore, the server, as an emotion recognition means, uses a generative AI model to analyze emotional states from facial expressions and recognizes emotions such as stress and anxiety. If an anomaly is detected as a result, it uses an alert sending means to generate warning information and transmit it to a terminal including smart glasses.

[0195] The terminal displays received warning information to the user as a pop-up notification, prompting a quick response. This pop-up provides a visual alert via the head-up display of a glasses-type information terminal. For example, in a shopping mall's surveillance system, if a person is repeatedly opening and closing their bag in an unusual manner, and the analysis indicates they are experiencing tension or anxiety, this information is immediately notified to security guards. An example prompt message might be: "This person is repeatedly opening and closing their bag, and the emotion analysis indicates they are experiencing tension. Please investigate the scene."

[0196] Real-time performance is crucial in the design of this system, as it allows servers to efficiently process data and generate appropriate alerts, enabling the implementation of swift and effective security measures.

[0197] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0198] Step 1:

[0199] The server receives video information of the target area in real time from the monitoring device. This video information is video data transmitted from the camera, and each frame is the subject of analysis. The input is video data from the camera, and the output is image frames ready for analysis. Specifically, stream-formatted data is captured as images.

[0200] Step 2:

[0201] The server analyzes the movement and position of individuals from the received video information. To this end, it utilizes OpenCV as a motion recognition tool, applying an algorithm to identify individuals within frames and track their movements. The input is video data from each frame, and the output is analytical information regarding movement and position. Specifically, image processing techniques are used to detect the joints and positions of each individual.

[0202] Step 3:

[0203] The server identifies abnormal behavior based on the results of behavior recognition. If a specific behavior pattern is deemed abnormal, the discrimination means detects it. The input is the analysis information from step 2, and the output is the result of the abnormality determination. Specifically, the degree of abnormality is evaluated by comparing it with a pre-set standard.

[0204] Step 4:

[0205] The server uses emotion recognition to analyze a person's emotional state from video information. Emotions such as stress and fear are identified through facial expression analysis using a generative AI model. The input is image data of a person's face, and the output is the result of the emotional state analysis. Specifically, the emotion analysis model is applied to quantify the degree of emotion.

[0206] Step 5:

[0207] The server generates an alert and sends it to the smart glasses when it detects abnormal behavior or emotion. The generated warning information is adjusted according to the urgency of the alert. The input is the result from steps 3 and 4, and the output is the warning message. Specifically, a prompt message is automatically generated according to the urgency.

[0208] Step 6:

[0209] The device receives alerts from the server and displays them to the user as pop-up notifications. The displayed information includes details of the detected anomaly and the necessary corrective actions. The input is warning information from the server, and the output is a visual notification to the user. Specifically, the alert is displayed on the smart glasses' display, which the user acknowledges.

[0210] This system enables swift safety measures based on real-time detection of abnormal behavior and emotions.

[0211] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0212] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include 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.

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

[0214] [Second Embodiment]

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

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

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

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

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

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

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

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

[0223] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

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

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

[0226] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0227] This invention relates to a system for monitoring multiple individuals over a wide area, and consists of a monitoring device, a server, a terminal, and a user. The monitoring device transmits video information acquired within the target area to the server in real time. The server applies a high-precision behavior recognition algorithm and analyzes the movements and locations of individuals from the video information to quickly detect abnormal behavior.

[0228] For example, if a camera installed in the pool area of ​​a childcare facility is used as a monitoring device, the server will detect if a child suddenly starts running by the poolside as an abnormal movement. When this movement is detected, the server will promptly generate an alert and notify the terminal of the user responsible for monitoring.

[0229] The alerts displayed on the device include detailed information such as the time, location, and video clips of the incident, allowing users to immediately understand the specific situation. By checking the alerts, users can take appropriate and prompt action.

[0230] Furthermore, the system has a function that automatically contacts external rescue personnel in the event of a serious anomaly. For example, it can immediately contact the nearest ambulance, preventing delays in emergency response. This enhances safety within the monitored area and also reduces the burden on monitoring personnel.

[0231] The introduction of such a system will enable highly accurate and efficient monitoring of a large number of individuals across a wide area.

[0232] The following describes the processing flow.

[0233] Step 1:

[0234] The server receives video information transmitted from the monitoring device in real time. The video information is streamed and buffered for processing within the server.

[0235] Step 2:

[0236] The server divides the received video information into frames and executes an action recognition algorithm. This allows the server to identify the actions and positions of people within each frame.

[0237] Step 3:

[0238] The server uses facial recognition technology to identify individual people. This step allows the server to track each person's activity history and easily detect anomalies related to specific individuals.

[0239] Step 4:

[0240] The server analyzes the behavioral patterns and compares them to pre-configured dangerous behavioral patterns. If abnormal behavior is detected, the server registers it as a flag.

[0241] Step 5:

[0242] When the server detects an anomaly, it generates an alert. This alert includes the time and location of the anomaly, as well as identification information of the parties involved.

[0243] Step 6:

[0244] The server sends the generated alert to the assigned person's terminal. The alert is immediately notified and displayed as a pop-up on the terminal.

[0245] Step 7:

[0246] The device receives alerts in real time and displays notifications to the user. The user can then review the alert details and take prompt action on-site.

[0247] Step 8:

[0248] If a user detects a serious anomaly while responding to an alert, the server automatically sends a notification to external rescue personnel. This automated notification enables a rapid response to the emergency.

[0249] Step 9:

[0250] The server saves all alert information and response history to a database. This information is used for later analysis and report generation.

[0251] (Example 1)

[0252] 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."

[0253] In monitoring large numbers of people, there is a need to detect abnormal behavior with high accuracy and speed, reduce the workload on personnel, and respond quickly. Current technologies are insufficient to meet these requirements, and further efficiency improvements and automation are particularly important for systems that operate in real time.

[0254] 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.

[0255] In this invention, the server includes means for receiving video data in real time, means for analyzing motion and location, and means for determining anomalies. This enables highly accurate motion recognition and anomaly detection, and improves safety within the monitored area through rapid response.

[0256] A "monitoring device" is a device that captures images of the situation within a target area in real time, generates video data, and transmits it.

[0257] "Video data" refers to data containing visual information acquired by devices such as cameras.

[0258] "Receiving in real time" means that video data is transmitted directly to a server or other device with almost no delay after it is captured.

[0259] "Behavioral analysis means" refers to technologies and algorithms for processing received video data and analyzing the actions and positions of a subject.

[0260] An "anomaly detection method" is a means of identifying unusual behavior from analyzed operational data and identifying events that trigger warnings.

[0261] A "warning transmission means" is a means that has the function of generating warning information when an abnormality is detected and transmitting it to an external party.

[0262] "User's device" refers to electronic devices such as smartphones and tablets used to receive and notify warning information.

[0263] "External rescue personnel" refers to external organizations or individuals that have been pre-registered to deal with emergencies.

[0264] "Identification target" refers to a person or object identified through the analysis of video data.

[0265] This invention provides a system for efficiently monitoring a wide-area target, and includes monitoring equipment, a server, a terminal, and a user. The monitoring equipment acquires video data within the target area in real time and provides a foundation for analyzing people's movements and locations. This equipment typically uses network cameras capable of processing 1080p high-resolution video.

[0266] The server receives video data transmitted from monitoring devices and performs high-precision behavioral analysis on that data. This behavioral analysis is carried out using deep learning frameworks such as TensorFlow or PyTorch. Through this analysis, the server analyzes the target's movements and position frame by frame and detects anomalies. For example, a sudden movement of running can be identified as an anomaly.

[0267] When an anomaly is detected, the server immediately generates warning information and sends a push notification to the device. This device is a smartphone or tablet carried by the user, with a dedicated application installed. The warning information includes the time the anomaly occurred, location information, and a link to a relevant video clip, allowing the user to quickly understand the situation and take immediate action.

[0268] In the event of a serious anomaly, the server has a function to automatically contact external rescue personnel. By making an emergency call using VoIP, it is possible to request assistance immediately.

[0269] For example, when used in a swimming pool at a childcare facility, the server would immediately detect any dangerous behavior by children at the poolside and transmit appropriate warnings to the user in charge of management.

[0270] An example of a prompt used in a generative AI model is, "Please explain the specific method of anomaly detection using a highly accurate behavior recognition algorithm." This system improves security within the monitoring area and streamlines monitoring operations.

[0271] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0272] Step 1:

[0273] The server receives video data transmitted in real time from the monitoring device. The input is digital video data, and the output is stored internally as frames of this data. The server receives this as streaming data and prepares it for the next analysis process.

[0274] Step 2:

[0275] The server performs behavioral analysis on the received video data. The input is the previously received video frames, and the output is the analyzed motion and location information. The server uses a deep learning framework to analyze the person's movements in real time and identify individual movements. In this process, a model using TensorFlow analyzes the movement patterns and classifies each movement in preparation for anomaly detection.

[0276] Step 3:

[0277] The server determines anomalies based on the results of analysis. The input is the result of behavior analysis, and the output is a determination of whether there is an anomaly or not. The server uses an anomaly detection algorithm to identify behaviors that deviate from normal operating patterns. For example, a sudden start behavior is generally regarded as an anomaly, and this determination is passed on to the next warning generation step.

[0278] Step 4:

[0279] When an anomaly is detected, the server immediately generates warning information and sends a push notification to the user's terminal. The input is the result of the anomaly determination, and the output is a warning notification to the user terminal. The warning information is generated as an alert message and includes the time of occurrence of the anomaly and location information, which contributes to a prompt response.

[0280] Step 5:

[0281] The user checks the alert notification received on the terminal. The input is the warning message that arrived at the terminal, and the output is the user's action based on the alert. The user judges the situation at the scene based on the notified information and takes emergency action to the scene or other countermeasures as necessary.

[0282] Step 6:

[0283] When a major anomaly is detected, the server automatically contacts external rescue personnel. The input is the determination of a major anomaly, and the output is the contact to the outside. By using the VoIP protocol to quickly make an emergency contact, the problem is reported without any time lag and immediate response is enabled.

[0284] (Application Example 1)

[0285] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0286] There is a demand to detect abnormal behavior in a wide monitoring area in real time and improve safety quickly and efficiently. On the other hand, in the current system, there is a problem of lack of responsiveness because rapid information communication after abnormal detection and notification to responders such as security guards are insufficient.

[0287] 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.

[0288] In this invention, the server includes means for processing video information acquired from a monitoring device, behavior recognition means for analyzing the actions and positions of objects from the video information, and discrimination means for determining whether the actions analyzed by the behavior recognition means are abnormal. Thereby, it becomes possible to automatically detect abnormal behavior and quickly transmit warning information to responders. In addition, notifications can be visually presented via a wearable device, enhancing responsiveness.

[0289] The "monitoring device" is a device for acquiring video information within a target area.

[0290] The "video information" is visual data acquired by a monitoring device and includes the actions and positions of people.

[0291] The "behavior recognition means" is means for analyzing the actions and positions of objects from video information and recognizing specific behaviors.

[0292] The "discrimination means" is means for determining whether the actions analyzed by the behavior recognition means are abnormal.

[0293] The "alert transmission means" is means for generating warning information when the discrimination means determines an abnormality and transmitting it to a predetermined notification destination.

[0294] The "wearable device" is a device worn on the body for presenting information and has a function of visually displaying warning information.

[0295] The system used to realize this application consists of a monitoring device, a server, and a wearable device. The server receives video information acquired from the monitoring device and analyzes the video information using a high-precision behavior recognition algorithm (e.g., a TensorFlow model) on a data processing platform (e.g., AWS Lambda) running on the cloud.

[0296] The server recognizes the target's movements and location from video information and determines whether abnormal behavior is present. During this process, it uses computer vision technologies such as OpenCV to process the data and analyze behavioral patterns. If an anomaly is detected, the server generates an alert and sends warning information to the responder's wearable device via a messaging platform such as Firebase.

[0297] As a concrete example, in a large shopping mall, security guards could wear smart glasses and utilize a surveillance system to enhance security within the commercial facility. If an individual exhibiting suspicious behavior is detected within the commercial facility, an alert would immediately appear on the security guard's smart glasses, allowing the guard to quickly verify and respond, thereby preventing crime.

[0298] An example of a prompt message for a generated AI model could be, "Detect any suspicious activity within the commercial facility and promptly notify security personnel." This would help improve security.

[0299] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0300] Step 1:

[0301] The server receives video information from the monitoring device in real time. The video information becomes the input, and at this stage it is still raw video data. The server temporarily stores this data in preparation for the next processing step.

[0302] Step 2:

[0303] The server preprocesses the received video information using OpenCV, which is computer vision technology. As a result, the resolution of the image is adjusted and noise is removed, and it is converted into a format suitable as input data for the action recognition model.

[0304] Step 3:

[0305] Based on the preprocessed video information, the server performs motion analysis using an action recognition model trained by TensorFlow. The input for the analysis is the preprocessed image data, and the output is the motion and position information for each frame.

[0306] Step 4:

[0307] Based on the analysis results, the server performs anomaly detection using a generative AI model. In this step, the current motion data with respect to past normal behavior patterns is used as input, and the presence or absence of abnormal behavior is determined as output.

[0308] Step 5:

[0309] When the server detects an abnormal operation, it generates alert information using Firebase and sends it to the notification terminal. The input is the result of the anomaly detection, and as output, a warning message to be sent to the wearable device is generated.

[0310] Step 6:

[0311] The terminal displays the received alert on the wearable device. The input is the alert information, and the output is a visual notification on the device. The user checks this and takes prompt action on-site if necessary.

[0312] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion recognition model 59 and perform specific processing using the user's emotion.

[0313] This invention begins with transmitting video information acquired from a target area in real time to a server using a monitoring device. The server receives this video information and applies an action recognition means to analyze the movements and positions of people in each frame. Based on the analysis, a discrimination means determines whether there are any abnormalities in the movements. This process also includes a function to perform facial recognition of people and enable individual identification.

[0314] Furthermore, in this invention, by combining an emotion engine, the server also analyzes the user's emotional state from the video information. This process makes it possible to recognize emotions such as excitement, stress, and anxiety from subtle changes in the user's facial expressions and behavior. The server integrates this emotional information and adjusts the alert content when an anomaly is detected. For example, if a highly urgent emotional state is recognized, a more attention-grabbing alert is generated.

[0315] Meanwhile, the device receives alerts sent from the server and displays them to the user as pop-up notifications. Specifically, considering a situation by a swimming pool, if a child makes a movement that suggests they might drown, the device analyzes the child's emotional state along with the movement. If there are signs of stress or fear, a high-priority alert is immediately displayed on the device.

[0316] Users can check alerts on their devices and take quick action based on the details, including emotional information. The server also has an option to automatically notify external rescue personnel of the detected information, triggering a rapid external response if necessary.

[0317] Thus, the system of the present invention achieves a more precise and responsive monitoring system by integrating behavioral and emotional recognition.

[0318] The following describes the processing flow.

[0319] Step 1:

[0320] The server receives video information in real time from the monitoring device. This information is converted to an appropriate format within the server for analysis.

[0321] Step 2:

[0322] The server breaks down the received video information frame by frame and applies an action recognition algorithm. This allows for a detailed analysis of the movements and positions of the people appearing in each frame.

[0323] Step 3:

[0324] The server uses an emotion engine to analyze a person's facial expressions and body language from video information to infer their emotional state. This analysis determines whether the person is experiencing excitement, stress, anxiety, or other emotions.

[0325] Step 4:

[0326] The server integrates the results of behavior recognition and emotion analysis, and uses a discrimination mechanism to evaluate whether or not there is abnormal behavior. If an abnormality is detected, it is flagged as an escalated abnormality.

[0327] Step 5:

[0328] When a server is detected as abnormal, an alert system generates warning information. The alert includes the time of detection, location, and urgency level based on the emotional state of the person affected.

[0329] Step 6:

[0330] The server generates alert information and sends it to the assigned person's terminal. The terminal receives this information and displays it as a notification to the user.

[0331] Step 7:

[0332] The user checks the alert from their device. The alert provides a description of the situation and details of their emotional state, which the user uses to decide how to respond.

[0333] Step 8:

[0334] If a user determines that a serious anomaly or emergency has occurred, an external rescue team will be immediately contacted, either manually or through the server's automated functions.

[0335] Step 9:

[0336] The server saves logs of these processes to a database, which are then used for later analysis and report generation. The saved data helps improve the system and enhance future anomaly detection capabilities.

[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] Conventional surveillance systems focus on detecting anomalies based on human movement, making it difficult to grasp the overall situation. Furthermore, because they do not consider emotional states, the appropriateness and speed of responses to emergencies may be insufficient. Therefore, there is a need for systems that can meticulously assess anomalies from video information and issue optimal warnings that also take emotional states into account.

[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 means for receiving visual information acquired from a video recording device in real time, motion recognition means for analyzing human movement and position from the visual information, and evaluation means. This makes it possible to determine with high accuracy whether there is an abnormality in a person's movement, and to adjust warning information more appropriately using emotion estimation means.

[0342] A "video recording device" is a device that acquires visual information from a target area and has the function of collecting data in real time and transmitting it to a server.

[0343] "Visual information" refers to digital images and video data acquired by video recording equipment, and is the subject of analysis.

[0344] "Motion recognition means" refers to a method or device that analyzes human movement and position from visual information to detect specific movement patterns.

[0345] "Evaluation means" refers to a method or device for determining whether the motion analyzed by the motion recognition means is abnormal.

[0346] "Instruction transmission means" refers to a method or device for generating warning information when the evaluation means detects an abnormality and transmitting it to other devices or users.

[0347] An "emotion inference method" is a technique or device that analyzes a person's emotional state from visual information and uses the results to make situational judgments or adjust warning information.

[0348] Modes for carrying out the invention

[0349] This system consists of a server, terminals, and users working together. The server plays a central role in receiving visual information acquired from video recording equipment in real time and performing various analyses. Specifically, the server receives visual information and analyzes human movement and position using motion recognition software. In this process, a server equipped with a high-performance GPU is used as hardware, while general motion recognition tools and AI models are used as software. For example, custom models based on TensorFlow or PyTorch are common.

[0350] Furthermore, the server analyzes the user's emotional state using emotion estimation tools such as Microsoft Cognitive Services and open-source facial expression analysis tools. This allows it to estimate the user's emotions not only from their actions but also from their facial expressions, and adjust the urgency of warning information as needed. Other hardware includes large-capacity storage and network equipment that enables high-speed data communication.

[0351] The terminal receives instructions from the server and conveys them intuitively to the user. For example, mobile devices such as smartphones and tablets are often used as terminals. These terminals send warnings to the user through pop-up notifications and audio alerts.

[0352] Users can check warnings from their devices and take appropriate action as needed. For example, if an anomaly is detected in a monitored area, users are required to take action in accordance with the warning in order to respond quickly.

[0353] For example, when this system is used for poolside monitoring, the server detects children exhibiting drowning-like behavior based on visual information obtained from video recording equipment. Furthermore, emotion inference measures identify fear and stress from the child's facial expressions, and a highly urgent warning generated based on this information is immediately displayed on the terminal. This allows the user to take swift action.

[0354] An example of a prompt for a generating AI model is the instruction, "Build a system that analyzes video data from a monitoring device, identifies specific actions and emotions, and generates warnings."

[0355] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0356] Step 1:

[0357] The server receives visual information in real time from video recording equipment. The input is raw data captured by the camera. The server converts this data into a format that is easy to analyze and prepares it for analysis. Specifically, this involves decompressing the data and splitting it into frames.

[0358] Step 2:

[0359] The server applies a motion recognition model to extract the person's movements and position from the received visual information. The input used is the video frame formatted in step 1. The server performs image recognition frame by frame through the AI ​​model, analyzes the motion patterns, and outputs the results as motion data. Specifically, actions such as where a person walks or raises their hand are recognized.

[0360] Step 3:

[0361] The server performs an evaluation to detect anomalies based on the operational data. The input is the operational data extracted in step 2. The server compares this data with a pre-configured operational pattern to determine whether it is normal or abnormal. If an abnormality is detected, alert data is generated. In this process, the number of abnormal operations that occurred within a specific time period is also taken into consideration.

[0362] Step 4:

[0363] The server analyzes a person's emotional state from visual information. To do this, it applies an emotion prediction model to evaluate the person's facial expressions. Visual information is used as input. The server performs emotion prediction and outputs emotion labels such as "joy," "fear," and "surprise." This process, performed simultaneously with motion detection, results in more comprehensive analysis.

[0364] Step 5:

[0365] The server integrates the generated behavioral data and emotional information to prepare for generating alert information. The input is the output data from steps 3 and 4. The server determines the urgency and priority of the anomaly and outputs it as an alert message. For example, if a child is drowning in a pool while also showing signs of fear, a high-urgency alert will be generated.

[0366] Step 6:

[0367] The device receives alerts sent from the server and displays a warning to the user. The input is the alert data from the server. Based on the notification settings, the device outputs a pop-up notification or an audio alert. This makes it possible to immediately attract the user's attention.

[0368] Step 7:

[0369] The user checks the alert displayed on the device and takes the necessary action. The input is the alert displayed on the device. The user is required to understand the content of the alert and its details, and take appropriate action, for example, in the case of an emergency. Specific actions may include reporting an anomaly or rushing to the scene.

[0370] (Application Example 2)

[0371] 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."

[0372] Modern surveillance systems focus on detecting anomalies based on human movement, but methods for detecting abnormal behavior accompanied by emotional changes are underdeveloped. Therefore, from a security and safety perspective, there is a need for an integrated surveillance system that can quickly identify anomalies not only in movement but also in emotion, and prompt immediate action.

[0373] 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.

[0374] In this invention, the server includes means for receiving video information acquired from a monitoring device in real time, means for recognizing a person's movements and position from the video information, and means for recognizing a person's emotional state from the video information. This enables immediate detection of abnormalities in movements and emotions, and allows for immediate response via smart glasses.

[0375] A "monitoring device" is a device that acquires video information from a target area and has the function of transmitting this information to a server in real time.

[0376] "Video information" refers to image and video data acquired by surveillance devices, and is the information that is subject to analysis.

[0377] "Action recognition means" refers to means that have the function of analyzing a person's movements and position from video information.

[0378] A "discrimination means" is a means of determining whether the behavior analyzed by the behavior recognition means is abnormal.

[0379] An "alert transmission means" is a means that has the function of generating and transmitting warning information when the discrimination means determines that an abnormality has occurred.

[0380] An "emotion recognition means" is a means that has the function of analyzing a person's emotional state from video information.

[0381] "Warning information" refers to information that includes the content of alerts generated when an anomaly is detected.

[0382] "Smart glasses" are glasses-type information terminals worn by security guards that function as display devices for showing warning information.

[0383] This invention relates to a surveillance system applied to security services. The server receives video information acquired from a target area by a monitoring device and analyzes the movements and positions of people in the video using an action recognition means. At this time, it uses an image processing library such as OpenCV to detect movements frame by frame and uses a discrimination means to determine whether there is an anomaly. Furthermore, the server, as an emotion recognition means, uses a generative AI model to analyze emotional states from facial expressions and recognizes emotions such as stress and anxiety. If an anomaly is detected as a result, it uses an alert sending means to generate warning information and transmit it to a terminal including smart glasses.

[0384] The terminal displays received warning information to the user as a pop-up notification, prompting a quick response. This pop-up provides a visual alert via the head-up display of a glasses-type information terminal. For example, in a shopping mall's surveillance system, if a person is repeatedly opening and closing their bag in an unusual manner, and the analysis indicates they are experiencing tension or anxiety, this information is immediately notified to security guards. An example prompt message might be: "This person is repeatedly opening and closing their bag, and the emotion analysis indicates they are experiencing tension. Please investigate the scene."

[0385] Real-time performance is crucial in the design of this system, as it allows servers to efficiently process data and generate appropriate alerts, enabling the implementation of swift and effective security measures.

[0386] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0387] Step 1:

[0388] The server receives video information of the target area in real time from the monitoring device. This video information is video data transmitted from the camera, and each frame is the subject of analysis. The input is video data from the camera, and the output is image frames ready for analysis. Specifically, stream-formatted data is captured as images.

[0389] Step 2:

[0390] The server analyzes the movement and position of individuals from the received video information. To this end, it utilizes OpenCV as a motion recognition tool, applying an algorithm to identify individuals within frames and track their movements. The input is video data from each frame, and the output is analytical information regarding movement and position. Specifically, image processing techniques are used to detect the joints and positions of each individual.

[0391] Step 3:

[0392] The server identifies abnormal behavior based on the results of behavior recognition. If a specific behavior pattern is deemed abnormal, the discrimination means detects it. The input is the analysis information from step 2, and the output is the result of the abnormality determination. Specifically, the degree of abnormality is evaluated by comparing it with a pre-set standard.

[0393] Step 4:

[0394] The server uses emotion recognition to analyze a person's emotional state from video information. Emotions such as stress and fear are identified through facial expression analysis using a generative AI model. The input is image data of a person's face, and the output is the result of the emotional state analysis. Specifically, the emotion analysis model is applied to quantify the degree of emotion.

[0395] Step 5:

[0396] The server generates an alert and sends it to the smart glasses when it detects abnormal behavior or emotion. The generated warning information is adjusted according to the urgency of the alert. The input is the result from steps 3 and 4, and the output is the warning message. Specifically, a prompt message is automatically generated according to the urgency.

[0397] Step 6:

[0398] The device receives alerts from the server and displays them to the user as pop-up notifications. The displayed information includes details of the detected anomaly and the necessary corrective actions. The input is warning information from the server, and the output is a visual notification to the user. Specifically, the alert is displayed on the smart glasses' display, which the user acknowledges.

[0399] This system enables swift safety measures based on real-time detection of abnormal behavior and emotions.

[0400] 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.

[0401] 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.

[0402] 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.

[0403] [Third Embodiment]

[0404] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0405] 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.

[0406] 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).

[0407] 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.

[0408] 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.

[0409] 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).

[0410] 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.

[0411] 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.

[0412] 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.

[0413] 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.

[0414] 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.

[0415] 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".

[0416] This invention relates to a system for monitoring multiple individuals over a wide area, and consists of a monitoring device, a server, a terminal, and a user. The monitoring device transmits video information acquired within the target area to the server in real time. The server applies a high-precision behavior recognition algorithm and analyzes the movements and locations of individuals from the video information to quickly detect abnormal behavior.

[0417] For example, if a camera installed in the pool area of ​​a childcare facility is used as a monitoring device, the server will detect if a child suddenly starts running by the poolside as an abnormal movement. When this movement is detected, the server will promptly generate an alert and notify the terminal of the user responsible for monitoring.

[0418] The alerts displayed on the device include detailed information such as the time, location, and video clips of the incident, allowing users to immediately understand the specific situation. By checking the alerts, users can take appropriate and prompt action.

[0419] Furthermore, the system has a function that automatically contacts external rescue personnel in the event of a serious anomaly. For example, it can immediately contact the nearest ambulance, preventing delays in emergency response. This enhances safety within the monitored area and also reduces the burden on monitoring personnel.

[0420] The introduction of such a system will enable highly accurate and efficient monitoring of a large number of individuals across a wide area.

[0421] The following describes the processing flow.

[0422] Step 1:

[0423] The server receives video information transmitted from the monitoring device in real time. The video information is streamed and buffered for processing within the server.

[0424] Step 2:

[0425] The server divides the received video information into frames and executes an action recognition algorithm. This allows the server to identify the actions and positions of people within each frame.

[0426] Step 3:

[0427] The server uses facial recognition technology to identify individual people. This step allows the server to track each person's activity history and easily detect anomalies related to specific individuals.

[0428] Step 4:

[0429] The server analyzes the behavioral patterns and compares them to pre-configured dangerous behavioral patterns. If abnormal behavior is detected, the server registers it as a flag.

[0430] Step 5:

[0431] When the server detects an anomaly, it generates an alert. This alert includes the time and location of the anomaly, as well as identification information of the parties involved.

[0432] Step 6:

[0433] The server sends the generated alert to the assigned person's terminal. The alert is immediately notified and displayed as a pop-up on the terminal.

[0434] Step 7:

[0435] The device receives alerts in real time and displays notifications to the user. The user can then review the alert details and take prompt action on-site.

[0436] Step 8:

[0437] If a user detects a serious anomaly while responding to an alert, the server automatically sends a notification to external rescue personnel. This automated notification enables a rapid response to the emergency.

[0438] Step 9:

[0439] The server saves all alert information and response history to a database. This information is used for later analysis and report generation.

[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] In monitoring large numbers of people, there is a need to detect abnormal behavior with high accuracy and speed, reduce the workload on personnel, and respond quickly. Current technologies are insufficient to meet these requirements, and further efficiency improvements and automation are particularly important for systems that operate in real time.

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

[0444] In this invention, the server includes means for receiving video data in real time, means for analyzing motion and location, and means for determining anomalies. This enables highly accurate motion recognition and anomaly detection, and improves safety within the monitored area through rapid response.

[0445] A "monitoring device" is a device that captures images of the situation within a target area in real time, generates video data, and transmits it.

[0446] "Video data" refers to data containing visual information acquired by devices such as cameras.

[0447] "Receiving in real time" means that video data is transmitted directly to a server or other device with almost no delay after it is captured.

[0448] "Behavioral analysis means" refers to technologies and algorithms for processing received video data and analyzing the actions and positions of a subject.

[0449] An "anomaly detection method" is a means of identifying unusual behavior from analyzed operational data and identifying events that trigger warnings.

[0450] A "warning transmission means" is a means that has the function of generating warning information when an abnormality is detected and transmitting it to an external party.

[0451] "User's device" refers to electronic devices such as smartphones and tablets used to receive and notify warning information.

[0452] "External rescue personnel" refers to external organizations or individuals that have been pre-registered to deal with emergencies.

[0453] "Identification target" refers to a person or object identified through the analysis of video data.

[0454] This invention provides a system for efficiently monitoring a wide-area target, and includes monitoring equipment, a server, a terminal, and a user. The monitoring equipment acquires video data within the target area in real time and provides a foundation for analyzing people's movements and locations. This equipment typically uses network cameras capable of processing 1080p high-resolution video.

[0455] The server receives video data transmitted from monitoring devices and performs high-precision behavioral analysis on that data. This behavioral analysis is carried out using deep learning frameworks such as TensorFlow or PyTorch. Through this analysis, the server analyzes the target's movements and position frame by frame and detects anomalies. For example, a sudden movement of running can be identified as an anomaly.

[0456] When an anomaly is detected, the server immediately generates warning information and sends a push notification to the device. This device is a smartphone or tablet carried by the user, with a dedicated application installed. The warning information includes the time the anomaly occurred, location information, and a link to a relevant video clip, allowing the user to quickly understand the situation and take immediate action.

[0457] In the event of a serious anomaly, the server has a function to automatically contact external rescue personnel. By making an emergency call using VoIP, it is possible to request assistance immediately.

[0458] For example, when used in a swimming pool at a childcare facility, the server would immediately detect any dangerous behavior by children at the poolside and transmit appropriate warnings to the user in charge of management.

[0459] An example of a prompt used in a generative AI model is, "Please explain the specific method of anomaly detection using a highly accurate behavior recognition algorithm." This system improves security within the monitoring area and streamlines monitoring operations.

[0460] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0461] Step 1:

[0462] The server receives video data transmitted in real time from the monitoring device. The input is digital video data, and the output is stored internally as frames of this data. The server receives this as streaming data and prepares it for the next analysis process.

[0463] Step 2:

[0464] The server performs behavioral analysis on the received video data. The input is the previously received video frames, and the output is the analyzed motion and location information. The server uses a deep learning framework to analyze the person's movements in real time and identify individual movements. In this process, a model using TensorFlow analyzes the movement patterns and classifies each movement in preparation for anomaly detection.

[0465] Step 3:

[0466] The server identifies anomalies based on the analysis results. The input is the result of the behavioral analysis, and the output is a determination of whether or not an anomaly has occurred. The server uses an anomaly detection algorithm to identify behavior that deviates from normal behavioral patterns. For example, suddenly starting to run is generally considered an anomaly, and this determination is passed on to the next warning generation process.

[0467] Step 4:

[0468] The server immediately generates warning information and sends a push notification to the user's device upon detecting an anomaly. The input is the result of the anomaly detection, and the output is the warning notification sent to the user's device. The warning information is generated as an alert message and includes the time and location information of the anomaly, contributing to a rapid response.

[0469] Step 5:

[0470] The user checks the alert notification received on their device. The input is the warning message received on the device, and the output is the user's action based on the alert. Based on the information received, the user assesses the situation on site and, if necessary, rushes to the site or takes other countermeasures.

[0471] Step 6:

[0472] The server automatically contacts external rescue personnel when a critical anomaly is detected. The input is the determination of a critical anomaly, and the output is the notification to the external party. By using the VoIP protocol to make emergency contacts quickly, problems can be reported with minimal time lag, enabling immediate response.

[0473] (Application Example 1)

[0474] 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."

[0475] There is a need to detect abnormal behavior in real time within a wide-ranging monitoring area and to improve security quickly and efficiently. However, current systems have a lack of responsiveness due to insufficient rapid information dissemination after anomaly detection and inadequate notification to responders such as security guards.

[0476] 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.

[0477] In this invention, the server includes means for processing video information acquired from a monitoring device, behavior recognition means for analyzing the movement and position of a target from the video information, and discrimination means for determining whether the movement analyzed by the behavior recognition means is abnormal. This makes it possible to automatically detect abnormal behavior and quickly send warning information to the responder. Furthermore, notifications can be visually presented via a wearable device to enhance responsiveness.

[0478] A "surveillance device" is a device used to acquire video information within a target area.

[0479] "Visual information" refers to visual data acquired by surveillance equipment, including the movements and positions of individuals.

[0480] "Action recognition means" refers to a means for analyzing the movement and position of an object from video information and recognizing a specific action.

[0481] "Discrimination means" refers to means for determining whether or not the behavior analyzed by the behavior recognition means is abnormal.

[0482] An "alert generation means" is a means for generating warning information when the discrimination means determines that an abnormality has occurred, and for sending it to a predetermined recipient.

[0483] A "wearable device" is a device worn on the body to display information, and has the function of visually displaying warning information.

[0484] The system used to realize this application consists of a monitoring device, a server, and a wearable device. The server receives video information acquired from the monitoring device and analyzes the video information using a high-precision behavior recognition algorithm (e.g., a TensorFlow model) on a data processing platform (e.g., AWS Lambda) running on the cloud.

[0485] The server recognizes the target's movements and location from video information and determines whether abnormal behavior is present. During this process, it uses computer vision technologies such as OpenCV to process the data and analyze behavioral patterns. If an anomaly is detected, the server generates an alert and sends warning information to the responder's wearable device via a messaging platform such as Firebase.

[0486] As a concrete example, in a large shopping mall, security guards could wear smart glasses and utilize a surveillance system to enhance security within the commercial facility. If an individual exhibiting suspicious behavior is detected within the commercial facility, an alert would immediately appear on the security guard's smart glasses, allowing the guard to quickly verify and respond, thereby preventing crime.

[0487] An example of a prompt message for a generated AI model could be, "Detect any suspicious activity within the commercial facility and promptly notify security personnel." This would help improve security.

[0488] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0489] Step 1:

[0490] The server receives video information from the monitoring device in real time. The video information becomes the input, and at this stage it is still raw video data. The server temporarily stores this data in preparation for the next processing step.

[0491] Step 2:

[0492] The server preprocesses the received video information using OpenCV, a computer vision technology. This adjusts the image resolution and removes noise, converting it into a format suitable as input data for the action recognition model.

[0493] Step 3:

[0494] Based on pre-processed video information, the server performs motion analysis using an action recognition model trained with TensorFlow. The input for the analysis is pre-processed image data, and the output is motion and position information for each frame.

[0495] Step 4:

[0496] Based on the analysis results, the server performs anomaly detection using a generated AI model. In this step, current behavior data against past normal behavior patterns is used as input, and the presence or absence of abnormal behavior is determined as output.

[0497] Step 5:

[0498] If the server detects abnormal operation, it uses Firebase to generate alert information and sends it to the notification device. The input is the result of the abnormality detection, and the output generates a warning message to be sent to the wearable device.

[0499] Step 6:

[0500] The terminal displays received alerts on a wearable device. The input is alert information, and the output is a visual notification on the device. The user reviews this and takes prompt action on-site as needed.

[0501] 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.

[0502] This invention begins with transmitting video information acquired from a target area in real time to a server using a monitoring device. The server receives this video information and applies an action recognition means to analyze the movements and positions of people in each frame. Based on the analysis, a discrimination means determines whether there are any abnormalities in the movements. This process also includes a function to perform facial recognition of people and enable individual identification.

[0503] Furthermore, in this invention, by combining an emotion engine, the server also analyzes the user's emotional state from the video information. This process makes it possible to recognize emotions such as excitement, stress, and anxiety from subtle changes in the user's facial expressions and behavior. The server integrates this emotional information and adjusts the alert content when an anomaly is detected. For example, if a highly urgent emotional state is recognized, a more attention-grabbing alert is generated.

[0504] Meanwhile, the device receives alerts sent from the server and displays them to the user as pop-up notifications. Specifically, considering a situation by a swimming pool, if a child makes a movement that suggests they might drown, the device analyzes the child's emotional state along with the movement. If there are signs of stress or fear, a high-priority alert is immediately displayed on the device.

[0505] Users can check alerts on their devices and take quick action based on the details, including emotional information. The server also has an option to automatically notify external rescue personnel of the detected information, triggering a rapid external response if necessary.

[0506] Thus, the system of the present invention achieves a more precise and responsive monitoring system by integrating behavioral and emotional recognition.

[0507] The following describes the processing flow.

[0508] Step 1:

[0509] The server receives video information in real time from the monitoring device. This information is converted to an appropriate format within the server for analysis.

[0510] Step 2:

[0511] The server breaks down the received video information frame by frame and applies an action recognition algorithm. This allows for a detailed analysis of the movements and positions of the people appearing in each frame.

[0512] Step 3:

[0513] The server uses an emotion engine to analyze a person's facial expressions and body language from video information to infer their emotional state. This analysis determines whether the person is experiencing excitement, stress, anxiety, or other emotions.

[0514] Step 4:

[0515] The server integrates the results of behavior recognition and emotion analysis, and uses a discrimination mechanism to evaluate whether or not there is abnormal behavior. If an abnormality is detected, it is flagged as an escalated abnormality.

[0516] Step 5:

[0517] When a server is detected as abnormal, an alert system generates warning information. The alert includes the time of detection, location, and urgency level based on the emotional state of the person affected.

[0518] Step 6:

[0519] The server generates alert information and sends it to the assigned person's terminal. The terminal receives this information and displays it as a notification to the user.

[0520] Step 7:

[0521] The user checks the alert from their device. The alert provides a description of the situation and details of their emotional state, which the user uses to decide how to respond.

[0522] Step 8:

[0523] If a user determines that a serious anomaly or emergency has occurred, an external rescue team will be immediately contacted, either manually or through the server's automated functions.

[0524] Step 9:

[0525] The server saves logs of these processes to a database, which are then used for later analysis and report generation. The saved data helps improve the system and enhance future anomaly detection capabilities.

[0526] (Example 2)

[0527] 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."

[0528] Conventional surveillance systems focus on detecting anomalies based on human movement, making it difficult to grasp the overall situation. Furthermore, because they do not consider emotional states, the appropriateness and speed of responses to emergencies may be insufficient. Therefore, there is a need for systems that can meticulously assess anomalies from video information and issue optimal warnings that also take emotional states into account.

[0529] 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.

[0530] In this invention, the server includes means for receiving visual information acquired from a video recording device in real time, motion recognition means for analyzing human movement and position from the visual information, and evaluation means. This makes it possible to determine with high accuracy whether there is an abnormality in a person's movement, and to adjust warning information more appropriately using emotion estimation means.

[0531] A "video recording device" is a device that acquires visual information from a target area and has the function of collecting data in real time and transmitting it to a server.

[0532] "Visual information" refers to digital images and video data acquired by video recording equipment, and is the subject of analysis.

[0533] "Motion recognition means" refers to a method or device that analyzes human movement and position from visual information to detect specific movement patterns.

[0534] "Evaluation means" refers to a method or device for determining whether the motion analyzed by the motion recognition means is abnormal.

[0535] "Instruction transmission means" refers to a method or device for generating warning information when the evaluation means detects an abnormality and transmitting it to other devices or users.

[0536] An "emotion inference method" is a technique or device that analyzes a person's emotional state from visual information and uses the results to make situational judgments or adjust warning information.

[0537] Modes for carrying out the invention

[0538] This system consists of a server, terminals, and users working together. The server plays a central role in receiving visual information acquired from video recording equipment in real time and performing various analyses. Specifically, the server receives visual information and analyzes human movement and position using motion recognition software. In this process, a server equipped with a high-performance GPU is used as hardware, while general motion recognition tools and AI models are used as software. For example, custom models based on TensorFlow or PyTorch are common.

[0539] Furthermore, the server analyzes the user's emotional state using emotion estimation tools such as Microsoft Cognitive Services and open-source facial expression analysis tools. This allows it to estimate the user's emotions not only from their actions but also from their facial expressions, and adjust the urgency of warning information as needed. Other hardware includes large-capacity storage and network equipment that enables high-speed data communication.

[0540] The terminal receives instructions from the server and conveys them intuitively to the user. For example, mobile devices such as smartphones and tablets are often used as terminals. These terminals send warnings to the user through pop-up notifications and audio alerts.

[0541] Users can check warnings from their devices and take appropriate action as needed. For example, if an anomaly is detected in a monitored area, users are required to take action in accordance with the warning in order to respond quickly.

[0542] For example, when this system is used for poolside monitoring, the server detects children exhibiting drowning-like behavior based on visual information obtained from video recording equipment. Furthermore, emotion inference measures identify fear and stress from the child's facial expressions, and a highly urgent warning generated based on this information is immediately displayed on the terminal. This allows the user to take swift action.

[0543] An example of a prompt for a generating AI model is the instruction, "Build a system that analyzes video data from a monitoring device, identifies specific actions and emotions, and generates warnings."

[0544] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0545] Step 1:

[0546] The server receives visual information in real time from video recording equipment. The input is raw data captured by the camera. The server converts this data into a format that is easy to analyze and prepares it for analysis. Specifically, this involves decompressing the data and splitting it into frames.

[0547] Step 2:

[0548] The server applies a motion recognition model to extract the person's movements and position from the received visual information. The input used is the video frame formatted in step 1. The server performs image recognition frame by frame through the AI ​​model, analyzes the motion patterns, and outputs the results as motion data. Specifically, actions such as where a person walks or raises their hand are recognized.

[0549] Step 3:

[0550] The server performs an evaluation to detect anomalies based on the operational data. The input is the operational data extracted in step 2. The server compares this data with a pre-configured operational pattern to determine whether it is normal or abnormal. If an abnormality is detected, alert data is generated. In this process, the number of abnormal operations that occurred within a specific time period is also taken into consideration.

[0551] Step 4:

[0552] The server analyzes a person's emotional state from visual information. To do this, it applies an emotion prediction model to evaluate the person's facial expressions. Visual information is used as input. The server performs emotion prediction and outputs emotion labels such as "joy," "fear," and "surprise." This process, performed simultaneously with motion detection, results in more comprehensive analysis.

[0553] Step 5:

[0554] The server integrates the generated behavioral data and emotional information to prepare for generating alert information. The input is the output data from steps 3 and 4. The server determines the urgency and priority of the anomaly and outputs it as an alert message. For example, if a child is drowning in a pool while also showing signs of fear, a high-urgency alert will be generated.

[0555] Step 6:

[0556] The device receives alerts sent from the server and displays a warning to the user. The input is the alert data from the server. Based on the notification settings, the device outputs a pop-up notification or an audio alert. This makes it possible to immediately attract the user's attention.

[0557] Step 7:

[0558] The user checks the alert displayed on the device and takes the necessary action. The input is the alert displayed on the device. The user is required to understand the content of the alert and its details, and take appropriate action, for example, in the case of an emergency. Specific actions may include reporting an anomaly or rushing to the scene.

[0559] (Application Example 2)

[0560] 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."

[0561] Modern surveillance systems focus on detecting anomalies based on human movement, but methods for detecting abnormal behavior accompanied by emotional changes are underdeveloped. Therefore, from a security and safety perspective, there is a need for an integrated surveillance system that can quickly identify anomalies not only in movement but also in emotion, and prompt immediate action.

[0562] 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.

[0563] In this invention, the server includes means for receiving video information acquired from a monitoring device in real time, means for recognizing a person's movements and position from the video information, and means for recognizing a person's emotional state from the video information. This enables immediate detection of abnormalities in movements and emotions, and allows for immediate response via smart glasses.

[0564] A "monitoring device" is a device that acquires video information from a target area and has the function of transmitting this information to a server in real time.

[0565] "Video information" refers to image and video data acquired by surveillance devices, and is the information that is subject to analysis.

[0566] "Action recognition means" refers to means that have the function of analyzing a person's movements and position from video information.

[0567] A "discrimination means" is a means of determining whether the behavior analyzed by the behavior recognition means is abnormal.

[0568] An "alert transmission means" is a means that has the function of generating and transmitting warning information when the discrimination means determines that an abnormality has occurred.

[0569] An "emotion recognition means" is a means that has the function of analyzing a person's emotional state from video information.

[0570] "Warning information" refers to information that includes the content of alerts generated when an anomaly is detected.

[0571] "Smart glasses" are glasses-type information terminals worn by security guards that function as display devices for showing warning information.

[0572] This invention relates to a surveillance system applied to security services. The server receives video information acquired from a target area by a monitoring device and analyzes the movements and positions of people in the video using an action recognition means. At this time, it uses an image processing library such as OpenCV to detect movements frame by frame and uses a discrimination means to determine whether there is an anomaly. Furthermore, the server, as an emotion recognition means, uses a generative AI model to analyze emotional states from facial expressions and recognizes emotions such as stress and anxiety. If an anomaly is detected as a result, it uses an alert sending means to generate warning information and transmit it to a terminal including smart glasses.

[0573] The terminal displays received warning information to the user as a pop-up notification, prompting a quick response. This pop-up provides a visual alert via the head-up display of a glasses-type information terminal. For example, in a shopping mall's surveillance system, if a person is repeatedly opening and closing their bag in an unusual manner, and the analysis indicates they are experiencing tension or anxiety, this information is immediately notified to security guards. An example prompt message might be: "This person is repeatedly opening and closing their bag, and the emotion analysis indicates they are experiencing tension. Please investigate the scene."

[0574] Real-time performance is crucial in the design of this system, as it allows servers to efficiently process data and generate appropriate alerts, enabling the implementation of swift and effective security measures.

[0575] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0576] Step 1:

[0577] The server receives video information of the target area in real time from the monitoring device. This video information is video data transmitted from the camera, and each frame is the subject of analysis. The input is video data from the camera, and the output is image frames ready for analysis. Specifically, stream-formatted data is captured as images.

[0578] Step 2:

[0579] The server analyzes the movement and position of individuals from the received video information. To this end, it utilizes OpenCV as a motion recognition tool, applying an algorithm to identify individuals within frames and track their movements. The input is video data from each frame, and the output is analytical information regarding movement and position. Specifically, image processing techniques are used to detect the joints and positions of each individual.

[0580] Step 3:

[0581] The server identifies abnormal behavior based on the results of behavior recognition. If a specific behavior pattern is deemed abnormal, the discrimination means detects it. The input is the analysis information from step 2, and the output is the result of the abnormality determination. Specifically, the degree of abnormality is evaluated by comparing it with a pre-set standard.

[0582] Step 4:

[0583] The server uses emotion recognition to analyze a person's emotional state from video information. Emotions such as stress and fear are identified through facial expression analysis using a generative AI model. The input is image data of a person's face, and the output is the result of the emotional state analysis. Specifically, the emotion analysis model is applied to quantify the degree of emotion.

[0584] Step 5:

[0585] The server generates an alert and sends it to the smart glasses when it detects abnormal behavior or emotion. The generated warning information is adjusted according to the urgency of the alert. The input is the result from steps 3 and 4, and the output is the warning message. Specifically, a prompt message is automatically generated according to the urgency.

[0586] Step 6:

[0587] The device receives alerts from the server and displays them to the user as pop-up notifications. The displayed information includes details of the detected anomaly and the necessary corrective actions. The input is warning information from the server, and the output is a visual notification to the user. Specifically, the alert is displayed on the smart glasses' display, which the user acknowledges.

[0588] This system enables swift safety measures based on real-time detection of abnormal behavior and emotions.

[0589] 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.

[0590] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include 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.

[0591] 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.

[0592] [Fourth Embodiment]

[0593] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0594] 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.

[0595] 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).

[0596] 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.

[0597] 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.

[0598] 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).

[0599] 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.

[0600] 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.

[0601] 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.

[0602] 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.

[0603] 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.

[0604] 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.

[0605] 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".

[0606] This invention relates to a system for monitoring multiple individuals over a wide area, and consists of a monitoring device, a server, a terminal, and a user. The monitoring device transmits video information acquired within the target area to the server in real time. The server applies a high-precision behavior recognition algorithm and analyzes the movements and locations of individuals from the video information to quickly detect abnormal behavior.

[0607] For example, if a camera installed in the pool area of ​​a childcare facility is used as a monitoring device, the server will detect if a child suddenly starts running by the poolside as an abnormal movement. When this movement is detected, the server will promptly generate an alert and notify the terminal of the user responsible for monitoring.

[0608] The alerts displayed on the device include detailed information such as the time, location, and video clips of the incident, allowing users to immediately understand the specific situation. By checking the alerts, users can take appropriate and prompt action.

[0609] Furthermore, the system has a function that automatically contacts external rescue personnel in the event of a serious anomaly. For example, it can immediately contact the nearest ambulance, preventing delays in emergency response. This enhances safety within the monitored area and also reduces the burden on monitoring personnel.

[0610] The introduction of such a system will enable highly accurate and efficient monitoring of a large number of individuals across a wide area.

[0611] The following describes the processing flow.

[0612] Step 1:

[0613] The server receives video information transmitted from the monitoring device in real time. The video information is streamed and buffered for processing within the server.

[0614] Step 2:

[0615] The server divides the received video information into frames and executes an action recognition algorithm. This allows the server to identify the actions and positions of people within each frame.

[0616] Step 3:

[0617] The server uses facial recognition technology to identify individual people. This step allows the server to track each person's activity history and easily detect anomalies related to specific individuals.

[0618] Step 4:

[0619] The server analyzes the behavioral patterns and compares them to pre-configured dangerous behavioral patterns. If abnormal behavior is detected, the server registers it as a flag.

[0620] Step 5:

[0621] When the server detects an anomaly, it generates an alert. This alert includes the time and location of the anomaly, as well as identification information of the parties involved.

[0622] Step 6:

[0623] The server sends the generated alert to the assigned person's terminal. The alert is immediately notified and displayed as a pop-up on the terminal.

[0624] Step 7:

[0625] The device receives alerts in real time and displays notifications to the user. The user can then review the alert details and take prompt action on-site.

[0626] Step 8:

[0627] If a user detects a serious anomaly while responding to an alert, the server automatically sends a notification to external rescue personnel. This automated notification enables a rapid response to the emergency.

[0628] Step 9:

[0629] The server saves all alert information and response history to a database. This information is used for later analysis and report generation.

[0630] (Example 1)

[0631] 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".

[0632] In monitoring large numbers of people, there is a need to detect abnormal behavior with high accuracy and speed, reduce the workload on personnel, and respond quickly. Current technologies are insufficient to meet these requirements, and further efficiency improvements and automation are particularly important for systems that operate in real time.

[0633] 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.

[0634] In this invention, the server includes means for receiving video data in real time, means for analyzing motion and location, and means for determining anomalies. This enables highly accurate motion recognition and anomaly detection, and improves safety within the monitored area through rapid response.

[0635] A "monitoring device" is a device that captures images of the situation within a target area in real time, generates video data, and transmits it.

[0636] "Video data" refers to data containing visual information acquired by devices such as cameras.

[0637] "Receiving in real time" means that video data is transmitted directly to a server or other device with almost no delay after it is captured.

[0638] "Behavioral analysis means" refers to technologies and algorithms for processing received video data and analyzing the actions and positions of a subject.

[0639] An "anomaly detection method" is a means of identifying unusual behavior from analyzed operational data and identifying events that trigger warnings.

[0640] A "warning transmission means" is a means that has the function of generating warning information when an abnormality is detected and transmitting it to an external party.

[0641] "User's device" refers to electronic devices such as smartphones and tablets used to receive and notify warning information.

[0642] "External rescue personnel" refers to external organizations or individuals that have been pre-registered to deal with emergencies.

[0643] "Identification target" refers to a person or object identified through the analysis of video data.

[0644] This invention provides a system for efficiently monitoring a wide-area target, and includes monitoring equipment, a server, a terminal, and a user. The monitoring equipment acquires video data within the target area in real time and provides a foundation for analyzing people's movements and locations. This equipment typically uses network cameras capable of processing 1080p high-resolution video.

[0645] The server receives video data transmitted from monitoring devices and performs high-precision behavioral analysis on that data. This behavioral analysis is carried out using deep learning frameworks such as TensorFlow or PyTorch. Through this analysis, the server analyzes the target's movements and position frame by frame and detects anomalies. For example, a sudden movement of running can be identified as an anomaly.

[0646] When an anomaly is detected, the server immediately generates warning information and sends a push notification to the device. This device is a smartphone or tablet carried by the user, with a dedicated application installed. The warning information includes the time the anomaly occurred, location information, and a link to a relevant video clip, allowing the user to quickly understand the situation and take immediate action.

[0647] In the event of a serious anomaly, the server has a function to automatically contact external rescue personnel. By making an emergency call using VoIP, it is possible to request assistance immediately.

[0648] For example, when used in a swimming pool at a childcare facility, the server would immediately detect any dangerous behavior by children at the poolside and transmit appropriate warnings to the user in charge of management.

[0649] An example of a prompt used in a generative AI model is, "Please explain the specific method of anomaly detection using a highly accurate behavior recognition algorithm." This system improves security within the monitoring area and streamlines monitoring operations.

[0650] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0651] Step 1:

[0652] The server receives video data transmitted in real time from the monitoring device. The input is digital video data, and the output is stored internally as frames of this data. The server receives this as streaming data and prepares it for the next analysis process.

[0653] Step 2:

[0654] The server performs behavioral analysis on the received video data. The input is the previously received video frames, and the output is the analyzed motion and location information. The server uses a deep learning framework to analyze the person's movements in real time and identify individual movements. In this process, a model using TensorFlow analyzes the movement patterns and classifies each movement in preparation for anomaly detection.

[0655] Step 3:

[0656] The server identifies anomalies based on the analysis results. The input is the result of the behavioral analysis, and the output is a determination of whether or not an anomaly has occurred. The server uses an anomaly detection algorithm to identify behavior that deviates from normal behavioral patterns. For example, suddenly starting to run is generally considered an anomaly, and this determination is passed on to the next warning generation process.

[0657] Step 4:

[0658] The server immediately generates warning information and sends a push notification to the user's device upon detecting an anomaly. The input is the result of the anomaly detection, and the output is the warning notification sent to the user's device. The warning information is generated as an alert message and includes the time and location information of the anomaly, contributing to a rapid response.

[0659] Step 5:

[0660] The user checks the alert notification received on their device. The input is the warning message received on the device, and the output is the user's action based on the alert. Based on the information received, the user assesses the situation on site and, if necessary, rushes to the site or takes other countermeasures.

[0661] Step 6:

[0662] The server automatically contacts external rescue personnel when a critical anomaly is detected. The input is the determination of a critical anomaly, and the output is the notification to the external party. By using the VoIP protocol to make emergency contacts quickly, problems can be reported with minimal time lag, enabling immediate response.

[0663] (Application Example 1)

[0664] 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".

[0665] There is a need to detect abnormal behavior in real time within a wide-ranging monitoring area and to improve security quickly and efficiently. However, current systems have a lack of responsiveness due to insufficient rapid information dissemination after anomaly detection and inadequate notification to responders such as security guards.

[0666] 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.

[0667] In this invention, the server includes means for processing video information acquired from a monitoring device, behavior recognition means for analyzing the movement and position of a target from the video information, and discrimination means for determining whether the movement analyzed by the behavior recognition means is abnormal. This makes it possible to automatically detect abnormal behavior and quickly send warning information to the responder. Furthermore, notifications can be visually presented via a wearable device to enhance responsiveness.

[0668] A "surveillance device" is a device used to acquire video information within a target area.

[0669] "Visual information" refers to visual data acquired by surveillance equipment, including the movements and positions of individuals.

[0670] "Action recognition means" refers to a means for analyzing the movement and position of an object from video information and recognizing a specific action.

[0671] "Discrimination means" refers to means for determining whether or not the behavior analyzed by the behavior recognition means is abnormal.

[0672] An "alert generation means" is a means for generating warning information when the discrimination means determines that an abnormality has occurred, and for sending it to a predetermined recipient.

[0673] A "wearable device" is a device worn on the body to display information, and has the function of visually displaying warning information.

[0674] The system used to realize this application consists of a monitoring device, a server, and a wearable device. The server receives video information acquired from the monitoring device and analyzes the video information using a high-precision behavior recognition algorithm (e.g., a TensorFlow model) on a data processing platform (e.g., AWS Lambda) running on the cloud.

[0675] The server recognizes the target's movements and location from video information and determines whether abnormal behavior is present. During this process, it uses computer vision technologies such as OpenCV to process the data and analyze behavioral patterns. If an anomaly is detected, the server generates an alert and sends warning information to the responder's wearable device via a messaging platform such as Firebase.

[0676] As a concrete example, in a large shopping mall, security guards could wear smart glasses and utilize a surveillance system to enhance security within the commercial facility. If an individual exhibiting suspicious behavior is detected within the commercial facility, an alert would immediately appear on the security guard's smart glasses, allowing the guard to quickly verify and respond, thereby preventing crime.

[0677] An example of a prompt message for a generated AI model could be, "Detect any suspicious activity within the commercial facility and promptly notify security personnel." This would help improve security.

[0678] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0679] Step 1:

[0680] The server receives video information from the monitoring device in real time. The video information becomes the input, and at this stage it is still raw video data. The server temporarily stores this data in preparation for the next processing step.

[0681] Step 2:

[0682] The server preprocesses the received video information using OpenCV, a computer vision technology. This adjusts the image resolution and removes noise, converting it into a format suitable as input data for the action recognition model.

[0683] Step 3:

[0684] Based on pre-processed video information, the server performs motion analysis using an action recognition model trained with TensorFlow. The input for the analysis is pre-processed image data, and the output is motion and position information for each frame.

[0685] Step 4:

[0686] Based on the analysis results, the server performs anomaly detection using a generated AI model. In this step, current behavior data against past normal behavior patterns is used as input, and the presence or absence of abnormal behavior is determined as output.

[0687] Step 5:

[0688] If the server detects abnormal operation, it uses Firebase to generate alert information and sends it to the notification device. The input is the result of the abnormality detection, and the output generates a warning message to be sent to the wearable device.

[0689] Step 6:

[0690] The terminal displays received alerts on a wearable device. The input is alert information, and the output is a visual notification on the device. The user reviews this and takes prompt action on-site as needed.

[0691] 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.

[0692] This invention begins with transmitting video information acquired from a target area in real time to a server using a monitoring device. The server receives this video information and applies an action recognition means to analyze the movements and positions of people in each frame. Based on the analysis, a discrimination means determines whether there are any abnormalities in the movements. This process also includes a function to perform facial recognition of people and enable individual identification.

[0693] Furthermore, in this invention, by combining an emotion engine, the server also analyzes the user's emotional state from the video information. This process makes it possible to recognize emotions such as excitement, stress, and anxiety from subtle changes in the user's facial expressions and behavior. The server integrates this emotional information and adjusts the alert content when an anomaly is detected. For example, if a highly urgent emotional state is recognized, a more attention-grabbing alert is generated.

[0694] Meanwhile, the device receives alerts sent from the server and displays them to the user as pop-up notifications. Specifically, considering a situation by a swimming pool, if a child makes a movement that suggests they might drown, the device analyzes the child's emotional state along with the movement. If there are signs of stress or fear, a high-priority alert is immediately displayed on the device.

[0695] Users can check alerts on their devices and take quick action based on the details, including emotional information. The server also has an option to automatically notify external rescue personnel of the detected information, triggering a rapid external response if necessary.

[0696] Thus, the system of the present invention achieves a more precise and responsive monitoring system by integrating behavioral and emotional recognition.

[0697] The following describes the processing flow.

[0698] Step 1:

[0699] The server receives video information in real time from the monitoring device. This information is converted to an appropriate format within the server for analysis.

[0700] Step 2:

[0701] The server breaks down the received video information frame by frame and applies an action recognition algorithm. This allows for a detailed analysis of the movements and positions of the people appearing in each frame.

[0702] Step 3:

[0703] The server uses an emotion engine to analyze a person's facial expressions and body language from video information to infer their emotional state. This analysis determines whether the person is experiencing excitement, stress, anxiety, or other emotions.

[0704] Step 4:

[0705] The server integrates the results of behavior recognition and emotion analysis, and uses a discrimination mechanism to evaluate whether or not there is abnormal behavior. If an abnormality is detected, it is flagged as an escalated abnormality.

[0706] Step 5:

[0707] When a server is detected as abnormal, an alert system generates warning information. The alert includes the time of detection, location, and urgency level based on the emotional state of the person affected.

[0708] Step 6:

[0709] The server generates alert information and sends it to the assigned person's terminal. The terminal receives this information and displays it as a notification to the user.

[0710] Step 7:

[0711] The user checks the alert from their device. The alert provides a description of the situation and details of their emotional state, which the user uses to decide how to respond.

[0712] Step 8:

[0713] If a user determines that a serious anomaly or emergency has occurred, an external rescue team will be immediately contacted, either manually or through the server's automated functions.

[0714] Step 9:

[0715] The server saves logs of these processes to a database, which are then used for later analysis and report generation. The saved data helps improve the system and enhance future anomaly detection capabilities.

[0716] (Example 2)

[0717] 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".

[0718] Conventional surveillance systems focus on detecting anomalies based on human movement, making it difficult to grasp the overall situation. Furthermore, because they do not consider emotional states, the appropriateness and speed of responses to emergencies may be insufficient. Therefore, there is a need for systems that can meticulously assess anomalies from video information and issue optimal warnings that also take emotional states into account.

[0719] 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.

[0720] In this invention, the server includes means for receiving visual information acquired from a video recording device in real time, motion recognition means for analyzing human movement and position from the visual information, and evaluation means. This makes it possible to determine with high accuracy whether there is an abnormality in a person's movement, and to adjust warning information more appropriately using emotion estimation means.

[0721] A "video recording device" is a device that acquires visual information from a target area and has the function of collecting data in real time and transmitting it to a server.

[0722] "Visual information" refers to digital images and video data acquired by video recording equipment, and is the subject of analysis.

[0723] "Motion recognition means" refers to a method or device that analyzes human movement and position from visual information to detect specific movement patterns.

[0724] "Evaluation means" refers to a method or device for determining whether the motion analyzed by the motion recognition means is abnormal.

[0725] "Instruction transmission means" refers to a method or device for generating warning information when the evaluation means detects an abnormality and transmitting it to other devices or users.

[0726] An "emotion inference method" is a technique or device that analyzes a person's emotional state from visual information and uses the results to make situational judgments or adjust warning information.

[0727] Modes for carrying out the invention

[0728] This system consists of a server, terminals, and users working together. The server plays a central role in receiving visual information acquired from video recording equipment in real time and performing various analyses. Specifically, the server receives visual information and analyzes human movement and position using motion recognition software. In this process, a server equipped with a high-performance GPU is used as hardware, while general motion recognition tools and AI models are used as software. For example, custom models based on TensorFlow or PyTorch are common.

[0729] Furthermore, the server analyzes the user's emotional state using emotion estimation tools such as Microsoft Cognitive Services and open-source facial expression analysis tools. This allows it to estimate the user's emotions not only from their actions but also from their facial expressions, and adjust the urgency of warning information as needed. Other hardware includes large-capacity storage and network equipment that enables high-speed data communication.

[0730] The terminal receives instructions from the server and conveys them intuitively to the user. For example, mobile devices such as smartphones and tablets are often used as terminals. These terminals send warnings to the user through pop-up notifications and audio alerts.

[0731] Users can check warnings from their devices and take appropriate action as needed. For example, if an anomaly is detected in a monitored area, users are required to take action in accordance with the warning in order to respond quickly.

[0732] For example, when this system is used for poolside monitoring, the server detects children exhibiting drowning-like behavior based on visual information obtained from video recording equipment. Furthermore, emotion inference measures identify fear and stress from the child's facial expressions, and a highly urgent warning generated based on this information is immediately displayed on the terminal. This allows the user to take swift action.

[0733] An example of a prompt for a generating AI model is the instruction, "Build a system that analyzes video data from a monitoring device, identifies specific actions and emotions, and generates warnings."

[0734] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0735] Step 1:

[0736] The server receives visual information in real time from video recording equipment. The input is raw data captured by the camera. The server converts this data into a format that is easy to analyze and prepares it for analysis. Specifically, this involves decompressing the data and splitting it into frames.

[0737] Step 2:

[0738] The server applies a motion recognition model to extract the person's movements and position from the received visual information. The input used is the video frame formatted in step 1. The server performs image recognition frame by frame through the AI ​​model, analyzes the motion patterns, and outputs the results as motion data. Specifically, actions such as where a person walks or raises their hand are recognized.

[0739] Step 3:

[0740] The server performs an evaluation to detect anomalies based on the operational data. The input is the operational data extracted in step 2. The server compares this data with a pre-configured operational pattern to determine whether it is normal or abnormal. If an abnormality is detected, alert data is generated. In this process, the number of abnormal operations that occurred within a specific time period is also taken into consideration.

[0741] Step 4:

[0742] The server analyzes a person's emotional state from visual information. To do this, it applies an emotion prediction model to evaluate the person's facial expressions. Visual information is used as input. The server performs emotion prediction and outputs emotion labels such as "joy," "fear," and "surprise." This process, performed simultaneously with motion detection, results in more comprehensive analysis.

[0743] Step 5:

[0744] The server integrates the generated behavioral data and emotional information to prepare for generating alert information. The input is the output data from steps 3 and 4. The server determines the urgency and priority of the anomaly and outputs it as an alert message. For example, if a child is drowning in a pool while also showing signs of fear, a high-urgency alert will be generated.

[0745] Step 6:

[0746] The device receives alerts sent from the server and displays a warning to the user. The input is the alert data from the server. Based on the notification settings, the device outputs a pop-up notification or an audio alert. This makes it possible to immediately attract the user's attention.

[0747] Step 7:

[0748] The user checks the alert displayed on the device and takes the necessary action. The input is the alert displayed on the device. The user is required to understand the content of the alert and its details, and take appropriate action, for example, in the case of an emergency. Specific actions may include reporting an anomaly or rushing to the scene.

[0749] (Application Example 2)

[0750] 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".

[0751] Modern surveillance systems focus on detecting anomalies based on human movement, but methods for detecting abnormal behavior accompanied by emotional changes are underdeveloped. Therefore, from a security and safety perspective, there is a need for an integrated surveillance system that can quickly identify anomalies not only in movement but also in emotion, and prompt immediate action.

[0752] 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.

[0753] In this invention, the server includes means for receiving video information acquired from a monitoring device in real time, means for recognizing a person's movements and position from the video information, and means for recognizing a person's emotional state from the video information. This enables immediate detection of abnormalities in movements and emotions, and allows for immediate response via smart glasses.

[0754] A "monitoring device" is a device that acquires video information from a target area and has the function of transmitting this information to a server in real time.

[0755] "Video information" refers to image and video data acquired by surveillance devices, and is the information that is subject to analysis.

[0756] "Action recognition means" refers to means that have the function of analyzing a person's movements and position from video information.

[0757] A "discrimination means" is a means of determining whether the behavior analyzed by the behavior recognition means is abnormal.

[0758] An "alert transmission means" is a means that has the function of generating and transmitting warning information when the discrimination means determines that an abnormality has occurred.

[0759] An "emotion recognition means" is a means that has the function of analyzing a person's emotional state from video information.

[0760] "Warning information" refers to information that includes the content of alerts generated when an anomaly is detected.

[0761] "Smart glasses" are glasses-type information terminals worn by security guards that function as display devices for showing warning information.

[0762] This invention relates to a surveillance system applied to security services. The server receives video information acquired from a target area by a monitoring device and analyzes the movements and positions of people in the video using an action recognition means. At this time, it uses an image processing library such as OpenCV to detect movements frame by frame and uses a discrimination means to determine whether there is an anomaly. Furthermore, the server, as an emotion recognition means, uses a generative AI model to analyze emotional states from facial expressions and recognizes emotions such as stress and anxiety. If an anomaly is detected as a result, it uses an alert sending means to generate warning information and transmit it to a terminal including smart glasses.

[0763] The terminal displays received warning information to the user as a pop-up notification, prompting a quick response. This pop-up provides a visual alert via the head-up display of a glasses-type information terminal. For example, in a shopping mall's surveillance system, if a person is repeatedly opening and closing their bag in an unusual manner, and the analysis indicates they are experiencing tension or anxiety, this information is immediately notified to security guards. An example prompt message might be: "This person is repeatedly opening and closing their bag, and the emotion analysis indicates they are experiencing tension. Please investigate the scene."

[0764] Real-time performance is crucial in the design of this system, as it allows servers to efficiently process data and generate appropriate alerts, enabling the implementation of swift and effective security measures.

[0765] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0766] Step 1:

[0767] The server receives video information of the target area in real time from the monitoring device. This video information is video data transmitted from the camera, and each frame is the subject of analysis. The input is video data from the camera, and the output is image frames ready for analysis. Specifically, stream-formatted data is captured as images.

[0768] Step 2:

[0769] The server analyzes the movement and position of individuals from the received video information. To this end, it utilizes OpenCV as a motion recognition tool, applying an algorithm to identify individuals within frames and track their movements. The input is video data from each frame, and the output is analytical information regarding movement and position. Specifically, image processing techniques are used to detect the joints and positions of each individual.

[0770] Step 3:

[0771] The server identifies abnormal behavior based on the results of behavior recognition. If a specific behavior pattern is deemed abnormal, the discrimination means detects it. The input is the analysis information from step 2, and the output is the result of the abnormality determination. Specifically, the degree of abnormality is evaluated by comparing it with a pre-set standard.

[0772] Step 4:

[0773] The server uses emotion recognition to analyze a person's emotional state from video information. Emotions such as stress and fear are identified through facial expression analysis using a generative AI model. The input is image data of a person's face, and the output is the result of the emotional state analysis. Specifically, the emotion analysis model is applied to quantify the degree of emotion.

[0774] Step 5:

[0775] The server generates an alert and sends it to the smart glasses when it detects abnormal behavior or emotion. The generated warning information is adjusted according to the urgency of the alert. The input is the result from steps 3 and 4, and the output is the warning message. Specifically, a prompt message is automatically generated according to the urgency.

[0776] Step 6:

[0777] The device receives alerts from the server and displays them to the user as pop-up notifications. The displayed information includes details of the detected anomaly and the necessary corrective actions. The input is warning information from the server, and the output is a visual notification to the user. Specifically, the alert is displayed on the smart glasses' display, which the user acknowledges.

[0778] This system enables swift safety measures based on real-time detection of abnormal behavior and emotions.

[0779] 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.

[0780] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include 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.

[0781] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0782] 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.

[0783] 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.

[0784] 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.

[0785] 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.

[0786] 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.

[0787] 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."

[0788] 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.

[0789] 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.

[0790] 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.

[0791] 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.

[0792] 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.

[0793] 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.

[0794] 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.

[0795] 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.

[0796] 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.

[0797] 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.

[0798] 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.

[0799] 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.

[0800] The following is further disclosed regarding the embodiments described above.

[0801] (Claim 1)

[0802] A means of receiving video information acquired from a monitoring device in real time,

[0803] Action recognition means for analyzing the movements and position of a person from the aforementioned video information,

[0804] A determination means for determining whether the action analyzed by the action recognition means is abnormal,

[0805] An alert issuing means generates and transmits warning information when the aforementioned discrimination means determines that an abnormality has occurred,

[0806] A system that includes this.

[0807] (Claim 2)

[0808] The system according to claim 1, further comprising means for automatically contacting an external rescue personnel when the alert issuing means detects an abnormality.

[0809] (Claim 3)

[0810] The system according to claim 1, wherein the discrimination means comprises means for identifying individual persons based on video information.

[0811] "Example 1"

[0812] (Claim 1)

[0813] A means of receiving video data acquired from a monitoring device in real time,

[0814] A behavioral analysis means for analyzing the movement and position of the target based on the aforementioned video data,

[0815] An anomaly detection means for determining whether the behavior analyzed by the aforementioned behavior analysis means is abnormal,

[0816] A warning issuing means that generates and notifies warning information when the abnormality detection means determines that an abnormality has occurred,

[0817] A means for sending the aforementioned warning information as a push notification to the user's terminal,

[0818] A system that includes this.

[0819] (Claim 2)

[0820] The system according to claim 1, further comprising means for automatically contacting an external rescue personnel when the warning issuing means detects an abnormality.

[0821] (Claim 3)

[0822] The system according to claim 1, wherein the anomaly detection means includes means for identifying an object to be identified based on video data.

[0823] "Application Example 1"

[0824] (Claim 1)

[0825] A means for processing video information acquired from a monitoring device,

[0826] Action recognition means for analyzing the movement and position of the target from the aforementioned video information,

[0827] A determination means for determining whether the action analyzed by the action recognition means is abnormal,

[0828] An alert sending means generates warning information and transmits it to a notification device when the aforementioned discrimination means determines that an abnormality has occurred,

[0829] A means of visually displaying warning information using a wearable device,

[0830] A system that includes this.

[0831] (Claim 2)

[0832] The system according to claim 1, further comprising means for automatically contacting an external responder when the alert issuing means detects an anomaly.

[0833] (Claim 3)

[0834] The system according to claim 1, wherein the discrimination means comprises means for identifying an individual based on video information.

[0835] "Example 2 of combining an emotion engine"

[0836] (Claim 1)

[0837] A means of receiving visual information acquired from video recording equipment in real time,

[0838] A motion recognition means for analyzing human movement and position from the aforementioned visual information,

[0839] An evaluation means for determining whether the operation analyzed by the operation recognition means is abnormal,

[0840] An instruction sending means that generates and transmits warning information when the evaluation means determines that an abnormality has occurred,

[0841] A means for analyzing a person's emotional state from the visual information using an emotion inference means,

[0842] A means for adjusting the content of warning information based on the analysis results by the emotion estimation means,

[0843] A system that includes this.

[0844] (Claim 2)

[0845] The system according to claim 1, further comprising means for automatically contacting an external support person when the instruction sending means detects an abnormality.

[0846] (Claim 3)

[0847] The system according to claim 1, wherein the evaluation means includes means for identifying an individual based on visual information.

[0848] "Application example 2 when combining with an emotional engine"

[0849] (Claim 1)

[0850] A means of receiving video information acquired from a monitoring device in real time,

[0851] Action recognition means for analyzing the movements and position of a person from the aforementioned video information,

[0852] A determination means for determining whether the action analyzed by the action recognition means is abnormal,

[0853] An alert issuing means generates and transmits warning information when the aforementioned discrimination means determines that an abnormality has occurred,

[0854] An emotion recognition means for analyzing a person's emotional state from the aforementioned video information,

[0855] A means for adjusting the content of warning information based on the emotional state analyzed by the aforementioned emotion recognition means,

[0856] A means of displaying alerts on smart glasses,

[0857] A system that includes this.

[0858] (Claim 2)

[0859] The system according to claim 1, further comprising means for automatically contacting external rescue when the alert issuing means detects an abnormality.

[0860] (Claim 3)

[0861] The system according to claim 1, wherein the discrimination means comprises means for identifying individual persons based on video information. [Explanation of symbols]

[0862] 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 of receiving video information acquired from a monitoring device in real time, Action recognition means for analyzing the movements and position of a person from the aforementioned video information, A determination means for determining whether the action analyzed by the action recognition means is abnormal, An alert issuing means generates and transmits warning information when the aforementioned discrimination means determines that an abnormality has occurred, A system that includes this.

2. The system according to claim 1, further comprising means for automatically contacting an external rescue personnel when the alert issuing means detects an abnormality.

3. The system according to claim 1, wherein the discrimination means comprises means for identifying individual persons based on video information.

Images