system

Abstract

We provide the system. [Solution] A means for acquiring image data obtained from a surveillance camera, A means for recognizing a person's face based on acquired image data, A means of classifying a person's emotions based on their recognized face, A means for analyzing classified emotion data and detecting abnormal emotional states, Means of reporting to the police based on detected abnormal emotional states, A means of sending an alert to people in the surrounding area, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In recent years, the importance of safety and security in public places has increased, but it is difficult for traditional monitoring systems to quickly detect and handle abnormal behaviors. Also, it is difficult for conventional technologies to detect human emotions and monitor drastic fluctuations in those emotions in real time. As a result, there is a problem that potential crimes and dangerous situations cannot be prevented in advance.

Means for Solving the Problems

[0005] This invention provides a system that analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. Specifically, the system includes means for acquiring image data from a surveillance camera, means for recognizing a person's face based on the acquired image data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotional data to detect abnormal emotional states, means for notifying the police based on the detected abnormal emotional state, and means for issuing alerts to people in the vicinity. This system makes it possible to immediately capture sudden changes in emotions and respond quickly.

[0006] A "surveillance camera" is a device installed in public places that captures images and videos in real time and records or transmits that data.

[0007] "Image data" refers to visual information acquired from devices such as surveillance cameras, represented in digital format.

[0008] "Methods for recognizing faces" refer to algorithms and software that automatically detect a person's face from image data and identify its shape and characteristics.

[0009] "Methods for classifying emotions" refer to algorithms and machine learning models that estimate a person's emotional state from a recognized facial image and classify it into multiple emotional categories (e.g., joy, anger, surprise, etc.).

[0010] "Emotional data" refers to information about emotional states classified based on a person's face detected by facial recognition technology.

[0011] "Means for detecting abnormal emotional states" refers to algorithms or software that analyze fluctuations in classified emotional data and detect abnormalities when those fluctuations exceed the normal range.

[0012] "Means of reporting to the police" refers to communication devices or software that automatically send notifications to the police when an abnormal emotional state is detected.

[0013] "Means of issuing alerts" refers to devices or systems that issue audio or visual warnings to people in the vicinity when an abnormal emotional state is detected. [Brief explanation of the drawing]

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

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

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

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

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

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

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

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

[0022] [First Embodiment]

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

[0024] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

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

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

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

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

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

[0031] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

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

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0035] The present invention's system analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[0036] 1. Collection of image data

[0037] server

[0038] The system connects to surveillance cameras via a network and acquires image data in real time. This data is temporarily stored in storage.

[0039] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[0040] 2. Sentiment analysis

[0041] server

[0042] A face recognition algorithm is applied to the stored image data to detect the faces of people in the images.

[0043] The detected facial images are input into a machine learning model to classify emotions. Emotional classifications include categories such as "joy," "anger," and "surprise."

[0044] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[0045] 3. Anomaly detection

[0046] server

[0047] Based on emotional data, it analyzes fluctuations in those emotions and detects abnormal emotional states that exceed a set threshold.

[0048] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[0049] 4. Reporting and issuing alerts

[0050] server

[0051] If an abnormal emotional state is detected, an automatic report will be sent to the police. The report will include location information, a person's characteristics, and their emotional state.

[0052] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[0053] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[0054] Thus, the system of the present invention operates in conjunction with surveillance cameras and improves safety in public places by analyzing emotional fluctuations in real time. This makes it possible to quickly capture sudden changes in emotions and respond promptly accordingly. Furthermore, since machine learning models are used to analyze emotional data, accuracy improves through continuous data learning, enabling more effective risk management.

[0055] As a concrete example, consider a scenario where a specific individual suddenly displays strong anger in a public space. The server first detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be a potential danger. In this case, the server automatically notifies the police and also sends a signal to surrounding alert systems. This allows for immediate action to be taken, preventing a dangerous situation from occurring.

[0056] The above describes the basic form and operation for carrying out the present invention. This system is expected to have a wide range of applications as an effective tool for enhancing public safety.

[0057] The following describes the processing flow.

[0058] Step 1:

[0059] server

[0060] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0061] Step 2:

[0062] server

[0063] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0064] Step 3:

[0065] server

[0066] The detected facial portion is extracted and input into a machine learning model (for example, a deep learning-based emotion classification model). This model analyzes the facial features and classifies the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[0067] Step 4:

[0068] server

[0069] Based on classified emotion data, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0070] Step 5:

[0071] server

[0072] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[0073] Step 6:

[0074] server

[0075] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[0076] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[0077] (Example 1)

[0078] 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."

[0079] Traditional surveillance systems often relied on manual processes for identifying individuals and detecting abnormal behavior, resulting in significant labor and time constraints. Furthermore, real-time analysis of abnormal emotional fluctuations and rapid response were challenging. Therefore, there was a need for effective methods to improve safety in public spaces.

[0080] 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.

[0081] In this invention, the server includes means for acquiring image data from a surveillance camera, means for detecting a person's face based on the stored image data, and means for inputting the detected face into a machine learning model to classify emotions. This makes it possible to detect abnormal emotional states in real time and automatically notify the police. Furthermore, by quickly issuing alerts to people in the vicinity, dangerous situations can be prevented.

[0082] A "surveillance camera" is a device that continuously photographs a designated area or object and acquires video data.

[0083] "Image data" refers to visual information acquired by cameras and other recording devices, stored in a digital format.

[0084] "Storage" refers to a memory device used to temporarily or permanently store digital data.

[0085] A "face recognition algorithm" is a computational method for detecting the face portion of a person from image data and identifying specific features.

[0086] A "machine learning model" is an algorithm or a set of algorithms that learns from data and makes predictions about new data.

[0087] "Emotion classification" is the process of categorizing the emotions a person is expressing based on characteristic data obtained from their face.

[0088] An "abnormal emotional state" refers to an emotional intensity or fluctuation that exceeds the normal range and may indicate a potential danger.

[0089] "Reporting" refers to the act of automatically issuing a warning to relevant authorities, such as the police, when a monitoring system detects an anomaly.

[0090] An "alert" is a warning signal that is issued to alert people in the vicinity when an abnormal situation is detected.

[0091] A "server" is a computer system that manages and controls data processing, storage, and communication on a network.

[0092] This invention relates to a system that analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of a surveillance camera, a server, and related software.

[0093] 1. System Configuration

[0094] The system includes surveillance cameras, servers, storage, and network infrastructure as its main components. The servers play a central role in acquiring image data from the surveillance cameras and storing that data in storage. Specific hardware and software used include surveillance cameras (e.g., network cameras), network infrastructure (e.g., routers), servers (e.g., general-purpose servers), and storage (e.g., cloud storage).

[0095] 2. Data Collection and Storage

[0096] The server connects to the surveillance camera via a network and acquires image data in real time. The acquired image data is stored in cloud storage. The server also adds a timestamp to the image data before saving it.

[0097] 3. Face detection and sentiment analysis

[0098] The server uses a face recognition algorithm (e.g., OpenCV) to detect human faces based on the stored image data. The detected face images are input into a machine learning model (e.g., TENSORFLOW®) to classify emotions into categories such as "joy," "anger," and "surprise."

[0099] 4. Detection of abnormal emotions

[0100] The server analyzes emotional fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold. If extreme emotional changes are observed in a short period of time, it is assessed as high risk.

[0101] 5. Notification and Alert System

[0102] The server automatically notifies the police if it detects an abnormal emotional state. The report includes location information, the person's characteristics, and their emotional state. Simultaneously, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[0103] Specific example

[0104] For example, if a particular individual suddenly displays strong anger in a public space, the server detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be high risk. In that case, the server automatically notifies the police and sends a warning signal to the surrounding alert system. This allows for immediate action and prevents dangerous situations from occurring.

[0105] Example of a prompt

[0106] "Collect image data from surveillance cameras in real time and save it on the server."

[0107] "Use image data stored on the server, apply a facial recognition algorithm, and input the results into a machine learning model to classify emotions."

[0108] "Based on the results of the sentiment analysis, analyze the emotional fluctuations and detect any anomalies that exceed the set threshold."

[0109] "If an abnormal emotional state is detected, automatically notify the police and send a signal to the surrounding alert system."

[0110] The above describes specific embodiments for carrying out the present invention. The present invention provides an excellent solution that enables a rapid response at the scene in order to enhance public safety.

[0111] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0112] Step 1: Collecting Image Data

[0113] The server connects to the surveillance cameras via the network and acquires image data in real time.

[0114] Input: Video data from surveillance cameras

[0115] Process: Periodically retrieve image data from surveillance cameras using HTTP requests.

[0116] Output: Acquired image data

[0117] Specific operation: The server sends a request to the surveillance camera every second and receives the latest image data. For example, "New image data has been obtained from the camera in front of the city hall."

[0118] Step 2: Saving Image Data

[0119] The server temporarily stores the acquired image data in cloud storage.

[0120] Input: Acquired image data

[0121] Processing: Save image data to cloud storage and add a timestamp to the file name.

[0122] Output: Saved image data file

[0123] Specific operation: The server saves the image data to a specific folder in storage and adds a timestamp, such as "2023-10-01_10-30-00.jpg".

[0124] Step 3: Face Detection

[0125] The server applies a face recognition algorithm (e.g., OpenCV) to the stored image data to detect human faces.

[0126] Input: Saved image data file

[0127] Processing: A face recognition algorithm detects faces in the image and draws a rectangle around them.

[0128] Output: Coordinate information of detected faces

[0129] Specific operation: The server uses OpenCV to recognize faces in the image and assumes that "a face has been detected from the upper left to the lower right of the image."

[0130] Step 4: Sentiment Analysis

[0131] The server inputs the detected facial images into a machine learning model (e.g., TensorFlow) to analyze their emotions.

[0132] Input: Face coordinate information and corresponding face image

[0133] Processing: The machine learning model is fed with facial images and classified into emotion categories (such as "joy," "anger," and "surprise").

[0134] Output: Detected sentiment data

[0135] Specific operation: The server extracts facial images and inputs them into a machine learning model to classify them as "a man's face showing the emotion of 'anger'."

[0136] Step 5: Anomaly Detection

[0137] The server analyzes fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold.

[0138] Input: Sentiment data

[0139] Processing: Analyze emotional fluctuations and evaluate rapid changes over short periods.

[0140] Output: Detection results for abnormal emotional states

[0141] Specific operation: The server analyzes emotional data from the past few minutes and detects high risk if there was a rapid shift from "joy" to "anger" in the past minute.

[0142] Step 6: Reporting and issuing alerts

[0143] When the server detects an abnormal emotional state, it automatically notifies the police and sends a signal to the surrounding alert system.

[0144] Input: Detection result of abnormal emotional state

[0145] Processing: Generates a report and sends it to the police, while simultaneously sending a signal to the surrounding alert system.

[0146] Output: Confirmation of notification transmission results and alert issuance.

[0147] Specific actions: The server uses Twilio to notify the police, reporting that "a man in front of City Hall is showing strong signs of anger." It also sends a signal to the Bosch emergency alert system, issuing an audio warning to those nearby to "be careful."

[0148] (Application Example 1)

[0149] 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."

[0150] In recent years, maintaining safety in public spaces and large facilities has become increasingly important. In particular, there is a need to prevent sudden incidents and troubles caused by rapid changes in individuals' emotions. However, conventional surveillance systems only acquire video data and do not support real-time emotion analysis or anomaly detection, making it difficult to take immediate and appropriate action. The present invention aims to solve these problems and provide a system that can more effectively ensure public safety.

[0151] 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.

[0152] In this invention, the server includes means for acquiring video data obtained from surveillance equipment, means for recognizing a person's face based on the acquired video data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying a notification device based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, and means for issuing warnings in real time via an application installed on a portable device. This enables a rapid response by analyzing a person's emotions in real time and detecting abnormalities.

[0153] "Surveillance equipment" refers to devices used to acquire video data, specifically surveillance cameras and other imaging equipment.

[0154] "Video data" refers to image and video information acquired by surveillance equipment.

[0155] "Means of recognizing a person's face" refers to algorithms or devices that have the function of detecting and identifying a specific human face from video data.

[0156] "Means of classifying emotions" refers to devices or software that analyze a person's emotions based on their recognized face and classify them into specific emotional categories (e.g., joy, anger, surprise, etc.).

[0157] "Emotional data" refers to data about a person's emotional state, obtained through methods of classifying emotions.

[0158] "Means for detecting abnormal emotional states" refers to devices or software that analyze emotional data and have the function of detecting abnormal emotional fluctuations that exceed a set threshold.

[0159] A "notification device" is a device or system that transmits information to relevant organizations or personnel when an abnormal emotional state is detected.

[0160] An "alert issuing mechanism" refers to a device or system that has the function of issuing a warning to people in the vicinity via sound or visual means when an anomaly is detected.

[0161] "Portable devices" refer to electronic devices that individuals can carry with them at all times, including smartphones and smart glasses.

[0162] An "application" is software that runs on a mobile device and provides a specific function.

[0163] "Real-time" refers to a state where the time between data acquisition and the availability of analysis results is extremely short, with virtually no delay.

[0164] The system for carrying out this invention is constructed using the following main hardware and software.

[0165] hardware

[0166] Surveillance equipment: Includes surveillance cameras and other recording devices for acquiring video data.

[0167] Terminal devices: Security staff will use smart glasses or smartphones.

[0168] Server: A computer device used for storing and analyzing video data.

[0169] software

[0170] Face recognition algorithm: An algorithm used to identify a person's face from video data.

[0171] Emotion analysis model: A model that analyzes and classifies a person's emotions based on machine learning.

[0172] Reporting system: A system that automates reporting based on detected abnormal emotional states.

[0173] Alert system: An audio or visual alert system used to warn people in the vicinity.

[0174] Mobile applications: Applications installed on the mobile devices of security staff.

[0175] Data processing and data calculation

[0176] The server streams video data acquired from surveillance equipment in real time and temporarily stores it in storage. Next, it uses a facial recognition algorithm to detect faces of people in the video. The detected facial image data is input into an emotion analysis model, where emotions are classified. The classified emotion data is analyzed, and if an abnormal emotional state exceeding a set threshold is detected, the notification system is activated, and an automatic notification and alert are issued.

[0177] Specific example

[0178] For example, imagine a security staff member at a shopping mall wearing these smart glasses while patrolling. Surveillance cameras and the smart glasses' cameras capture video data and send it to a server. The server uses a facial recognition algorithm to detect people's faces and an emotion analysis model to classify those people's emotions. If the emotion of "anger" changes rapidly within a short period of time and exceeds a threshold, the notification system is automatically activated, an alert is sent to the security center, and a warning is displayed on the smart glasses. This allows security staff to respond quickly to the scene.

[0179] Example of a prompt

[0180] "We are developing a real-time emotional anomaly detection app for smart glasses worn by security staff in shopping malls. This app analyzes emotions based on video data captured by the glasses' camera and automatically issues warnings and notifications when abnormal emotional fluctuations are detected. Please generate a specific code example."

[0181] Thus, the system for implementing this invention ensures safety in public places and large facilities, and enables the detection and response to emotional abnormalities in real time.

[0182] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0183] Step 1:

[0184] The server acquires video data in real time from surveillance equipment (such as surveillance cameras and smart glasses cameras). The input is video data, and the output is video data temporarily stored in storage. Specifically, the surveillance equipment sends frame-by-frame image data captured by the equipment to the server via the network, and the server receives and stores this data.

[0185] Step 2:

[0186] The server applies a face recognition algorithm to the acquired video data to detect human faces. The input is the video data saved in step 1, and the output is the detected face image data. Specifically, the server uses a face recognition library such as OpenCV to detect human faces in the video and extracts face location information and feature data.

[0187] Step 3:

[0188] The server inputs the detected facial image data into an emotion analysis model to classify emotions. The input is the facial image data detected in step 2, and the output is the classified emotion data. Specifically, the server uses a generative AI model to assign emotion categories such as "joy," "anger," and "surprise" to the facial image data.

[0189] Step 4:

[0190] The server analyzes the classified emotion data and detects abnormal emotional states. The input is the emotion data obtained in step 3, and the output is the detection result of abnormal emotional states. Specifically, the server compares the data with a set threshold to detect sudden changes in emotion or specific abnormal emotional states.

[0191] Step 5:

[0192] When the server detects an abnormal emotional state, it notifies the relevant agency (e.g., security center) through the notification system. The input is the abnormal emotional state detected in step 4, and the output is the notification message. Specifically, the server uses a communication protocol (e.g., an HTTP request) to send details of the detected abnormal emotional state to the notification recipient.

[0193] Step 6:

[0194] The server simultaneously issues an alert to people in the surrounding area. The input is the abnormal emotional state detected in step 4, and the output is an audible or visual warning message. Specifically, the server sends a signal to nearby alert systems (e.g., speakers, display panels) to emit an audible or visual alert.

[0195] Step 7:

[0196] The terminal sends real-time alerts to security staff through an application installed on a mobile device (e.g., smart glasses, smartphone). The input is data of the abnormal emotional state detected in step 4, and the output is a warning message displayed on the mobile device. Specifically, the application receives the data and performs actions such as sending a pop-up notification or audio notification to the user.

[0197] 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.

[0198] The system of this invention analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[0199] 1. Collection of image data

[0200] server

[0201] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0202] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[0203] 2. Face recognition

[0204] server

[0205] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0206] 3. Emotion classification

[0207] server

[0208] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[0209] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[0210] 4. Anomaly detection

[0211] server

[0212] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0213] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[0214] 5. Reporting and issuing alerts

[0215] server

[0216] If an abnormal emotional state is detected, the system automatically notifies the police. The report includes location information, a person's characteristics, and the detected emotional state.

[0217] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[0218] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[0219] Specific example

[0220] As a concrete example, consider a scenario where a specific individual suddenly displays intense anger in a public space. The server first detects the person's face and applies a facial recognition algorithm. Next, an emotion engine analyzes the facial image and classifies the emotion as "anger." If the extreme anger exceeds a threshold in a short period of time, the server determines it to be high risk and automatically notifies the police. In addition, an alert system installed in the vicinity is activated, issuing audio and visual warnings to people nearby. This allows for prompt action and prevents a dangerous situation from occurring.

[0221] As described above, the system of the present invention enhances safety in public places by working in conjunction with surveillance cameras and analyzing emotional fluctuations using an emotion engine. Since machine learning is used for analyzing emotional data, it offers high accuracy, and further improvements are expected through continuous learning. This makes it possible to detect abnormal emotional fluctuations in real time and respond quickly.

[0222] The following describes the processing flow.

[0223] Step 1:

[0224] server

[0225] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0226] Step 2:

[0227] server

[0228] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0229] Step 3:

[0230] server

[0231] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[0232] Step 4:

[0233] server

[0234] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0235] Step 5:

[0236] server

[0237] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[0238] Step 6:

[0239] server

[0240] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[0241] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[0242] (Example 2)

[0243] 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".

[0244] There is a need to improve safety in public places and prevent dangerous situations caused by sudden emotional fluctuations. However, conventional surveillance systems have difficulty analyzing a person's emotional state in real time and issuing appropriate warnings. As a result, their effectiveness in preventing unexpected incidents and crimes is limited. To solve this problem, advanced analysis technology based on image data and a mechanism that can rapidly detect abnormal emotional fluctuations are necessary.

[0245] 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.

[0246] In this invention, the server includes means for acquiring media data obtained from a monitoring device, means for detecting a person's face captured based on the acquired media data, means for analyzing the detected face and classifying the person's emotional state, means for analyzing the classified emotional data over time and detecting abnormal emotional fluctuations, means for notifying a safety agency based on the detected abnormal emotional fluctuations, and means for issuing a warning to the surrounding public. This makes it possible to quickly detect dangerous situations in public places and respond automatically.

[0247] A "monitoring device" is a device used to continuously observe and record events within a specific area.

[0248] "Media data" refers to digital data that includes visual information such as images and videos.

[0249] "Captured" means extracting and saving a specific object from an image or video.

[0250] "Detecting" means using an algorithm to identify specific information and extract it.

[0251] "Analyzing" means breaking down data and examining its contents in detail.

[0252] "Emotional state" refers to a person's psychological and emotional state, and includes "joy," "anger," "surprise," and "sadness."

[0253] "Classifying" means dividing detected data into specific categories.

[0254] "Over time" means observing fluctuations or changes in data over a certain period of time.

[0255] "Abnormal emotional fluctuations" refer to sudden and significant emotional changes that are different from the norm.

[0256] "Means of detection" refers to methods and techniques for finding specific information.

[0257] A "security agency" refers to an organization or institution, such as the police or security companies, that is responsible for ensuring public safety.

[0258] "To notify" means to transmit information.

[0259] "The public" refers to an unspecified large number of people.

[0260] "To issue a warning" means to send a message to inform someone of danger or a problem.

[0261] The system of the present invention analyzes a person's emotional state in real time based on media data acquired from a monitoring device and detects abnormal emotional fluctuations. This system is implemented through the following series of operations.

[0262] First, the server retrieves media data from the monitoring device via the network. The server uses HTTP requests to obtain images and videos from the monitoring device's API and temporarily stores this data in local storage. At this time, a timestamp is added to the media data.

[0263] Next, the server analyzes the acquired media data and detects human faces. Image processing algorithms such as the OpenCV library are used for face detection. OpenCV converts the image to grayscale and uses the Haar Cascade classifier to determine the position and size of human faces. The detected face portions are individually extracted and proceed to the next analysis step.

[0264] The server then inputs the detected facial images into the emotion engine. The emotion engine uses a learning model such as TensorFlow to analyze the emotional state. The emotional state is classified into categories such as "joy," "anger," "surprise," and "sadness." The classified emotion data is temporarily stored and monitored for changes over time.

[0265] The server analyzes changes in emotional data over time and detects abnormal emotional fluctuations based on a specific threshold. This threshold determines a high-risk situation when the set emotional score exceeds a certain standard.

[0266] If abnormal emotional fluctuations are detected, the server automatically notifies security authorities. The notification includes location information, characteristics of the detected person, and emotional state. The server also sends signals to surrounding alert systems to issue audio and visual warnings to the surrounding public.

[0267] For example, if a person standing in front of City Hall suddenly displays strong anger, the server detects their face and uses an emotion engine to identify the emotion of "anger." If extreme anger is detected in a short period of time and its fluctuation exceeds a threshold, the system automatically notifies the police and warns people in the vicinity, enabling a swift response.

[0268] This system enables the early detection of dangerous situations in public places and allows for automatic and rapid response.

[0269] Example of a prompt

[0270] The system classifies emotions from facial images of men captured by surveillance cameras in front of the city hall, and if high-risk emotions are detected, it notifies the police.

[0271] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0272] Step 1:

[0273] The server acquires media data from monitoring devices via the network.

[0274] Input: Real-time image data or video data transmitted from a monitoring device.

[0275] Process: The server periodically sends HTTP requests to obtain image data from the API of the monitoring device. At this time, the acquired data is saved with a timestamp attached.

[0276] Output: Image data file with timestamp.

[0277] Step 2:

[0278] The server detects the faces of people from the saved media data.

[0279] Input: Saved image data file.

[0280] Process: The server uses the OpenCV library to convert the image data to grayscale and performs face detection using the Haar Cascade classifier. It identifies the position and size of the detected face and cuts out the face part.

[0281] Output: Detected face image file.

[0282] Step 3:

[0283] The server analyzes the detected face image and classifies the emotional state.

[0284] Input: Cut-out face image file.

[0285] Process: The server loads the TensorFlow learning model and analyzes the face image. It resizes the face image to an appropriate size and inputs it into the learning model. The emotion engine classifies it into categories such as "happiness", "anger", "surprise", "sadness", etc.

[0286] Output: Data indicating the emotional state (e.g., result of "anger").

[0287] Step 4:

[0288] The server analyzes the emotion data and detects abnormal emotion fluctuations over time.

[0289] Input: Classified sentiment data.

[0290] Processing: The server monitors fluctuations in sentiment data and sets specific thresholds. If the set sentiment score exceeds the threshold or if there are sudden fluctuations, it is judged as high risk.

[0291] Output: Detection result of abnormal emotional fluctuations (e.g., "High Risk").

[0292] Step 5:

[0293] If the server detects abnormal emotional fluctuations, it will notify security authorities and issue a warning to the surrounding public.

[0294] Input: Abnormal emotional fluctuation detection results, location information, characteristics of the detected person.

[0295] Processing: The server sends the notification to the security agency. Furthermore, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[0296] Output: Notification to safety agencies, warning signal to alert systems.

[0297] (Application Example 2)

[0298] 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 device 14 will be referred to as the "terminal."

[0299] Conventional surveillance systems use image data acquired from surveillance cameras to recognize faces and analyze emotions, but real-time detection and response to abnormal emotional states are difficult. Furthermore, information about detected anomalies is not transmitted quickly, leading to delays in real-time response. In particular, real-time detection and notification mechanisms on portable devices such as smart glasses are not yet established, necessitating rapid risk detection and response.

[0300] In Application Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for acquiring image data obtained from a surveillance camera, means for recognizing a person's face based on the acquired image data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying the police based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, means for recognizing the faces of people in the vicinity in real time using a camera mounted on smart glasses and analyzing their emotions, means for automatically notifying the security center when an abnormal emotional state exceeding a specific threshold is detected in the emotion analysis, and means for recording the detected emotion data and using it for analysis and countermeasures at a later date. This enables real-time detection of emotional states and rapid notification in the event of an anomaly.

[0301] A "surveillance camera" is a device that photographs a specific area and acquires the footage in real time or at regular intervals.

[0302] "Image data" refers to digital data of still images or videos acquired by cameras or other recording devices.

[0303] "Human face" refers to the portion of a person's face within image data and is the object of identification and recognition.

[0304] "Means of face recognition" refers to algorithms or devices that detect a person's face from image data and identify its location and characteristics.

[0305] "Methods for classifying emotions" refer to technologies that analyze a person's emotional state from their recognized face and classify it into categories such as "joy," "anger," "surprise," and "sadness."

[0306] "Emotional data" refers to digital information that indicates a person's emotional state, generated through methods of classifying emotions.

[0307] "Abnormal emotional state" refers to a state where the fluctuations in emotional data rapidly increase or decrease beyond the normal range, indicating an elevated risk.

[0308] "Means of reporting to the police" refers to a system or device that automatically makes an emergency contact to a public agency such as the police when an abnormal emotional state is detected.

[0309] "Means of alerting people in the vicinity" refers to a technology that warns people in the vicinity by voice or visually when an abnormal emotional state is detected.

[0310] "Smart glasses" are wearable devices equipped with a camera, display, and communication function, which can provide visual assistance while displaying various information.

[0311] "Means of real-time recognition" refers to a technology that uses the camera installed in smart glasses to detect a person's face on the spot and analyze it immediately.

[0312] "Means of automatically reporting to the security center" refers to a system or device that automatically notifies the security center according to a certain protocol when an abnormal emotional state is detected.

[0313] "Means of recording data" refers to a technology or device that stores emotional data and information on abnormal detections for later analysis and countermeasure applications.

[0314] The system of the present invention is for quickly responding by analyzing a person's emotional state in real time based on the image data obtained from surveillance cameras and detecting abnormal emotional fluctuations. This system is composed of surveillance cameras, smart glasses, a server, an emotion analysis engine, a reporting system, and an alert system.

[0315] Collection of image data

[0316] The server connects to the surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The acquired image data is temporarily stored in storage.

[0317] Face recognition

[0318] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images.

[0319] emotion classification

[0320] The server extracts the recognized portion of the face and inputs it into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[0321] Anomaly detection

[0322] The server analyzes emotional fluctuations over time based on the emotional data generated by the emotion engine and sets a specific threshold. If there is a sudden emotional fluctuation that exceeds that threshold, it is judged to be an anomaly.

[0323] Reporting and alerting

[0324] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, a person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings.

[0325] Applications of smart glasses

[0326] The camera integrated into the smart glasses has the ability to recognize the faces of people in the surroundings in real time and analyze their emotions. The detected emotion data is sent to a server, and if an abnormal emotional state exceeding a certain threshold is detected based on the analysis results, an alert is automatically sent to the security center. Furthermore, the detected emotion data is recorded along with the date and time and used for later analysis and countermeasures.

[0327] Specific example

[0328] As a concrete example, imagine a scenario at a city event where security guards are wearing smart glasses. As the security guard monitors the crowd, a man suddenly displays strong anger. At this moment, the smart glasses' system detects the emotion in real time and identifies it as "anger." This data is transmitted to the security center, which automatically sends an alert. As a result, the security guard can quickly issue a warning and take necessary measures.

[0329] Example of a prompt

[0330] "A 31-year-old man is standing in front of the city hall, showing strong signs of anger. Please take immediate action."

[0331] "A woman in her late 30s or early 40s showed strong signs of sadness at the event venue. We request assistance from security personnel."

[0332] As described above, the collaboration between smart glasses and a server enables real-time detection of emotional states and rapid response. This effectively enhances safety in public places.

[0333] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0334] Step 1:

[0335] Image data collection

[0336] The server receives image data from surveillance cameras over the network. Input is streaming video transmitted from the surveillance cameras or image data that is periodically polled. This data is temporarily stored in storage. Output is image data for facial recognition.

[0337] Step 2:

[0338] Face recognition

[0339] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images. The input is image data acquired from a surveillance camera, and the output is image data of the detected face portion.

[0340] Step 3:

[0341] emotion classification

[0342] The server inputs the image data of the recognized face into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state. The input is the facial image data obtained in the face recognition step, and the output is the classified emotion data.

[0343] Step 4:

[0344] Anomaly detection

[0345] The server analyzes emotional fluctuations over time based on emotional data generated by the emotion engine. A specific threshold is set, and if a sudden emotional shift exceeding that threshold occurs, it is judged as abnormal. The input is the emotional data obtained in the emotion classification step, and the output is warning data for the detected abnormal emotional state.

[0346] Step 5:

[0347] Reporting and alerting

[0348] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, the person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings. The input is the warning data obtained in the anomaly detection step, and the output is the report and alert system activation data.

[0349] Step 6:

[0350] Data transmission for smart glasses

[0351] The camera integrated into the smart glasses recognizes the faces of people in the surroundings in real time and analyzes their emotions. The detected emotion data is sent to a server. The input is image data acquired from the smart glasses' camera, and the output is emotion data sent to the server.

[0352] Step 7:

[0353] Recording and analysis of emotional data

[0354] The server records detected emotion data along with the date and time, and uses it for later analysis and countermeasures. The input is emotion data transmitted from smart glasses and surveillance cameras, and the output is a saved file of the recorded emotion data.

[0355] Through the above processing steps, the smart glasses and server work together to enable real-time detection of emotional states and rapid response.

[0356] 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.

[0357] 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.

[0358] 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.

[0359] [Second Embodiment]

[0360] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0361] 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.

[0362] 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).

[0363] 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.

[0364] 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.

[0365] 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).

[0366] 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.

[0367] 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.

[0368] 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.

[0369] 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.

[0370] 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.

[0371] 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".

[0372] The present invention's system analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[0373] 1. Collection of image data

[0374] server

[0375] The system connects to surveillance cameras via a network and acquires image data in real time. This data is temporarily stored in storage.

[0376] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[0377] 2. Sentiment analysis

[0378] server

[0379] A face recognition algorithm is applied to the stored image data to detect the faces of people in the images.

[0380] The detected facial images are input into a machine learning model to classify emotions. Emotional classifications include categories such as "joy," "anger," and "surprise."

[0381] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[0382] 3. Anomaly detection

[0383] server

[0384] Based on emotional data, it analyzes fluctuations in those emotions and detects abnormal emotional states that exceed a set threshold.

[0385] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[0386] 4. Reporting and issuing alerts

[0387] server

[0388] If an abnormal emotional state is detected, an automatic report will be sent to the police. The report will include location information, a person's characteristics, and their emotional state.

[0389] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[0390] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[0391] Thus, the system of the present invention operates in conjunction with surveillance cameras and improves safety in public places by analyzing emotional fluctuations in real time. This makes it possible to quickly capture sudden changes in emotions and respond promptly accordingly. Furthermore, since machine learning models are used to analyze emotional data, accuracy improves through continuous data learning, enabling more effective risk management.

[0392] As a concrete example, consider a scenario where a specific individual suddenly displays strong anger in a public space. The server first detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be a potential danger. In this case, the server automatically notifies the police and also sends a signal to surrounding alert systems. This allows for immediate action to be taken, preventing a dangerous situation from occurring.

[0393] The above describes the basic form and operation for carrying out the present invention. This system is expected to have a wide range of applications as an effective tool for enhancing public safety.

[0394] The following describes the processing flow.

[0395] Step 1:

[0396] server

[0397] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0398] Step 2:

[0399] server

[0400] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0401] Step 3:

[0402] server

[0403] The detected facial portion is extracted and input into a machine learning model (for example, a deep learning-based emotion classification model). This model analyzes the facial features and classifies the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[0404] Step 4:

[0405] server

[0406] Based on classified emotion data, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0407] Step 5:

[0408] server

[0409] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[0410] Step 6:

[0411] server

[0412] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[0413] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[0414] (Example 1)

[0415] 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".

[0416] Traditional surveillance systems often relied on manual processes for identifying individuals and detecting abnormal behavior, resulting in significant labor and time constraints. Furthermore, real-time analysis of abnormal emotional fluctuations and rapid response were challenging. Therefore, there was a need for effective methods to improve safety in public spaces.

[0417] 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.

[0418] In this invention, the server includes means for acquiring image data from a surveillance camera, means for detecting a person's face based on the stored image data, and means for inputting the detected face into a machine learning model to classify emotions. This makes it possible to detect abnormal emotional states in real time and automatically notify the police. Furthermore, by quickly issuing alerts to people in the vicinity, dangerous situations can be prevented.

[0419] A "surveillance camera" is a device that continuously photographs a designated area or object and acquires video data.

[0420] "Image data" refers to visual information acquired by cameras and other recording devices, stored in a digital format.

[0421] "Storage" refers to a memory device used to temporarily or permanently store digital data.

[0422] A "face recognition algorithm" is a computational method for detecting the face portion of a person from image data and identifying specific features.

[0423] A "machine learning model" is an algorithm or a set of algorithms that learns from data and makes predictions about new data.

[0424] "Emotion classification" is the process of categorizing the emotions a person is expressing based on characteristic data obtained from their face.

[0425] An "abnormal emotional state" refers to an emotional intensity or fluctuation that exceeds the normal range and may indicate a potential danger.

[0426] "Reporting" refers to the act of automatically issuing a warning to relevant authorities, such as the police, when a monitoring system detects an anomaly.

[0427] An "alert" is a warning signal that is issued to alert people in the vicinity when an abnormal situation is detected.

[0428] A "server" is a computer system that manages and controls data processing, storage, and communication on a network.

[0429] This invention relates to a system that analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of a surveillance camera, a server, and related software.

[0430] 1. System Configuration

[0431] The system includes surveillance cameras, servers, storage, and network infrastructure as its main components. The servers play a central role in acquiring image data from the surveillance cameras and storing that data in storage. Specific hardware and software used include surveillance cameras (e.g., network cameras), network infrastructure (e.g., routers), servers (e.g., general-purpose servers), and storage (e.g., cloud storage).

[0432] 2. Data Collection and Storage

[0433] The server connects to the surveillance camera via a network and acquires image data in real time. The acquired image data is stored in cloud storage. The server also adds a timestamp to the image data before saving it.

[0434] 3. Face detection and sentiment analysis

[0435] The server uses a face recognition algorithm (e.g., OpenCV) to detect human faces based on the stored image data. The detected face images are then input into a machine learning model (e.g., TensorFlow), where emotions are categorized into categories such as "joy," "anger," and "surprise."

[0436] 4. Detection of abnormal emotions

[0437] The server analyzes emotional fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold. If extreme emotional changes are observed in a short period of time, it is assessed as high risk.

[0438] 5. Notification and Alert System

[0439] The server automatically notifies the police if it detects an abnormal emotional state. The report includes location information, the person's characteristics, and their emotional state. Simultaneously, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[0440] Specific example

[0441] For example, if a particular individual suddenly displays strong anger in a public space, the server detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be high risk. In that case, the server automatically notifies the police and sends a warning signal to the surrounding alert system. This allows for immediate action and prevents dangerous situations from occurring.

[0442] Example of a prompt

[0443] "Collect image data from surveillance cameras in real time and save it on the server."

[0444] "Use image data stored on the server, apply a facial recognition algorithm, and input the results into a machine learning model to classify emotions."

[0445] "Based on the results of the sentiment analysis, analyze the emotional fluctuations and detect any anomalies that exceed the set threshold."

[0446] "If an abnormal emotional state is detected, automatically notify the police and send a signal to the surrounding alert system."

[0447] The above describes specific embodiments for carrying out the present invention. The present invention provides an excellent solution that enables a rapid response at the scene in order to enhance public safety.

[0448] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0449] Step 1: Collecting Image Data

[0450] The server connects to the surveillance cameras via the network and acquires image data in real time.

[0451] Input: Video data from surveillance cameras

[0452] Process: Periodically retrieve image data from surveillance cameras using HTTP requests.

[0453] Output: Acquired image data

[0454] Specific operation: The server sends a request to the surveillance camera every second and receives the latest image data. For example, "New image data has been obtained from the camera in front of the city hall."

[0455] Step 2: Saving Image Data

[0456] The server temporarily stores the acquired image data in cloud storage.

[0457] Input: Acquired image data

[0458] Processing: Save image data to cloud storage and add a timestamp to the file name.

[0459] Output: Saved image data file

[0460] Specific operation: The server saves the image data to a specific folder in storage and adds a timestamp, such as "2023-10-01_10-30-00.jpg".

[0461] Step 3: Face Detection

[0462] The server applies a face recognition algorithm (e.g., OpenCV) to the stored image data to detect human faces.

[0463] Input: Saved image data file

[0464] Processing: A face recognition algorithm detects faces in the image and draws a rectangle around them.

[0465] Output: Coordinate information of detected faces

[0466] Specific operation: The server uses OpenCV to recognize faces in the image and assumes that "a face has been detected from the upper left to the lower right of the image."

[0467] Step 4: Sentiment Analysis

[0468] The server inputs the detected facial images into a machine learning model (e.g., TensorFlow) to analyze their emotions.

[0469] Input: Face coordinate information and corresponding face image

[0470] Processing: The machine learning model is fed with facial images and classified into emotion categories (such as "joy," "anger," and "surprise").

[0471] Output: Detected sentiment data

[0472] Specific operation: The server extracts facial images and inputs them into a machine learning model to classify them as "a man's face showing the emotion of 'anger'."

[0473] Step 5: Anomaly Detection

[0474] The server analyzes fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold.

[0475] Input: Sentiment data

[0476] Processing: Analyze emotional fluctuations and evaluate rapid changes over short periods.

[0477] Output: Detection results for abnormal emotional states

[0478] Specific operation: The server analyzes emotional data from the past few minutes and detects high risk if there was a rapid shift from "joy" to "anger" in the past minute.

[0479] Step 6: Reporting and issuing alerts

[0480] When the server detects an abnormal emotional state, it automatically notifies the police and sends a signal to the surrounding alert system.

[0481] Input: Detection result of abnormal emotional state

[0482] Processing: Generates a report and sends it to the police, while simultaneously sending a signal to the surrounding alert system.

[0483] Output: Confirmation of notification transmission results and alert issuance.

[0484] Specific actions: The server uses Twilio to notify the police, reporting that "a man in front of City Hall is showing strong signs of anger." It also sends a signal to the Bosch emergency alert system, issuing an audio warning to those nearby to "be careful."

[0485] (Application Example 1)

[0486] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0487] In recent years, maintaining safety in public spaces and large facilities has become increasingly important. In particular, there is a need to prevent sudden incidents and troubles caused by rapid changes in individuals' emotions. However, conventional surveillance systems only acquire video data and do not support real-time emotion analysis or anomaly detection, making it difficult to take immediate and appropriate action. The present invention aims to solve these problems and provide a system that can more effectively ensure public safety.

[0488] 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.

[0489] In this invention, the server includes means for acquiring video data obtained from surveillance equipment, means for recognizing a person's face based on the acquired video data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying a notification device based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, and means for issuing warnings in real time via an application installed on a portable device. This enables a rapid response by analyzing a person's emotions in real time and detecting abnormalities.

[0490] "Surveillance equipment" refers to devices used to acquire video data, specifically surveillance cameras and other imaging equipment.

[0491] "Video data" refers to image and video information acquired by surveillance equipment.

[0492] "Means of recognizing a person's face" refers to algorithms or devices that have the function of detecting and identifying a specific human face from video data.

[0493] "Means of classifying emotions" refers to devices or software that analyze a person's emotions based on their recognized face and classify them into specific emotional categories (e.g., joy, anger, surprise, etc.).

[0494] "Emotional data" refers to data about a person's emotional state, obtained through methods of classifying emotions.

[0495] "Means for detecting abnormal emotional states" refers to devices or software that analyze emotional data and have the function of detecting abnormal emotional fluctuations that exceed a set threshold.

[0496] A "notification device" is a device or system that transmits information to relevant organizations or personnel when an abnormal emotional state is detected.

[0497] An "alert issuing mechanism" refers to a device or system that has the function of issuing a warning to people in the vicinity via sound or visual means when an anomaly is detected.

[0498] "Portable devices" refer to electronic devices that individuals can carry with them at all times, including smartphones and smart glasses.

[0499] An "application" is software that runs on a mobile device and provides a specific function.

[0500] "Real-time" refers to a state where the time between data acquisition and the availability of analysis results is extremely short, with virtually no delay.

[0501] The system for carrying out this invention is constructed using the following main hardware and software.

[0502] hardware

[0503] Surveillance equipment: Includes surveillance cameras and other recording devices for acquiring video data.

[0504] Terminal devices: Security staff will use smart glasses or smartphones.

[0505] Server: A computer device used for storing and analyzing video data.

[0506] software

[0507] Face recognition algorithm: An algorithm used to identify a person's face from video data.

[0508] Emotion analysis model: A model that analyzes and classifies a person's emotions based on machine learning.

[0509] Reporting system: A system that automates reporting based on detected abnormal emotional states.

[0510] Alert system: An audio or visual alert system used to warn people in the vicinity.

[0511] Mobile applications: Applications installed on the mobile devices of security staff.

[0512] Data processing and data calculation

[0513] The server streams video data acquired from surveillance equipment in real time and temporarily stores it in storage. Next, it uses a facial recognition algorithm to detect faces of people in the video. The detected facial image data is input into an emotion analysis model, where emotions are classified. The classified emotion data is analyzed, and if an abnormal emotional state exceeding a set threshold is detected, the notification system is activated, and an automatic notification and alert are issued.

[0514] Specific example

[0515] For example, imagine a security staff member at a shopping mall wearing these smart glasses while patrolling. Surveillance cameras and the smart glasses' cameras capture video data and send it to a server. The server uses a facial recognition algorithm to detect people's faces and an emotion analysis model to classify those people's emotions. If the emotion of "anger" changes rapidly within a short period of time and exceeds a threshold, the notification system is automatically activated, an alert is sent to the security center, and a warning is displayed on the smart glasses. This allows security staff to respond quickly to the scene.

[0516] Example of a prompt

[0517] "We are developing a real-time emotional anomaly detection app for smart glasses worn by security staff in shopping malls. This app analyzes emotions based on video data captured by the glasses' camera and automatically issues warnings and notifications when abnormal emotional fluctuations are detected. Please generate a specific code example."

[0518] Thus, the system for implementing this invention ensures safety in public places and large facilities, and enables the detection and response to emotional abnormalities in real time.

[0519] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0520] Step 1:

[0521] The server acquires video data in real time from surveillance equipment (such as surveillance cameras and smart glasses cameras). The input is video data, and the output is video data temporarily stored in storage. Specifically, the surveillance equipment sends frame-by-frame image data captured by the equipment to the server via the network, and the server receives and stores this data.

[0522] Step 2:

[0523] The server applies a face recognition algorithm to the acquired video data to detect human faces. The input is the video data saved in step 1, and the output is the detected face image data. Specifically, the server uses a face recognition library such as OpenCV to detect human faces in the video and extracts face location information and feature data.

[0524] Step 3:

[0525] The server inputs the detected facial image data into an emotion analysis model to classify emotions. The input is the facial image data detected in step 2, and the output is the classified emotion data. Specifically, the server uses a generative AI model to assign emotion categories such as "joy," "anger," and "surprise" to the facial image data.

[0526] Step 4:

[0527] The server analyzes the classified emotion data and detects abnormal emotional states. The input is the emotion data obtained in step 3, and the output is the detection result of abnormal emotional states. Specifically, the server compares the data with a set threshold to detect sudden changes in emotion or specific abnormal emotional states.

[0528] Step 5:

[0529] When the server detects an abnormal emotional state, it notifies the relevant agency (e.g., security center) through the notification system. The input is the abnormal emotional state detected in step 4, and the output is the notification message. Specifically, the server uses a communication protocol (e.g., an HTTP request) to send details of the detected abnormal emotional state to the notification recipient.

[0530] Step 6:

[0531] The server simultaneously issues an alert to people in the surrounding area. The input is the abnormal emotional state detected in step 4, and the output is an audible or visual warning message. Specifically, the server sends a signal to nearby alert systems (e.g., speakers, display panels) to emit an audible or visual alert.

[0532] Step 7:

[0533] The terminal sends real-time alerts to security staff through an application installed on a mobile device (e.g., smart glasses, smartphone). The input is data of the abnormal emotional state detected in step 4, and the output is a warning message displayed on the mobile device. Specifically, the application receives the data and performs actions such as sending a pop-up notification or audio notification to the user.

[0534] 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.

[0535] The system of this invention analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[0536] 1. Collection of image data

[0537] server

[0538] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0539] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[0540] 2. Face recognition

[0541] server

[0542] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0543] 3. Emotion classification

[0544] server

[0545] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[0546] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[0547] 4. Anomaly detection

[0548] server

[0549] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0550] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[0551] 5. Reporting and issuing alerts

[0552] server

[0553] If an abnormal emotional state is detected, the system automatically notifies the police. The report includes location information, a person's characteristics, and the detected emotional state.

[0554] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[0555] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[0556] Specific example

[0557] As a concrete example, consider a scenario where a specific individual suddenly displays intense anger in a public space. The server first detects the person's face and applies a facial recognition algorithm. Next, an emotion engine analyzes the facial image and classifies the emotion as "anger." If the extreme anger exceeds a threshold in a short period of time, the server determines it to be high risk and automatically notifies the police. In addition, an alert system installed in the vicinity is activated, issuing audio and visual warnings to people nearby. This allows for prompt action and prevents a dangerous situation from occurring.

[0558] As described above, the system of the present invention enhances safety in public places by working in conjunction with surveillance cameras and analyzing emotional fluctuations using an emotion engine. Since machine learning is used for analyzing emotional data, it offers high accuracy, and further improvements are expected through continuous learning. This makes it possible to detect abnormal emotional fluctuations in real time and respond quickly.

[0559] The following describes the processing flow.

[0560] Step 1:

[0561] server

[0562] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0563] Step 2:

[0564] server

[0565] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0566] Step 3:

[0567] server

[0568] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[0569] Step 4:

[0570] server

[0571] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0572] Step 5:

[0573] server

[0574] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[0575] Step 6:

[0576] server

[0577] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[0578] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[0579] (Example 2)

[0580] 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".

[0581] There is a need to improve safety in public places and prevent dangerous situations caused by sudden emotional fluctuations. However, conventional surveillance systems have difficulty analyzing a person's emotional state in real time and issuing appropriate warnings. As a result, their effectiveness in preventing unexpected incidents and crimes is limited. To solve this problem, advanced analysis technology based on image data and a mechanism that can rapidly detect abnormal emotional fluctuations are necessary.

[0582] 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.

[0583] In this invention, the server includes means for acquiring media data obtained from a monitoring device, means for detecting a person's face captured based on the acquired media data, means for analyzing the detected face and classifying the person's emotional state, means for analyzing the classified emotional data over time and detecting abnormal emotional fluctuations, means for notifying a safety agency based on the detected abnormal emotional fluctuations, and means for issuing a warning to the surrounding public. This makes it possible to quickly detect dangerous situations in public places and respond automatically.

[0584] A "monitoring device" is a device used to continuously observe and record events within a specific area.

[0585] "Media data" refers to digital data that includes visual information such as images and videos.

[0586] "Captured" means extracting and saving a specific object from an image or video.

[0587] "Detecting" means using an algorithm to identify specific information and extract it.

[0588] "Analyzing" means breaking down data and examining its contents in detail.

[0589] "Emotional state" refers to a person's psychological and emotional state, and includes "joy," "anger," "surprise," and "sadness."

[0590] "Classifying" means dividing detected data into specific categories.

[0591] "Over time" means observing fluctuations or changes in data over a certain period of time.

[0592] "Abnormal emotional fluctuations" refer to sudden and significant emotional changes that are different from the norm.

[0593] "Means of detection" refers to methods and techniques for finding specific information.

[0594] A "security agency" refers to an organization or institution, such as the police or security companies, that is responsible for ensuring public safety.

[0595] "To notify" means to transmit information.

[0596] "The public" refers to an unspecified large number of people.

[0597] "To issue a warning" means to send a message to inform someone of danger or a problem.

[0598] The system of the present invention analyzes a person's emotional state in real time based on media data acquired from a monitoring device and detects abnormal emotional fluctuations. This system is implemented through the following series of operations.

[0599] First, the server retrieves media data from the monitoring device via the network. The server uses HTTP requests to obtain images and videos from the monitoring device's API and temporarily stores this data in local storage. At this time, a timestamp is added to the media data.

[0600] Next, the server analyzes the acquired media data and detects human faces. Image processing algorithms such as the OpenCV library are used for face detection. OpenCV converts the image to grayscale and uses the Haar Cascade classifier to determine the position and size of human faces. The detected face portions are individually extracted and proceed to the next analysis step.

[0601] The server then inputs the detected facial images into the emotion engine. The emotion engine uses a learning model such as TensorFlow to analyze the emotional state. The emotional state is classified into categories such as "joy," "anger," "surprise," and "sadness." The classified emotion data is temporarily stored and monitored for changes over time.

[0602] The server analyzes changes in emotional data over time and detects abnormal emotional fluctuations based on a specific threshold. This threshold determines a high-risk situation when the set emotional score exceeds a certain standard.

[0603] If abnormal emotional fluctuations are detected, the server automatically notifies security authorities. The notification includes location information, characteristics of the detected person, and emotional state. The server also sends signals to surrounding alert systems to issue audio and visual warnings to the surrounding public.

[0604] For example, if a person standing in front of City Hall suddenly displays strong anger, the server detects their face and uses an emotion engine to identify the emotion of "anger." If extreme anger is detected in a short period of time and its fluctuation exceeds a threshold, the system automatically notifies the police and warns people in the vicinity, enabling a swift response.

[0605] This system enables the early detection of dangerous situations in public places and allows for automatic and rapid response.

[0606] Example of a prompt

[0607] The system classifies emotions from facial images of men captured by surveillance cameras in front of the city hall, and if high-risk emotions are detected, it notifies the police.

[0608] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0609] Step 1:

[0610] The server acquires media data from monitoring devices via the network.

[0611] Input: Real-time image data or video data transmitted from a monitoring device.

[0612] Processing: The server periodically sends HTTP requests to retrieve image data from the monitoring device's API. The retrieved data is then time-stamped and saved.

[0613] Output: Image data file with timestamp.

[0614] Step 2:

[0615] The server detects human faces from stored media data.

[0616] Input: Saved image data file.

[0617] Processing: The server uses the OpenCV library to convert image data to grayscale and performs face detection using the Haar Cascade classifier. It identifies the position and size of the detected faces and extracts the face portion.

[0618] Output: Detected face image file.

[0619] Step 3:

[0620] The server analyzes the detected facial images and classifies their emotional states.

[0621] Input: Extracted face image file.

[0622] Processing: The server loads a TensorFlow trained model and analyzes the face image. The face image is resized to an appropriate size and input into the trained model. The emotion engine classifies the image into categories such as "joy," "anger," "surprise," and "sadness."

[0623] Output: Data indicating emotional state (e.g., result "anger").

[0624] Step 4:

[0625] The server analyzes emotional data and detects abnormal emotional fluctuations over time.

[0626] Input: Classified sentiment data.

[0627] Processing: The server monitors fluctuations in sentiment data and sets specific thresholds. If the set sentiment score exceeds the threshold or if there are sudden fluctuations, it is judged as high risk.

[0628] Output: Detection result of abnormal emotional fluctuations (e.g., "High Risk").

[0629] Step 5:

[0630] If the server detects abnormal emotional fluctuations, it will notify security authorities and issue a warning to the surrounding public.

[0631] Input: Abnormal emotional fluctuation detection results, location information, characteristics of the detected person.

[0632] Processing: The server sends the notification to the security agency. Furthermore, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[0633] Output: Notification to safety agencies, warning signal to alert systems.

[0634] (Application Example 2)

[0635] 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."

[0636] Conventional surveillance systems use image data acquired from surveillance cameras to recognize faces and analyze emotions, but real-time detection and response to abnormal emotional states are difficult. Furthermore, information about detected anomalies is not transmitted quickly, leading to delays in real-time response. In particular, real-time detection and notification mechanisms on portable devices such as smart glasses are not yet established, necessitating rapid risk detection and response.

[0637] In Application Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for acquiring image data obtained from a surveillance camera, means for recognizing a person's face based on the acquired image data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying the police based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, means for recognizing the faces of people in the vicinity in real time using a camera mounted on smart glasses and analyzing their emotions, means for automatically notifying the security center when an abnormal emotional state exceeding a specific threshold is detected in the emotion analysis, and means for recording the detected emotion data and using it for analysis and countermeasures at a later date. This enables real-time detection of emotional states and rapid notification in the event of an anomaly.

[0638] A "surveillance camera" is a device that photographs a specific area and acquires the footage in real time or at regular intervals.

[0639] "Image data" refers to digital data of still images or videos acquired by cameras or other recording devices.

[0640] "Human face" refers to the portion of a person's face within image data and is the object of identification and recognition.

[0641] "Means of face recognition" refers to algorithms or devices that detect a person's face from image data and identify its location and characteristics.

[0642] "Methods for classifying emotions" refer to technologies that analyze a person's emotional state from their recognized face and classify it into categories such as "joy," "anger," "surprise," and "sadness."

[0643] "Emotional data" refers to digital information that indicates a person's emotional state, generated through methods of classifying emotions.

[0644] An "abnormal emotional state" refers to a condition where fluctuations in emotional data increase or decrease rapidly beyond the normal range, leading to an increased risk.

[0645] "Means of reporting to the police" refers to a system or device that automatically makes an emergency contact with the police or other public authorities when an abnormal emotional state is detected.

[0646] "Means of alerting people in the vicinity" refers to technology that notifies people in the vicinity of an abnormal emotional state through sound or visual means.

[0647] "Smart glasses" are wearable devices equipped with cameras, displays, and communication functions that can provide visual assistance by displaying various information.

[0648] "Real-time recognition" refers to a technology that uses a camera mounted on smart glasses to detect a person's face on the spot and analyze it immediately.

[0649] "Means of automatically notifying the security center" refers to a system or device that automatically notifies the security center according to a specific protocol when an abnormal emotional state is detected.

[0650] "Means of recording data" refers to technologies or devices that store emotional data or anomaly detection information so that it can be used for analysis and countermeasures at a later date.

[0651] The present invention provides a system for rapidly responding to abnormal emotional fluctuations by analyzing a person's emotional state in real time based on image data acquired from a surveillance camera. This system consists of a surveillance camera, smart glasses, a server, an emotion analysis engine, a notification system, and an alert system.

[0652] Image data collection

[0653] The server connects to the surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The acquired image data is temporarily stored in storage.

[0654] Face recognition

[0655] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images.

[0656] emotion classification

[0657] The server extracts the recognized portion of the face and inputs it into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[0658] Anomaly detection

[0659] The server analyzes emotional fluctuations over time based on the emotional data generated by the emotion engine and sets a specific threshold. If there is a sudden emotional fluctuation that exceeds that threshold, it is judged to be an anomaly.

[0660] Reporting and alerting

[0661] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, a person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings.

[0662] Applications of smart glasses

[0663] The camera integrated into the smart glasses has the ability to recognize the faces of people in the surroundings in real time and analyze their emotions. The detected emotion data is sent to a server, and if an abnormal emotional state exceeding a certain threshold is detected based on the analysis results, an alert is automatically sent to the security center. Furthermore, the detected emotion data is recorded along with the date and time and used for later analysis and countermeasures.

[0664] Specific example

[0665] As a concrete example, imagine a scenario at a city event where security guards are wearing smart glasses. As the security guard monitors the crowd, a man suddenly displays strong anger. At this moment, the smart glasses' system detects the emotion in real time and identifies it as "anger." This data is transmitted to the security center, which automatically sends an alert. As a result, the security guard can quickly issue a warning and take necessary measures.

[0666] Example of a prompt

[0667] "A 31-year-old man is standing in front of the city hall, showing strong signs of anger. Please take immediate action."

[0668] "A woman in her late 30s or early 40s showed strong signs of sadness at the event venue. We request assistance from security personnel."

[0669] As described above, the collaboration between smart glasses and a server enables real-time detection of emotional states and rapid response. This effectively enhances safety in public places.

[0670] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0671] Step 1:

[0672] Image data collection

[0673] The server receives image data from surveillance cameras over the network. Input is streaming video transmitted from the surveillance cameras or image data that is periodically polled. This data is temporarily stored in storage. Output is image data for facial recognition.

[0674] Step 2:

[0675] Face recognition

[0676] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images. The input is image data acquired from a surveillance camera, and the output is image data of the detected face portion.

[0677] Step 3:

[0678] emotion classification

[0679] The server inputs the image data of the recognized face into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state. The input is the facial image data obtained in the face recognition step, and the output is the classified emotion data.

[0680] Step 4:

[0681] Anomaly detection

[0682] The server analyzes emotional fluctuations over time based on emotional data generated by the emotion engine. A specific threshold is set, and if a sudden emotional shift exceeding that threshold occurs, it is judged as abnormal. The input is the emotional data obtained in the emotion classification step, and the output is warning data for the detected abnormal emotional state.

[0683] Step 5:

[0684] Reporting and alerting

[0685] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, the person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings. The input is the warning data obtained in the anomaly detection step, and the output is the report and alert system activation data.

[0686] Step 6:

[0687] Data transmission for smart glasses

[0688] The camera integrated into the smart glasses recognizes the faces of people in the surroundings in real time and analyzes their emotions. The detected emotion data is sent to a server. The input is image data acquired from the smart glasses' camera, and the output is emotion data sent to the server.

[0689] Step 7:

[0690] Recording and analysis of emotional data

[0691] The server records detected emotion data along with the date and time, and uses it for later analysis and countermeasures. The input is emotion data transmitted from smart glasses and surveillance cameras, and the output is a saved file of the recorded emotion data.

[0692] Through the above processing steps, the smart glasses and server work together to enable real-time detection of emotional states and rapid response.

[0693] 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.

[0694] 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.

[0695] 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.

[0696] [Third Embodiment]

[0697] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0698] 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.

[0699] 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).

[0700] 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.

[0701] 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.

[0702] 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).

[0703] 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.

[0704] 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.

[0705] 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.

[0706] 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.

[0707] 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.

[0708] 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".

[0709] The present invention's system analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[0710] 1. Collection of image data

[0711] server

[0712] The system connects to surveillance cameras via a network and acquires image data in real time. This data is temporarily stored in storage.

[0713] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[0714] 2. Sentiment analysis

[0715] server

[0716] A face recognition algorithm is applied to the stored image data to detect the faces of people in the images.

[0717] The detected facial images are input into a machine learning model to classify emotions. Emotional classifications include categories such as "joy," "anger," and "surprise."

[0718] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[0719] 3. Anomaly detection

[0720] server

[0721] Based on emotional data, it analyzes fluctuations in those emotions and detects abnormal emotional states that exceed a set threshold.

[0722] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[0723] 4. Reporting and issuing alerts

[0724] server

[0725] If an abnormal emotional state is detected, an automatic report will be sent to the police. The report will include location information, a person's characteristics, and their emotional state.

[0726] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[0727] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[0728] Thus, the system of the present invention operates in conjunction with surveillance cameras and improves safety in public places by analyzing emotional fluctuations in real time. This makes it possible to quickly capture sudden changes in emotions and respond promptly accordingly. Furthermore, since machine learning models are used to analyze emotional data, accuracy improves through continuous data learning, enabling more effective risk management.

[0729] As a concrete example, consider a scenario where a specific individual suddenly displays strong anger in a public space. The server first detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be a potential danger. In this case, the server automatically notifies the police and also sends a signal to surrounding alert systems. This allows for immediate action to be taken, preventing a dangerous situation from occurring.

[0730] The above describes the basic form and operation for carrying out the present invention. This system is expected to have a wide range of applications as an effective tool for enhancing public safety.

[0731] The following describes the processing flow.

[0732] Step 1:

[0733] server

[0734] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0735] Step 2:

[0736] server

[0737] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0738] Step 3:

[0739] server

[0740] The detected facial portion is extracted and input into a machine learning model (for example, a deep learning-based emotion classification model). This model analyzes the facial features and classifies the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[0741] Step 4:

[0742] server

[0743] Based on classified emotion data, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0744] Step 5:

[0745] server

[0746] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[0747] Step 6:

[0748] server

[0749] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[0750] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[0751] (Example 1)

[0752] 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."

[0753] Traditional surveillance systems often relied on manual processes for identifying individuals and detecting abnormal behavior, resulting in significant labor and time constraints. Furthermore, real-time analysis of abnormal emotional fluctuations and rapid response were challenging. Therefore, there was a need for effective methods to improve safety in public spaces.

[0754] 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.

[0755] In this invention, the server includes means for acquiring image data from a surveillance camera, means for detecting a person's face based on the stored image data, and means for inputting the detected face into a machine learning model to classify emotions. This makes it possible to detect abnormal emotional states in real time and automatically notify the police. Furthermore, by quickly issuing alerts to people in the vicinity, dangerous situations can be prevented.

[0756] A "surveillance camera" is a device that continuously photographs a designated area or object and acquires video data.

[0757] "Image data" refers to visual information acquired by cameras and other recording devices, stored in a digital format.

[0758] "Storage" refers to a memory device used to temporarily or permanently store digital data.

[0759] A "face recognition algorithm" is a computational method for detecting the face portion of a person from image data and identifying specific features.

[0760] A "machine learning model" is an algorithm or a set of algorithms that learns from data and makes predictions about new data.

[0761] "Emotion classification" is the process of categorizing the emotions a person is expressing based on characteristic data obtained from their face.

[0762] An "abnormal emotional state" refers to an emotional intensity or fluctuation that exceeds the normal range and may indicate a potential danger.

[0763] "Reporting" refers to the act of automatically issuing a warning to relevant authorities, such as the police, when a monitoring system detects an anomaly.

[0764] An "alert" is a warning signal that is issued to alert people in the vicinity when an abnormal situation is detected.

[0765] A "server" is a computer system that manages and controls data processing, storage, and communication on a network.

[0766] This invention relates to a system that analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of a surveillance camera, a server, and related software.

[0767] 1. System Configuration

[0768] The system includes surveillance cameras, servers, storage, and network infrastructure as its main components. The servers play a central role in acquiring image data from the surveillance cameras and storing that data in storage. Specific hardware and software used include surveillance cameras (e.g., network cameras), network infrastructure (e.g., routers), servers (e.g., general-purpose servers), and storage (e.g., cloud storage).

[0769] 2. Data Collection and Storage

[0770] The server connects to the surveillance camera via a network and acquires image data in real time. The acquired image data is stored in cloud storage. The server also adds a timestamp to the image data before saving it.

[0771] 3. Face detection and sentiment analysis

[0772] The server uses a face recognition algorithm (e.g., OpenCV) to detect human faces based on the stored image data. The detected face images are then input into a machine learning model (e.g., TensorFlow), where emotions are categorized into categories such as "joy," "anger," and "surprise."

[0773] 4. Detection of abnormal emotions

[0774] The server analyzes emotional fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold. If extreme emotional changes are observed in a short period of time, it is assessed as high risk.

[0775] 5. Notification and Alert System

[0776] The server automatically notifies the police if it detects an abnormal emotional state. The report includes location information, the person's characteristics, and their emotional state. Simultaneously, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[0777] Specific example

[0778] For example, if a particular individual suddenly displays strong anger in a public space, the server detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be high risk. In that case, the server automatically notifies the police and sends a warning signal to the surrounding alert system. This allows for immediate action and prevents dangerous situations from occurring.

[0779] Example of a prompt

[0780] "Collect image data from surveillance cameras in real time and save it on the server."

[0781] "Use image data stored on the server, apply a facial recognition algorithm, and input the results into a machine learning model to classify emotions."

[0782] "Based on the results of the sentiment analysis, analyze the emotional fluctuations and detect any anomalies that exceed the set threshold."

[0783] "If an abnormal emotional state is detected, automatically notify the police and send a signal to the surrounding alert system."

[0784] The above describes specific embodiments for carrying out the present invention. The present invention provides an excellent solution that enables a rapid response at the scene in order to enhance public safety.

[0785] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0786] Step 1: Collecting Image Data

[0787] The server connects to the surveillance cameras via the network and acquires image data in real time.

[0788] Input: Video data from surveillance cameras

[0789] Process: Periodically retrieve image data from surveillance cameras using HTTP requests.

[0790] Output: Acquired image data

[0791] Specific operation: The server sends a request to the surveillance camera every second and receives the latest image data. For example, "New image data has been obtained from the camera in front of the city hall."

[0792] Step 2: Saving Image Data

[0793] The server temporarily stores the acquired image data in cloud storage.

[0794] Input: Acquired image data

[0795] Processing: Save image data to cloud storage and add a timestamp to the file name.

[0796] Output: Saved image data file

[0797] Specific operation: The server saves the image data to a specific folder in storage and adds a timestamp, such as "2023-10-01_10-30-00.jpg".

[0798] Step 3: Face Detection

[0799] The server applies a face recognition algorithm (e.g., OpenCV) to the stored image data to detect human faces.

[0800] Input: Saved image data file

[0801] Processing: A face recognition algorithm detects faces in the image and draws a rectangle around them.

[0802] Output: Coordinate information of detected faces

[0803] Specific operation: The server uses OpenCV to recognize faces in the image and assumes that "a face has been detected from the upper left to the lower right of the image."

[0804] Step 4: Sentiment Analysis

[0805] The server inputs the detected facial images into a machine learning model (e.g., TensorFlow) to analyze their emotions.

[0806] Input: Face coordinate information and corresponding face image

[0807] Processing: The machine learning model is fed with facial images and classified into emotion categories (such as "joy," "anger," and "surprise").

[0808] Output: Detected sentiment data

[0809] Specific operation: The server extracts facial images and inputs them into a machine learning model to classify them as "a man's face showing the emotion of 'anger'."

[0810] Step 5: Anomaly Detection

[0811] The server analyzes fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold.

[0812] Input: Sentiment data

[0813] Processing: Analyze emotional fluctuations and evaluate rapid changes over short periods.

[0814] Output: Detection results for abnormal emotional states

[0815] Specific operation: The server analyzes emotional data from the past few minutes and detects high risk if there was a rapid shift from "joy" to "anger" in the past minute.

[0816] Step 6: Reporting and issuing alerts

[0817] When the server detects an abnormal emotional state, it automatically notifies the police and sends a signal to the surrounding alert system.

[0818] Input: Detection result of abnormal emotional state

[0819] Processing: Generates a report and sends it to the police, while simultaneously sending a signal to the surrounding alert system.

[0820] Output: Confirmation of notification transmission results and alert issuance.

[0821] Specific actions: The server uses Twilio to notify the police, reporting that "a man in front of City Hall is showing strong signs of anger." It also sends a signal to the Bosch emergency alert system, issuing an audio warning to those nearby to "be careful."

[0822] (Application Example 1)

[0823] 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."

[0824] In recent years, maintaining safety in public spaces and large facilities has become increasingly important. In particular, there is a need to prevent sudden incidents and troubles caused by rapid changes in individuals' emotions. However, conventional surveillance systems only acquire video data and do not support real-time emotion analysis or anomaly detection, making it difficult to take immediate and appropriate action. The present invention aims to solve these problems and provide a system that can more effectively ensure public safety.

[0825] 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.

[0826] In this invention, the server includes means for acquiring video data obtained from surveillance equipment, means for recognizing a person's face based on the acquired video data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying a notification device based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, and means for issuing warnings in real time via an application installed on a portable device. This enables a rapid response by analyzing a person's emotions in real time and detecting abnormalities.

[0827] "Surveillance equipment" refers to devices used to acquire video data, specifically surveillance cameras and other imaging equipment.

[0828] "Video data" refers to image and video information acquired by surveillance equipment.

[0829] "Means of recognizing a person's face" refers to algorithms or devices that have the function of detecting and identifying a specific human face from video data.

[0830] "Means of classifying emotions" refers to devices or software that analyze a person's emotions based on their recognized face and classify them into specific emotional categories (e.g., joy, anger, surprise, etc.).

[0831] "Emotional data" refers to data about a person's emotional state, obtained through methods of classifying emotions.

[0832] "Means for detecting abnormal emotional states" refers to devices or software that analyze emotional data and have the function of detecting abnormal emotional fluctuations that exceed a set threshold.

[0833] A "notification device" is a device or system that transmits information to relevant organizations or personnel when an abnormal emotional state is detected.

[0834] An "alert issuing mechanism" refers to a device or system that has the function of issuing a warning to people in the vicinity via sound or visual means when an anomaly is detected.

[0835] "Portable devices" refer to electronic devices that individuals can carry with them at all times, including smartphones and smart glasses.

[0836] An "application" is software that runs on a mobile device and provides a specific function.

[0837] "Real-time" refers to a state where the time between data acquisition and the availability of analysis results is extremely short, with virtually no delay.

[0838] The system for carrying out this invention is constructed using the following main hardware and software.

[0839] hardware

[0840] Surveillance equipment: Includes surveillance cameras and other recording devices for acquiring video data.

[0841] Terminal devices: Security staff will use smart glasses or smartphones.

[0842] Server: A computer device used for storing and analyzing video data.

[0843] software

[0844] Face recognition algorithm: An algorithm used to identify a person's face from video data.

[0845] Emotion analysis model: A model that analyzes and classifies a person's emotions based on machine learning.

[0846] Reporting system: A system that automates reporting based on detected abnormal emotional states.

[0847] Alert system: An audio or visual alert system used to warn people in the vicinity.

[0848] Mobile applications: Applications installed on the mobile devices of security staff.

[0849] Data processing and data calculation

[0850] The server streams video data acquired from surveillance equipment in real time and temporarily stores it in storage. Next, it uses a facial recognition algorithm to detect faces of people in the video. The detected facial image data is input into an emotion analysis model, where emotions are classified. The classified emotion data is analyzed, and if an abnormal emotional state exceeding a set threshold is detected, the notification system is activated, and an automatic notification and alert are issued.

[0851] Specific example

[0852] For example, imagine a security staff member at a shopping mall wearing these smart glasses while patrolling. Surveillance cameras and the smart glasses' cameras capture video data and send it to a server. The server uses a facial recognition algorithm to detect people's faces and an emotion analysis model to classify those people's emotions. If the emotion of "anger" changes rapidly within a short period of time and exceeds a threshold, the notification system is automatically activated, an alert is sent to the security center, and a warning is displayed on the smart glasses. This allows security staff to respond quickly to the scene.

[0853] Example of a prompt

[0854] "We are developing a real-time emotional anomaly detection app for smart glasses worn by security staff in shopping malls. This app analyzes emotions based on video data captured by the glasses' camera and automatically issues warnings and notifications when abnormal emotional fluctuations are detected. Please generate a specific code example."

[0855] Thus, the system for implementing this invention ensures safety in public places and large facilities, and enables the detection and response to emotional abnormalities in real time.

[0856] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0857] Step 1:

[0858] The server acquires video data in real time from surveillance equipment (such as surveillance cameras and smart glasses cameras). The input is video data, and the output is video data temporarily stored in storage. Specifically, the surveillance equipment sends frame-by-frame image data captured by the equipment to the server via the network, and the server receives and stores this data.

[0859] Step 2:

[0860] The server applies a face recognition algorithm to the acquired video data to detect human faces. The input is the video data saved in step 1, and the output is the detected face image data. Specifically, the server uses a face recognition library such as OpenCV to detect human faces in the video and extracts face location information and feature data.

[0861] Step 3:

[0862] The server inputs the detected facial image data into an emotion analysis model to classify emotions. The input is the facial image data detected in step 2, and the output is the classified emotion data. Specifically, the server uses a generative AI model to assign emotion categories such as "joy," "anger," and "surprise" to the facial image data.

[0863] Step 4:

[0864] The server analyzes the classified emotion data and detects abnormal emotional states. The input is the emotion data obtained in step 3, and the output is the detection result of abnormal emotional states. Specifically, the server compares the data with a set threshold to detect sudden changes in emotion or specific abnormal emotional states.

[0865] Step 5:

[0866] When the server detects an abnormal emotional state, it notifies the relevant agency (e.g., security center) through the notification system. The input is the abnormal emotional state detected in step 4, and the output is the notification message. Specifically, the server uses a communication protocol (e.g., an HTTP request) to send details of the detected abnormal emotional state to the notification recipient.

[0867] Step 6:

[0868] The server simultaneously issues an alert to people in the surrounding area. The input is the abnormal emotional state detected in step 4, and the output is an audible or visual warning message. Specifically, the server sends a signal to nearby alert systems (e.g., speakers, display panels) to emit an audible or visual alert.

[0869] Step 7:

[0870] The terminal sends real-time alerts to security staff through an application installed on a mobile device (e.g., smart glasses, smartphone). The input is data of the abnormal emotional state detected in step 4, and the output is a warning message displayed on the mobile device. Specifically, the application receives the data and performs actions such as sending a pop-up notification or audio notification to the user.

[0871] 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.

[0872] The system of this invention analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[0873] 1. Collection of image data

[0874] server

[0875] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0876] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[0877] 2. Face recognition

[0878] server

[0879] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0880] 3. Emotion classification

[0881] server

[0882] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[0883] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[0884] 4. Anomaly detection

[0885] server

[0886] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0887] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[0888] 5. Reporting and issuing alerts

[0889] server

[0890] If an abnormal emotional state is detected, the system automatically notifies the police. The report includes location information, a person's characteristics, and the detected emotional state.

[0891] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[0892] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[0893] Specific example

[0894] As a concrete example, consider a scenario where a specific individual suddenly displays intense anger in a public space. The server first detects the person's face and applies a facial recognition algorithm. Next, an emotion engine analyzes the facial image and classifies the emotion as "anger." If the extreme anger exceeds a threshold in a short period of time, the server determines it to be high risk and automatically notifies the police. In addition, an alert system installed in the vicinity is activated, issuing audio and visual warnings to people nearby. This allows for prompt action and prevents a dangerous situation from occurring.

[0895] As described above, the system of the present invention enhances safety in public places by working in conjunction with surveillance cameras and analyzing emotional fluctuations using an emotion engine. Since machine learning is used for analyzing emotional data, it offers high accuracy, and further improvements are expected through continuous learning. This makes it possible to detect abnormal emotional fluctuations in real time and respond quickly.

[0896] The following describes the processing flow.

[0897] Step 1:

[0898] server

[0899] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[0900] Step 2:

[0901] server

[0902] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[0903] Step 3:

[0904] server

[0905] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[0906] Step 4:

[0907] server

[0908] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[0909] Step 5:

[0910] server

[0911] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[0912] Step 6:

[0913] server

[0914] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[0915] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[0916] (Example 2)

[0917] 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."

[0918] There is a need to improve safety in public places and prevent dangerous situations caused by sudden emotional fluctuations. However, conventional surveillance systems have difficulty analyzing a person's emotional state in real time and issuing appropriate warnings. As a result, their effectiveness in preventing unexpected incidents and crimes is limited. To solve this problem, advanced analysis technology based on image data and a mechanism that can rapidly detect abnormal emotional fluctuations are necessary.

[0919] 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.

[0920] In this invention, the server includes means for acquiring media data obtained from a monitoring device, means for detecting a person's face captured based on the acquired media data, means for analyzing the detected face and classifying the person's emotional state, means for analyzing the classified emotional data over time and detecting abnormal emotional fluctuations, means for notifying a safety agency based on the detected abnormal emotional fluctuations, and means for issuing a warning to the surrounding public. This makes it possible to quickly detect dangerous situations in public places and respond automatically.

[0921] A "monitoring device" is a device used to continuously observe and record events within a specific area.

[0922] "Media data" refers to digital data that includes visual information such as images and videos.

[0923] "Captured" means extracting and saving a specific object from an image or video.

[0924] "Detecting" means using an algorithm to identify specific information and extract it.

[0925] "Analyzing" means breaking down data and examining its contents in detail.

[0926] "Emotional state" refers to a person's psychological and emotional state, and includes "joy," "anger," "surprise," and "sadness."

[0927] "Classifying" means dividing detected data into specific categories.

[0928] "Over time" means observing fluctuations or changes in data over a certain period of time.

[0929] "Abnormal emotional fluctuations" refer to sudden and significant emotional changes that are different from the norm.

[0930] "Means of detection" refers to methods and techniques for finding specific information.

[0931] A "security agency" refers to an organization or institution, such as the police or security companies, that is responsible for ensuring public safety.

[0932] "To notify" means to transmit information.

[0933] "The public" refers to an unspecified large number of people.

[0934] "To issue a warning" means to send a message to inform someone of danger or a problem.

[0935] The system of the present invention analyzes a person's emotional state in real time based on media data acquired from a monitoring device and detects abnormal emotional fluctuations. This system is implemented through the following series of operations.

[0936] First, the server retrieves media data from the monitoring device via the network. The server uses HTTP requests to obtain images and videos from the monitoring device's API and temporarily stores this data in local storage. At this time, a timestamp is added to the media data.

[0937] Next, the server analyzes the acquired media data and detects human faces. Image processing algorithms such as the OpenCV library are used for face detection. OpenCV converts the image to grayscale and uses the Haar Cascade classifier to determine the position and size of human faces. The detected face portions are individually extracted and proceed to the next analysis step.

[0938] The server then inputs the detected facial images into the emotion engine. The emotion engine uses a learning model such as TensorFlow to analyze the emotional state. The emotional state is classified into categories such as "joy," "anger," "surprise," and "sadness." The classified emotion data is temporarily stored and monitored for changes over time.

[0939] The server analyzes changes in emotional data over time and detects abnormal emotional fluctuations based on a specific threshold. This threshold determines a high-risk situation when the set emotional score exceeds a certain standard.

[0940] If abnormal emotional fluctuations are detected, the server automatically notifies security authorities. The notification includes location information, characteristics of the detected person, and emotional state. The server also sends signals to surrounding alert systems to issue audio and visual warnings to the surrounding public.

[0941] For example, if a person standing in front of City Hall suddenly displays strong anger, the server detects their face and uses an emotion engine to identify the emotion of "anger." If extreme anger is detected in a short period of time and its fluctuation exceeds a threshold, the system automatically notifies the police and warns people in the vicinity, enabling a swift response.

[0942] This system enables the early detection of dangerous situations in public places and allows for automatic and rapid response.

[0943] Example of a prompt

[0944] The system classifies emotions from facial images of men captured by surveillance cameras in front of the city hall, and if high-risk emotions are detected, it notifies the police.

[0945] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0946] Step 1:

[0947] The server acquires media data from monitoring devices via the network.

[0948] Input: Real-time image data or video data transmitted from a monitoring device.

[0949] Processing: The server periodically sends HTTP requests to retrieve image data from the monitoring device's API. The retrieved data is then time-stamped and saved.

[0950] Output: Image data file with timestamp.

[0951] Step 2:

[0952] The server detects human faces from stored media data.

[0953] Input: Saved image data file.

[0954] Processing: The server uses the OpenCV library to convert image data to grayscale and performs face detection using the Haar Cascade classifier. It identifies the position and size of the detected faces and extracts the face portion.

[0955] Output: Detected face image file.

[0956] Step 3:

[0957] The server analyzes the detected facial images and classifies their emotional states.

[0958] Input: Extracted face image file.

[0959] Processing: The server loads a TensorFlow trained model and analyzes the face image. The face image is resized to an appropriate size and input into the trained model. The emotion engine classifies the image into categories such as "joy," "anger," "surprise," and "sadness."

[0960] Output: Data indicating emotional state (e.g., result "anger").

[0961] Step 4:

[0962] The server analyzes emotional data and detects abnormal emotional fluctuations over time.

[0963] Input: Classified sentiment data.

[0964] Processing: The server monitors fluctuations in sentiment data and sets specific thresholds. If the set sentiment score exceeds the threshold or if there are sudden fluctuations, it is judged as high risk.

[0965] Output: Detection result of abnormal emotional fluctuations (e.g., "High Risk").

[0966] Step 5:

[0967] If the server detects abnormal emotional fluctuations, it will notify security authorities and issue a warning to the surrounding public.

[0968] Input: Abnormal emotional fluctuation detection results, location information, characteristics of the detected person.

[0969] Processing: The server sends the notification to the security agency. Furthermore, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[0970] Output: Notification to safety agencies, warning signal to alert systems.

[0971] (Application Example 2)

[0972] 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."

[0973] Conventional surveillance systems use image data acquired from surveillance cameras to recognize faces and analyze emotions, but real-time detection and response to abnormal emotional states are difficult. Furthermore, information about detected anomalies is not transmitted quickly, leading to delays in real-time response. In particular, real-time detection and notification mechanisms on portable devices such as smart glasses are not yet established, necessitating rapid risk detection and response.

[0974] In Application Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for acquiring image data obtained from a surveillance camera, means for recognizing a person's face based on the acquired image data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying the police based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, means for recognizing the faces of people in the vicinity in real time using a camera mounted on smart glasses and analyzing their emotions, means for automatically notifying the security center when an abnormal emotional state exceeding a specific threshold is detected in the emotion analysis, and means for recording the detected emotion data and using it for analysis and countermeasures at a later date. This enables real-time detection of emotional states and rapid notification in the event of an anomaly.

[0975] A "surveillance camera" is a device that photographs a specific area and acquires the footage in real time or at regular intervals.

[0976] "Image data" refers to digital data of still images or videos acquired by cameras or other recording devices.

[0977] "Human face" refers to the portion of a person's face within image data and is the object of identification and recognition.

[0978] "Means of face recognition" refers to algorithms or devices that detect a person's face from image data and identify its location and characteristics.

[0979] "Methods for classifying emotions" refer to technologies that analyze a person's emotional state from their recognized face and classify it into categories such as "joy," "anger," "surprise," and "sadness."

[0980] "Emotional data" refers to digital information that indicates a person's emotional state, generated through methods of classifying emotions.

[0981] An "abnormal emotional state" refers to a condition where fluctuations in emotional data increase or decrease rapidly beyond the normal range, leading to an increased risk.

[0982] "Means of reporting to the police" refers to a system or device that automatically makes an emergency contact with the police or other public authorities when an abnormal emotional state is detected.

[0983] "Means of alerting people in the vicinity" refers to technology that notifies people in the vicinity of an abnormal emotional state through sound or visual means.

[0984] "Smart glasses" are wearable devices equipped with cameras, displays, and communication functions that can provide visual assistance by displaying various information.

[0985] "Real-time recognition" refers to a technology that uses a camera mounted on smart glasses to detect a person's face on the spot and analyze it immediately.

[0986] "Means of automatically notifying the security center" refers to a system or device that automatically notifies the security center according to a specific protocol when an abnormal emotional state is detected.

[0987] "Means of recording data" refers to technologies or devices that store emotional data or anomaly detection information so that it can be used for analysis and countermeasures at a later date.

[0988] The present invention provides a system for rapidly responding to abnormal emotional fluctuations by analyzing a person's emotional state in real time based on image data acquired from a surveillance camera. This system consists of a surveillance camera, smart glasses, a server, an emotion analysis engine, a notification system, and an alert system.

[0989] Image data collection

[0990] The server connects to the surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The acquired image data is temporarily stored in storage.

[0991] Face recognition

[0992] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images.

[0993] emotion classification

[0994] The server extracts the recognized portion of the face and inputs it into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[0995] Anomaly detection

[0996] The server analyzes emotional fluctuations over time based on the emotional data generated by the emotion engine and sets a specific threshold. If there is a sudden emotional fluctuation that exceeds that threshold, it is judged to be an anomaly.

[0997] Reporting and alerting

[0998] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, a person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings.

[0999] Applications of smart glasses

[1000] The camera integrated into the smart glasses has the ability to recognize the faces of people in the surroundings in real time and analyze their emotions. The detected emotion data is sent to a server, and if an abnormal emotional state exceeding a certain threshold is detected based on the analysis results, an alert is automatically sent to the security center. Furthermore, the detected emotion data is recorded along with the date and time and used for later analysis and countermeasures.

[1001] Specific example

[1002] As a concrete example, imagine a scenario at a city event where security guards are wearing smart glasses. As the security guard monitors the crowd, a man suddenly displays strong anger. At this moment, the smart glasses' system detects the emotion in real time and identifies it as "anger." This data is transmitted to the security center, which automatically sends an alert. As a result, the security guard can quickly issue a warning and take necessary measures.

[1003] Example of a prompt

[1004] "A 31-year-old man is standing in front of the city hall, showing strong signs of anger. Please take immediate action."

[1005] "A woman in her late 30s or early 40s showed strong signs of sadness at the event venue. We request assistance from security personnel."

[1006] As described above, the collaboration between smart glasses and a server enables real-time detection of emotional states and rapid response. This effectively enhances safety in public places.

[1007] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1008] Step 1:

[1009] Image data collection

[1010] The server receives image data from surveillance cameras over the network. Input is streaming video transmitted from the surveillance cameras or image data that is periodically polled. This data is temporarily stored in storage. Output is image data for facial recognition.

[1011] Step 2:

[1012] Face recognition

[1013] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images. The input is image data acquired from a surveillance camera, and the output is image data of the detected face portion.

[1014] Step 3:

[1015] emotion classification

[1016] The server inputs the image data of the recognized face into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state. The input is the facial image data obtained in the face recognition step, and the output is the classified emotion data.

[1017] Step 4:

[1018] Anomaly detection

[1019] The server analyzes emotional fluctuations over time based on emotional data generated by the emotion engine. A specific threshold is set, and if a sudden emotional shift exceeding that threshold occurs, it is judged as abnormal. The input is the emotional data obtained in the emotion classification step, and the output is warning data for the detected abnormal emotional state.

[1020] Step 5:

[1021] Reporting and alerting

[1022] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, the person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings. The input is the warning data obtained in the anomaly detection step, and the output is the report and alert system activation data.

[1023] Step 6:

[1024] Data transmission for smart glasses

[1025] The camera integrated into the smart glasses recognizes the faces of people in the surroundings in real time and analyzes their emotions. The detected emotion data is sent to a server. The input is image data acquired from the smart glasses' camera, and the output is emotion data sent to the server.

[1026] Step 7:

[1027] Recording and analysis of emotional data

[1028] The server records detected emotion data along with the date and time, and uses it for later analysis and countermeasures. The input is emotion data transmitted from smart glasses and surveillance cameras, and the output is a saved file of the recorded emotion data.

[1029] Through the above processing steps, the smart glasses and server work together to enable real-time detection of emotional states and rapid response.

[1030] 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.

[1031] 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.

[1032] 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.

[1033] [Fourth Embodiment]

[1034] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1035] 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.

[1036] 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).

[1037] 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.

[1038] 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.

[1039] 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).

[1040] 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.

[1041] 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.

[1042] 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.

[1043] 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.

[1044] 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.

[1045] 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.

[1046] 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".

[1047] The present invention's system analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[1048] 1. Collection of image data

[1049] server

[1050] The system connects to surveillance cameras via a network and acquires image data in real time. This data is temporarily stored in storage.

[1051] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[1052] 2. Sentiment analysis

[1053] server

[1054] A face recognition algorithm is applied to the stored image data to detect the faces of people in the images.

[1055] The detected facial images are input into a machine learning model to classify emotions. Emotional classifications include categories such as "joy," "anger," and "surprise."

[1056] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[1057] 3. Anomaly detection

[1058] server

[1059] Based on emotional data, it analyzes fluctuations in those emotions and detects abnormal emotional states that exceed a set threshold.

[1060] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[1061] 4. Reporting and issuing alerts

[1062] server

[1063] If an abnormal emotional state is detected, an automatic report will be sent to the police. The report will include location information, a person's characteristics, and their emotional state.

[1064] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[1065] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[1066] Thus, the system of the present invention operates in conjunction with surveillance cameras and improves safety in public places by analyzing emotional fluctuations in real time. This makes it possible to quickly capture sudden changes in emotions and respond promptly accordingly. Furthermore, since machine learning models are used to analyze emotional data, accuracy improves through continuous data learning, enabling more effective risk management.

[1067] As a concrete example, consider a scenario where a specific individual suddenly displays strong anger in a public space. The server first detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be a potential danger. In this case, the server automatically notifies the police and also sends a signal to surrounding alert systems. This allows for immediate action to be taken, preventing a dangerous situation from occurring.

[1068] The above describes the basic form and operation for carrying out the present invention. This system is expected to have a wide range of applications as an effective tool for enhancing public safety.

[1069] The following describes the processing flow.

[1070] Step 1:

[1071] server

[1072] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[1073] Step 2:

[1074] server

[1075] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[1076] Step 3:

[1077] server

[1078] The detected facial portion is extracted and input into a machine learning model (for example, a deep learning-based emotion classification model). This model analyzes the facial features and classifies the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[1079] Step 4:

[1080] server

[1081] Based on classified emotion data, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[1082] Step 5:

[1083] server

[1084] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[1085] Step 6:

[1086] server

[1087] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[1088] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[1089] (Example 1)

[1090] 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".

[1091] Traditional surveillance systems often relied on manual processes for identifying individuals and detecting abnormal behavior, resulting in significant labor and time constraints. Furthermore, real-time analysis of abnormal emotional fluctuations and rapid response were challenging. Therefore, there was a need for effective methods to improve safety in public spaces.

[1092] 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.

[1093] In this invention, the server includes means for acquiring image data from a surveillance camera, means for detecting a person's face based on the stored image data, and means for inputting the detected face into a machine learning model to classify emotions. This makes it possible to detect abnormal emotional states in real time and automatically notify the police. Furthermore, by quickly issuing alerts to people in the vicinity, dangerous situations can be prevented.

[1094] A "surveillance camera" is a device that continuously photographs a designated area or object and acquires video data.

[1095] "Image data" refers to visual information acquired by cameras and other recording devices, stored in a digital format.

[1096] "Storage" refers to a memory device used to temporarily or permanently store digital data.

[1097] A "face recognition algorithm" is a computational method for detecting the face portion of a person from image data and identifying specific features.

[1098] A "machine learning model" is an algorithm or a set of algorithms that learns from data and makes predictions about new data.

[1099] "Emotion classification" is the process of categorizing the emotions a person is expressing based on characteristic data obtained from their face.

[1100] An "abnormal emotional state" refers to an emotional intensity or fluctuation that exceeds the normal range and may indicate a potential danger.

[1101] "Reporting" refers to the act of automatically issuing a warning to relevant authorities, such as the police, when a monitoring system detects an anomaly.

[1102] An "alert" is a warning signal that is issued to alert people in the vicinity when an abnormal situation is detected.

[1103] A "server" is a computer system that manages and controls data processing, storage, and communication on a network.

[1104] This invention relates to a system that analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of a surveillance camera, a server, and related software.

[1105] 1. System Configuration

[1106] The system includes surveillance cameras, servers, storage, and network infrastructure as its main components. The servers play a central role in acquiring image data from the surveillance cameras and storing that data in storage. Specific hardware and software used include surveillance cameras (e.g., network cameras), network infrastructure (e.g., routers), servers (e.g., general-purpose servers), and storage (e.g., cloud storage).

[1107] 2. Data Collection and Storage

[1108] The server connects to the surveillance camera via a network and acquires image data in real time. The acquired image data is stored in cloud storage. The server also adds a timestamp to the image data before saving it.

[1109] 3. Face detection and sentiment analysis

[1110] The server uses a face recognition algorithm (e.g., OpenCV) to detect human faces based on the stored image data. The detected face images are then input into a machine learning model (e.g., TensorFlow), where emotions are categorized into categories such as "joy," "anger," and "surprise."

[1111] 4. Detection of abnormal emotions

[1112] The server analyzes emotional fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold. If extreme emotional changes are observed in a short period of time, it is assessed as high risk.

[1113] 5. Notification and Alert System

[1114] The server automatically notifies the police if it detects an abnormal emotional state. The report includes location information, the person's characteristics, and their emotional state. Simultaneously, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[1115] Specific example

[1116] For example, if a particular individual suddenly displays strong anger in a public space, the server detects the person's face and performs an emotion analysis. If the emotion of "anger" fluctuates rapidly in a short period of time, it is judged to be high risk. In that case, the server automatically notifies the police and sends a warning signal to the surrounding alert system. This allows for immediate action and prevents dangerous situations from occurring.

[1117] Example of a prompt

[1118] "Collect image data from surveillance cameras in real time and save it on the server."

[1119] "Use image data stored on the server, apply a facial recognition algorithm, and input the results into a machine learning model to classify emotions."

[1120] "Based on the results of the sentiment analysis, analyze the emotional fluctuations and detect any anomalies that exceed the set threshold."

[1121] "If an abnormal emotional state is detected, automatically notify the police and send a signal to the surrounding alert system."

[1122] The above describes specific embodiments for carrying out the present invention. The present invention provides an excellent solution that enables a rapid response at the scene in order to enhance public safety.

[1123] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1124] Step 1: Collecting Image Data

[1125] The server connects to the surveillance cameras via the network and acquires image data in real time.

[1126] Input: Video data from surveillance cameras

[1127] Process: Periodically retrieve image data from surveillance cameras using HTTP requests.

[1128] Output: Acquired image data

[1129] Specific operation: The server sends a request to the surveillance camera every second and receives the latest image data. For example, "New image data has been obtained from the camera in front of the city hall."

[1130] Step 2: Saving Image Data

[1131] The server temporarily stores the acquired image data in cloud storage.

[1132] Input: Acquired image data

[1133] Processing: Save image data to cloud storage and add a timestamp to the file name.

[1134] Output: Saved image data file

[1135] Specific operation: The server saves the image data to a specific folder in storage and adds a timestamp, such as "2023-10-01_10-30-00.jpg".

[1136] Step 3: Face Detection

[1137] The server applies a face recognition algorithm (e.g., OpenCV) to the stored image data to detect human faces.

[1138] Input: Saved image data file

[1139] Processing: A face recognition algorithm detects faces in the image and draws a rectangle around them.

[1140] Output: Coordinate information of detected faces

[1141] Specific operation: The server uses OpenCV to recognize faces in the image and assumes that "a face has been detected from the upper left to the lower right of the image."

[1142] Step 4: Sentiment Analysis

[1143] The server inputs the detected facial images into a machine learning model (e.g., TensorFlow) to analyze their emotions.

[1144] Input: Face coordinate information and corresponding face image

[1145] Processing: The machine learning model is fed with facial images and classified into emotion categories (such as "joy," "anger," and "surprise").

[1146] Output: Detected sentiment data

[1147] Specific operation: The server extracts facial images and inputs them into a machine learning model to classify them as "a man's face showing the emotion of 'anger'."

[1148] Step 5: Anomaly Detection

[1149] The server analyzes fluctuations based on emotional data and detects abnormal emotional states that exceed a set threshold.

[1150] Input: Sentiment data

[1151] Processing: Analyze emotional fluctuations and evaluate rapid changes over short periods.

[1152] Output: Detection results for abnormal emotional states

[1153] Specific operation: The server analyzes emotional data from the past few minutes and detects high risk if there was a rapid shift from "joy" to "anger" in the past minute.

[1154] Step 6: Reporting and issuing alerts

[1155] When the server detects an abnormal emotional state, it automatically notifies the police and sends a signal to the surrounding alert system.

[1156] Input: Detection result of abnormal emotional state

[1157] Processing: Generates a report and sends it to the police, while simultaneously sending a signal to the surrounding alert system.

[1158] Output: Confirmation of notification transmission results and alert issuance.

[1159] Specific actions: The server uses Twilio to notify the police, reporting that "a man in front of City Hall is showing strong signs of anger." It also sends a signal to the Bosch emergency alert system, issuing an audio warning to those nearby to "be careful."

[1160] (Application Example 1)

[1161] 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".

[1162] In recent years, maintaining safety in public spaces and large facilities has become increasingly important. In particular, there is a need to prevent sudden incidents and troubles caused by rapid changes in individuals' emotions. However, conventional surveillance systems only acquire video data and do not support real-time emotion analysis or anomaly detection, making it difficult to take immediate and appropriate action. The present invention aims to solve these problems and provide a system that can more effectively ensure public safety.

[1163] 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.

[1164] In this invention, the server includes means for acquiring video data obtained from surveillance equipment, means for recognizing a person's face based on the acquired video data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying a notification device based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, and means for issuing warnings in real time via an application installed on a portable device. This enables a rapid response by analyzing a person's emotions in real time and detecting abnormalities.

[1165] "Surveillance equipment" refers to devices used to acquire video data, specifically surveillance cameras and other imaging equipment.

[1166] "Video data" refers to image and video information acquired by surveillance equipment.

[1167] "Means of recognizing a person's face" refers to algorithms or devices that have the function of detecting and identifying a specific human face from video data.

[1168] "Means of classifying emotions" refers to devices or software that analyze a person's emotions based on their recognized face and classify them into specific emotional categories (e.g., joy, anger, surprise, etc.).

[1169] "Emotional data" refers to data about a person's emotional state, obtained through methods of classifying emotions.

[1170] "Means for detecting abnormal emotional states" refers to devices or software that analyze emotional data and have the function of detecting abnormal emotional fluctuations that exceed a set threshold.

[1171] A "notification device" is a device or system that transmits information to relevant organizations or personnel when an abnormal emotional state is detected.

[1172] An "alert issuing mechanism" refers to a device or system that has the function of issuing a warning to people in the vicinity via sound or visual means when an anomaly is detected.

[1173] "Portable devices" refer to electronic devices that individuals can carry with them at all times, including smartphones and smart glasses.

[1174] An "application" is software that runs on a mobile device and provides a specific function.

[1175] "Real-time" refers to a state where the time between data acquisition and the availability of analysis results is extremely short, with virtually no delay.

[1176] The system for carrying out this invention is constructed using the following main hardware and software.

[1177] hardware

[1178] Surveillance equipment: Includes surveillance cameras and other recording devices for acquiring video data.

[1179] Terminal devices: Security staff will use smart glasses or smartphones.

[1180] Server: A computer device used for storing and analyzing video data.

[1181] software

[1182] Face recognition algorithm: An algorithm used to identify a person's face from video data.

[1183] Emotion analysis model: A model that analyzes and classifies a person's emotions based on machine learning.

[1184] Reporting system: A system that automates reporting based on detected abnormal emotional states.

[1185] Alert system: An audio or visual alert system used to warn people in the vicinity.

[1186] Mobile applications: Applications installed on the mobile devices of security staff.

[1187] Data processing and data calculation

[1188] The server streams video data acquired from surveillance equipment in real time and temporarily stores it in storage. Next, it uses a facial recognition algorithm to detect faces of people in the video. The detected facial image data is input into an emotion analysis model, where emotions are classified. The classified emotion data is analyzed, and if an abnormal emotional state exceeding a set threshold is detected, the notification system is activated, and an automatic notification and alert are issued.

[1189] Specific example

[1190] For example, imagine a security staff member at a shopping mall wearing these smart glasses while patrolling. Surveillance cameras and the smart glasses' cameras capture video data and send it to a server. The server uses a facial recognition algorithm to detect people's faces and an emotion analysis model to classify those people's emotions. If the emotion of "anger" changes rapidly within a short period of time and exceeds a threshold, the notification system is automatically activated, an alert is sent to the security center, and a warning is displayed on the smart glasses. This allows security staff to respond quickly to the scene.

[1191] Example of a prompt

[1192] "We are developing a real-time emotional anomaly detection app for smart glasses worn by security staff in shopping malls. This app analyzes emotions based on video data captured by the glasses' camera and automatically issues warnings and notifications when abnormal emotional fluctuations are detected. Please generate a specific code example."

[1193] Thus, the system for implementing this invention ensures safety in public places and large facilities, and enables the detection and response to emotional abnormalities in real time.

[1194] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1195] Step 1:

[1196] The server acquires video data in real time from surveillance equipment (such as surveillance cameras and smart glasses cameras). The input is video data, and the output is video data temporarily stored in storage. Specifically, the surveillance equipment sends frame-by-frame image data captured by the equipment to the server via the network, and the server receives and stores this data.

[1197] Step 2:

[1198] The server applies a face recognition algorithm to the acquired video data to detect human faces. The input is the video data saved in step 1, and the output is the detected face image data. Specifically, the server uses a face recognition library such as OpenCV to detect human faces in the video and extracts face location information and feature data.

[1199] Step 3:

[1200] The server inputs the detected facial image data into an emotion analysis model to classify emotions. The input is the facial image data detected in step 2, and the output is the classified emotion data. Specifically, the server uses a generative AI model to assign emotion categories such as "joy," "anger," and "surprise" to the facial image data.

[1201] Step 4:

[1202] The server analyzes the classified emotion data and detects abnormal emotional states. The input is the emotion data obtained in step 3, and the output is the detection result of abnormal emotional states. Specifically, the server compares the data with a set threshold to detect sudden changes in emotion or specific abnormal emotional states.

[1203] Step 5:

[1204] When the server detects an abnormal emotional state, it notifies the relevant agency (e.g., security center) through the notification system. The input is the abnormal emotional state detected in step 4, and the output is the notification message. Specifically, the server uses a communication protocol (e.g., an HTTP request) to send details of the detected abnormal emotional state to the notification recipient.

[1205] Step 6:

[1206] The server simultaneously issues an alert to people in the surrounding area. The input is the abnormal emotional state detected in step 4, and the output is an audible or visual warning message. Specifically, the server sends a signal to nearby alert systems (e.g., speakers, display panels) to emit an audible or visual alert.

[1207] Step 7:

[1208] The terminal sends real-time alerts to security staff through an application installed on a mobile device (e.g., smart glasses, smartphone). The input is data of the abnormal emotional state detected in step 4, and the output is a warning message displayed on the mobile device. Specifically, the application receives the data and performs actions such as sending a pop-up notification or audio notification to the user.

[1209] 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.

[1210] The system of this invention analyzes a person's emotional state in real time based on image data acquired from a surveillance camera and detects abnormal emotional fluctuations. This system consists of the following main functions and operations.

[1211] 1. Collection of image data

[1212] server

[1213] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[1214] For example, suppose a camera in front of the city hall captured the face of a 31-year-old man.

[1215] 2. Face recognition

[1216] server

[1217] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[1218] 3. Emotion classification

[1219] server

[1220] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[1221] For example, we might determine that the 31-year-old man mentioned above is showing the emotion of "anger."

[1222] 4. Anomaly detection

[1223] server

[1224] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[1225] For example, if extreme anger is detected in a short period of time, and the fluctuation exceeds a threshold, it is judged to be a high-risk factor.

[1226] 5. Reporting and issuing alerts

[1227] server

[1228] If an abnormal emotional state is detected, the system automatically notifies the police. The report includes location information, a person's characteristics, and the detected emotional state.

[1229] At the same time, it sends signals to surrounding alert systems, issuing warnings via voice and visuals.

[1230] For example, you could report to the police that "a 31-year-old man is showing strong anger in front of the city hall" and warn nearby citizens to be cautious.

[1231] Specific example

[1232] As a concrete example, consider a scenario where a specific individual suddenly displays intense anger in a public space. The server first detects the person's face and applies a facial recognition algorithm. Next, an emotion engine analyzes the facial image and classifies the emotion as "anger." If the extreme anger exceeds a threshold in a short period of time, the server determines it to be high risk and automatically notifies the police. In addition, an alert system installed in the vicinity is activated, issuing audio and visual warnings to people nearby. This allows for prompt action and prevents a dangerous situation from occurring.

[1233] As described above, the system of the present invention enhances safety in public places by working in conjunction with surveillance cameras and analyzing emotional fluctuations using an emotion engine. Since machine learning is used for analyzing emotional data, it offers high accuracy, and further improvements are expected through continuous learning. This makes it possible to detect abnormal emotional fluctuations in real time and respond quickly.

[1234] The following describes the processing flow.

[1235] Step 1:

[1236] server

[1237] The system connects to surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The server temporarily stores the received image data in storage.

[1238] Step 2:

[1239] server

[1240] A face recognition algorithm is applied to the stored image data to detect human faces in the images. Specifically, the position and size of faces are determined using libraries such as OpenCV.

[1241] Step 3:

[1242] server

[1243] The detected facial portion is extracted and input into the emotion engine. The emotion engine uses machine learning algorithms to analyze facial features and classify emotional states into categories such as "joy," "anger," "surprise," and "sadness."

[1244] Step 4:

[1245] server

[1246] Based on the emotional data generated by the emotion engine, the system analyzes emotional fluctuations over time. A specific threshold is set, and the system checks for any sudden emotional shifts that exceed that threshold.

[1247] Step 5:

[1248] server

[1249] If a sudden emotional shift exceeds a set threshold, the system will be deemed high-risk and will automatically notify the police. The report will include location information, the person's characteristics, and the detected emotional state.

[1250] Step 6:

[1251] server

[1252] Simultaneously with notifying the police, a signal is sent to surrounding alert systems, issuing audio and visual warnings. Based on this signal, people in the vicinity are alerted to the unusual situation, ensuring their safety.

[1253] The processing steps described above allow the system to monitor a person's emotions in real time and quickly detect and respond to abnormal emotional fluctuations. This significantly improves safety in public places.

[1254] (Example 2)

[1255] 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".

[1256] There is a need to improve safety in public places and prevent dangerous situations caused by sudden emotional fluctuations. However, conventional surveillance systems have difficulty analyzing a person's emotional state in real time and issuing appropriate warnings. As a result, their effectiveness in preventing unexpected incidents and crimes is limited. To solve this problem, advanced analysis technology based on image data and a mechanism that can rapidly detect abnormal emotional fluctuations are necessary.

[1257] 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.

[1258] In this invention, the server includes means for acquiring media data obtained from a monitoring device, means for detecting a person's face captured based on the acquired media data, means for analyzing the detected face and classifying the person's emotional state, means for analyzing the classified emotional data over time and detecting abnormal emotional fluctuations, means for notifying a safety agency based on the detected abnormal emotional fluctuations, and means for issuing a warning to the surrounding public. This makes it possible to quickly detect dangerous situations in public places and respond automatically.

[1259] A "monitoring device" is a device used to continuously observe and record events within a specific area.

[1260] "Media data" refers to digital data that includes visual information such as images and videos.

[1261] "Captured" means extracting and saving a specific object from an image or video.

[1262] "Detecting" means using an algorithm to identify specific information and extract it.

[1263] "Analyzing" means breaking down data and examining its contents in detail.

[1264] "Emotional state" refers to a person's psychological and emotional state, and includes "joy," "anger," "surprise," and "sadness."

[1265] "Classifying" means dividing detected data into specific categories.

[1266] "Over time" means observing fluctuations or changes in data over a certain period of time.

[1267] "Abnormal emotional fluctuations" refer to sudden and significant emotional changes that are different from the norm.

[1268] "Means of detection" refers to methods and techniques for finding specific information.

[1269] A "security agency" refers to an organization or institution, such as the police or security companies, that is responsible for ensuring public safety.

[1270] "To notify" means to transmit information.

[1271] "The public" refers to an unspecified large number of people.

[1272] "To issue a warning" means to send a message to inform someone of danger or a problem.

[1273] The system of the present invention analyzes a person's emotional state in real time based on media data acquired from a monitoring device and detects abnormal emotional fluctuations. This system is implemented through the following series of operations.

[1274] First, the server retrieves media data from the monitoring device via the network. The server uses HTTP requests to obtain images and videos from the monitoring device's API and temporarily stores this data in local storage. At this time, a timestamp is added to the media data.

[1275] Next, the server analyzes the acquired media data and detects human faces. Image processing algorithms such as the OpenCV library are used for face detection. OpenCV converts the image to grayscale and uses the Haar Cascade classifier to determine the position and size of human faces. The detected face portions are individually extracted and proceed to the next analysis step.

[1276] The server then inputs the detected facial images into the emotion engine. The emotion engine uses a learning model such as TensorFlow to analyze the emotional state. The emotional state is classified into categories such as "joy," "anger," "surprise," and "sadness." The classified emotion data is temporarily stored and monitored for changes over time.

[1277] The server analyzes changes in emotional data over time and detects abnormal emotional fluctuations based on a specific threshold. This threshold determines a high-risk situation when the set emotional score exceeds a certain standard.

[1278] If abnormal emotional fluctuations are detected, the server automatically notifies security authorities. The notification includes location information, characteristics of the detected person, and emotional state. The server also sends signals to surrounding alert systems to issue audio and visual warnings to the surrounding public.

[1279] For example, if a person standing in front of City Hall suddenly displays strong anger, the server detects their face and uses an emotion engine to identify the emotion of "anger." If extreme anger is detected in a short period of time and its fluctuation exceeds a threshold, the system automatically notifies the police and warns people in the vicinity, enabling a swift response.

[1280] This system enables the early detection of dangerous situations in public places and allows for automatic and rapid response.

[1281] Example of a prompt

[1282] The system classifies emotions from facial images of men captured by surveillance cameras in front of the city hall, and if high-risk emotions are detected, it notifies the police.

[1283] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1284] Step 1:

[1285] The server acquires media data from monitoring devices via the network.

[1286] Input: Real-time image data or video data transmitted from a monitoring device.

[1287] Processing: The server periodically sends HTTP requests to retrieve image data from the monitoring device's API. The retrieved data is then time-stamped and saved.

[1288] Output: Image data file with timestamp.

[1289] Step 2:

[1290] The server detects human faces from stored media data.

[1291] Input: Saved image data file.

[1292] Processing: The server uses the OpenCV library to convert image data to grayscale and performs face detection using the Haar Cascade classifier. It identifies the position and size of the detected faces and extracts the face portion.

[1293] Output: Detected face image file.

[1294] Step 3:

[1295] The server analyzes the detected facial images and classifies their emotional states.

[1296] Input: Extracted face image file.

[1297] Processing: The server loads a TensorFlow trained model and analyzes the face image. The face image is resized to an appropriate size and input into the trained model. The emotion engine classifies the image into categories such as "joy," "anger," "surprise," and "sadness."

[1298] Output: Data indicating emotional state (e.g., result "anger").

[1299] Step 4:

[1300] The server analyzes emotional data and detects abnormal emotional fluctuations over time.

[1301] Input: Classified sentiment data.

[1302] Processing: The server monitors fluctuations in sentiment data and sets specific thresholds. If the set sentiment score exceeds the threshold or if there are sudden fluctuations, it is judged as high risk.

[1303] Output: Detection result of abnormal emotional fluctuations (e.g., "High Risk").

[1304] Step 5:

[1305] If the server detects abnormal emotional fluctuations, it will notify security authorities and issue a warning to the surrounding public.

[1306] Input: Abnormal emotional fluctuation detection results, location information, characteristics of the detected person.

[1307] Processing: The server sends the notification to the security agency. Furthermore, it sends signals to surrounding alert systems, issuing audio and visual warnings.

[1308] Output: Notification to safety agencies, warning signal to alert systems.

[1309] (Application Example 2)

[1310] 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".

[1311] Conventional surveillance systems use image data acquired from surveillance cameras to recognize faces and analyze emotions, but real-time detection and response to abnormal emotional states are difficult. Furthermore, information about detected anomalies is not transmitted quickly, leading to delays in real-time response. In particular, real-time detection and notification mechanisms on portable devices such as smart glasses are not yet established, necessitating rapid risk detection and response.

[1312] In Application Example 2, the identification processing by the identification processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for acquiring image data obtained from a surveillance camera, means for recognizing a person's face based on the acquired image data, means for classifying the person's emotions based on the recognized face, means for analyzing the classified emotion data and detecting abnormal emotional states, means for notifying the police based on the detected abnormal emotional state, means for issuing alerts to people in the vicinity, means for recognizing the faces of people in the vicinity in real time using a camera mounted on smart glasses and analyzing their emotions, means for automatically notifying the security center when an abnormal emotional state exceeding a specific threshold is detected in the emotion analysis, and means for recording the detected emotion data and using it for analysis and countermeasures at a later date. This enables real-time detection of emotional states and rapid notification in the event of an anomaly.

[1313] A "surveillance camera" is a device that photographs a specific area and acquires the footage in real time or at regular intervals.

[1314] "Image data" refers to digital data of still images or videos acquired by cameras or other recording devices.

[1315] "Human face" refers to the portion of a person's face within image data and is the object of identification and recognition.

[1316] "Means of face recognition" refers to algorithms or devices that detect a person's face from image data and identify its location and characteristics.

[1317] "Methods for classifying emotions" refer to technologies that analyze a person's emotional state from their recognized face and classify it into categories such as "joy," "anger," "surprise," and "sadness."

[1318] "Emotional data" refers to digital information that indicates a person's emotional state, generated through methods of classifying emotions.

[1319] An "abnormal emotional state" refers to a condition where fluctuations in emotional data increase or decrease rapidly beyond the normal range, leading to an increased risk.

[1320] "Means of reporting to the police" refers to a system or device that automatically makes an emergency contact with the police or other public authorities when an abnormal emotional state is detected.

[1321] "Means of alerting people in the vicinity" refers to technology that notifies people in the vicinity of an abnormal emotional state through sound or visual means.

[1322] "Smart glasses" are wearable devices equipped with cameras, displays, and communication functions that can provide visual assistance by displaying various information.

[1323] "Real-time recognition" refers to a technology that uses a camera mounted on smart glasses to detect a person's face on the spot and analyze it immediately.

[1324] "Means of automatically notifying the security center" refers to a system or device that automatically notifies the security center according to a specific protocol when an abnormal emotional state is detected.

[1325] "Means of recording data" refers to technologies or devices that store emotional data or anomaly detection information so that it can be used for analysis and countermeasures at a later date.

[1326] The present invention provides a system for rapidly responding to abnormal emotional fluctuations by analyzing a person's emotional state in real time based on image data acquired from a surveillance camera. This system consists of a surveillance camera, smart glasses, a server, an emotion analysis engine, a notification system, and an alert system.

[1327] Image data collection

[1328] The server connects to the surveillance cameras via a network and periodically polls for image data or receives real-time streaming. The acquired image data is temporarily stored in storage.

[1329] Face recognition

[1330] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images.

[1331] emotion classification

[1332] The server extracts the recognized portion of the face and inputs it into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state into categories such as "joy," "anger," "surprise," and "sadness."

[1333] Anomaly detection

[1334] The server analyzes emotional fluctuations over time based on the emotional data generated by the emotion engine and sets a specific threshold. If there is a sudden emotional fluctuation that exceeds that threshold, it is judged to be an anomaly.

[1335] Reporting and alerting

[1336] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, a person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings.

[1337] Applications of smart glasses

[1338] The camera integrated into the smart glasses has the ability to recognize the faces of people in the surroundings in real time and analyze their emotions. The detected emotion data is sent to a server, and if an abnormal emotional state exceeding a certain threshold is detected based on the analysis results, an alert is automatically sent to the security center. Furthermore, the detected emotion data is recorded along with the date and time and used for later analysis and countermeasures.

[1339] Specific example

[1340] As a concrete example, imagine a scenario at a city event where security guards are wearing smart glasses. As the security guard monitors the crowd, a man suddenly displays strong anger. At this moment, the smart glasses' system detects the emotion in real time and identifies it as "anger." This data is transmitted to the security center, which automatically sends an alert. As a result, the security guard can quickly issue a warning and take necessary measures.

[1341] Example of a prompt

[1342] "A 31-year-old man is standing in front of the city hall, showing strong signs of anger. Please take immediate action."

[1343] "A woman in her late 30s or early 40s showed strong signs of sadness at the event venue. We request assistance from security personnel."

[1344] As described above, the collaboration between smart glasses and a server enables real-time detection of emotional states and rapid response. This effectively enhances safety in public places.

[1345] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1346] Step 1:

[1347] Image data collection

[1348] The server receives image data from surveillance cameras over the network. Input is streaming video transmitted from the surveillance cameras or image data that is periodically polled. This data is temporarily stored in storage. Output is image data for facial recognition.

[1349] Step 2:

[1350] Face recognition

[1351] The server applies a face recognition algorithm to the stored image data. Specifically, it uses the OpenCV library to determine the position and size of people's faces in the images. The input is image data acquired from a surveillance camera, and the output is image data of the detected face portion.

[1352] Step 3:

[1353] emotion classification

[1354] The server inputs the image data of the recognized face into the emotion engine. The emotion engine uses a machine learning algorithm (using a TensorFlow model) to analyze the facial features and classify the emotional state. The input is the facial image data obtained in the face recognition step, and the output is the classified emotion data.

[1355] Step 4:

[1356] Anomaly detection

[1357] The server analyzes emotional fluctuations over time based on emotional data generated by the emotion engine. A specific threshold is set, and if a sudden emotional shift exceeding that threshold occurs, it is judged as abnormal. The input is the emotional data obtained in the emotion classification step, and the output is warning data for the detected abnormal emotional state.

[1358] Step 5:

[1359] Reporting and alerting

[1360] The server automatically notifies the police if an abnormal emotional state is detected. The report includes location information, the person's characteristics, and the detected emotional state. It also sends signals to surrounding alert systems, issuing audio and visual warnings. The input is the warning data obtained in the anomaly detection step, and the output is the report and alert system activation data.

[1361] Step 6:

[1362] Data transmission for smart glasses

[1363] The camera integrated into the smart glasses recognizes the faces of people in the surroundings in real time and analyzes their emotions. The detected emotion data is sent to a server. The input is image data acquired from the smart glasses' camera, and the output is emotion data sent to the server.

[1364] Step 7:

[1365] Recording and analysis of emotional data

[1366] The server records detected emotion data along with the date and time, and uses it for later analysis and countermeasures. The input is emotion data transmitted from smart glasses and surveillance cameras, and the output is a saved file of the recorded emotion data.

[1367] Through the above processing steps, the smart glasses and server work together to enable real-time detection of emotional states and rapid response.

[1368] 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.

[1369] 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.

[1370] 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.

[1371] 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.

[1372] 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.

[1373] 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.

[1374] 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.

[1375] 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.

[1376] 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."

[1377] 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.

[1378] 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.

[1379] 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.

[1380] 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.

[1381] 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.

[1382] 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.

[1383] 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.

[1384] 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.

[1385] 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.

[1386] 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.

[1387] 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.

[1388] 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.

[1389] The following is further disclosed regarding the embodiments described above.

[1390] (Claim 1)

[1391] A means of obtaining image data acquired from surveillance cameras,

[1392] A means for recognizing a person's face based on acquired image data,

[1393] A means of classifying a person's emotions based on their recognized face,

[1394] A means for analyzing classified emotion data and detecting abnormal emotional states,

[1395] Means of reporting to the police based on detected abnormal emotional states,

[1396] A means of sending an alert to people in the surrounding area,

[1397] A system that includes this.

[1398] (Claim 2)

[1399] The system according to claim 1, which detects a person's face from image data using a face recognition algorithm.

[1400] (Claim 3)

[1401] The system according to claim 1, which uses a machine learning model to classify emotions from recognized faces.

[1402] "Example 1"

[1403] (Claim 1)

[1404] A means of obtaining image data acquired from surveillance cameras,

[1405] A means of saving the acquired image data to storage,

[1406] A means for detecting a person's face based on stored image data,

[1407] A method for inputting detected faces into a machine learning model to classify emotions,

[1408] A means for analyzing classified emotion data and detecting abnormal emotional states that exceed a set threshold,

[1409] A means of automatically notifying the police based on detected abnormal emotional states,

[1410] A means of sending an alert to people in the surrounding area,

[1411] A system that includes this.

[1412] (Claim 2)

[1413] The system according to claim 1, which detects a person's face from image data using a face recognition algorithm.

[1414] (Claim 3)

[1415] The system according to claim 1, which uses a machine learning model to classify emotions from detected faces.

[1416] "Application Example 1"

[1417] (Claim 1)

[1418] A means of acquiring video data obtained from surveillance equipment,

[1419] A means for recognizing a person's face based on acquired video data,

[1420] A means of classifying a person's emotions based on their recognized face,

[1421] A means for analyzing classified emotion data and detecting abnormal emotional states,

[1422] A means of notifying a notification device based on the detected abnormal emotional state,

[1423] A means of sending an alert to people in the surrounding area,

[1424] A means of issuing real-time warnings via an application installed on a mobile device,

[1425] A system that includes this.

[1426] (Claim 2)

[1427] The system according to claim 1, which detects a person's face from video data using a face recognition algorithm.

[1428] (Claim 3)

[1429] The system according to claim 1, which uses a machine learning model to classify emotions from recognized faces.

[1430] "Example 2 of combining an emotion engine"

[1431] (Claim 1)

[1432] A means of acquiring media data obtained from a monitoring device,

[1433] A means for detecting a person's face captured based on acquired media data,

[1434] A means for analyzing the detected face and classifying the emotional state of the person,

[1435] A means for analyzing classified emotion data over time to detect abnormal emotion fluctuations,

[1436] A means of notifying safety agencies based on detected abnormal emotional fluctuations,

[1437] A means of issuing a warning to the surrounding public,

[1438] A system that includes this.

[1439] (Claim 2)

[1440] The system according to claim 1, which uses an image processing algorithm to detect a person's face from media data.

[1441] (Claim 3)

[1442] The system according to claim 1, which uses a learning model to classify emotional states from detected faces.

[1443] "Application example 2 of combining emotional engines"

[1444] (Claim 1)

[1445] A means of obtaining image data acquired from surveillance cameras,

[1446] A means for recognizing a person's face based on acquired image data,

[1447] A means of classifying a person's emotions based on their recognized face,

[1448] A means for analyzing classified emotion data and detecting abnormal emotional states,

[1449] Means of reporting to the police based on detected abnormal emotional states,

[1450] A means of sending an alert to people in the surrounding area,

[1451] A method for recognizing the faces of people in the surroundings in real time using a camera built into smart glasses and analyzing their emotions,

[1452] In emotion analysis, when an abnormal emotional state exceeding a specific threshold is detected, a means of automatically notifying the security center is provided.

[1453] A method for recording detected emotion data and using it for later analysis and countermeasures,

[1454] A system that includes this.

[1455] (Claim 2)

[1456] The system according to claim 1, which detects a person's face from image data using a face recognition algorithm.

[1457] (Claim 3)

[1458] The system according to claim 1, which uses a machine learning model to classify emotions from recognized faces. [Explanation of symbols]

[1459] 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 obtaining image data acquired from surveillance cameras, A means for recognizing a person's face based on acquired image data, A means of classifying a person's emotions based on their recognized face, A means for analyzing classified emotion data and detecting abnormal emotional states, Means of reporting to the police based on detected abnormal emotional states, A means of sending an alert to people in the surrounding area, A system that includes this.

2. The system according to claim 1, which detects a person's face from image data using a face recognition algorithm.

3. The system according to claim 1, which uses a machine learning model to classify emotions from recognized faces.

Images