system

Abstract

Provide a system. A data collection means; data analysis means; An anomaly detection means; an alert notification means; Data storage means A system including:

Description

[Technical Field]

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

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

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

[0004] Abuse in daycare centers and nursing homes is a serious social problem. These facilities house many children and elderly people, and their safety and welfare are of paramount importance. However, due to staff shortages and management difficulties, abuse can sometimes go unnoticed. Current monitoring systems require constant human supervision, which places a strain on human resources. Furthermore, when abuse does occur, immediate response is often not possible, putting victims at risk of serious harm. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides a system including a data collection means, a data analysis means, an anomaly detection means, an alert notification means, and a data storage means. This system uses an AI algorithm to detect anomalies in audio and video data and immediately notify the facility manager. The data collection means continuously collects audio and video data within the facility, and the data analysis means analyzes the collected data in real time. If abnormal behavior or sounds are detected in the data analyzed by the anomaly detection means, the alert notification means sends a notification to the facility manager, urging them to take prompt action. Furthermore, the data storage means stores the collected data for a certain period of time, allowing it to be used for later verification and report creation. This enables early detection of abuse in daycare centers and elderly care facilities and rapid response, thereby protecting victims and ensuring their safety.

[0006] "Data Collection Implement" means a combination of hardware and software for collecting audio and video data within a facility in real time.

[0007] "Data analysis means" means a device that contains AI algorithms and related software for analyzing collected audio and video data and detecting anomalies.

[0008] The "abnormality detection means" is a function that identifies abnormal behavior or abnormal sounds from the analyzed data and determines whether or not there is an abnormality.

[0009] The "alert notification means" is a system for immediately sending a notification to facility managers and related parties when an abnormality is detected.

[0010] "Data storage means" refers to devices and software that store collected audio and video data for a certain period of time and keep it in a state that makes it available for later verification and report creation.

[0011] "Alert information" is detailed information that is generated when an abnormality is detected, and includes the type of abnormality, the time of occurrence, the location of occurrence, and so on.

[0012] "Notification Method" refers to the notification method used by the Alert Notification Means, and may include one or more of email, text message, and / or dedicated in-app notification. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0014] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0017] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0018] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0019] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0026] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0033] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0034] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[0035] Account Creation and Initial Setup

[0036] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and then send it to the server.

[0037] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[0038] Device Registration and Connection

[0039] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[0040] After a device boots up, it automatically sends identifying information to a server, which uses this information to associate the device with a specific user account.

[0041] Data Collection and Monitoring

[0042] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[0043] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[0044] Data analysis and anomaly detection

[0045] The server uses AI algorithms to analyze the received audio and video data, with the goal of detecting any abnormal behavior or sounds that deviate from normal patterns of behavior.

[0046] If the server detects any abnormal behavior or sound, it generates an alert containing information about the abnormality, including the type of abnormality, the time of occurrence, and the location where it occurred.

[0047] Alert Notifications

[0048] The server generates an alert and sends it via email, SMS, or a dedicated app, or a combination of these.

[0049] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, enabling them to take prompt action.

[0050] Data storage and report generation

[0051] The server stores the collected audio and video data for a certain period of time, and the stored data is used for later verification and report creation.

[0052] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[0053] Specific examples

[0054] For example, let's assume that a staff member at a nursing home is verbally abusing an elderly person. In this case, the system works as follows:

[0055] 1. The device captures audio data within the facility and sends it to the server.

[0056] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[0057] 3. The server detects the abnormality and generates alert information.

[0058] 4. The server sends a notification to the facility manager via email or a dedicated app.

[0059] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[0060] In this way, the present invention provides a system that utilizes AI technology to ensure safety within facilities and enables early detection of abuse and rapid response.

[0061] The processing flow will be explained below.

[0062] Step 1:

[0063] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0064] Step 2:

[0065] The server creates an account based on the received information, stores it in the database, and sends authentication information (user ID and password) to the user's email address.

[0066] Step 3:

[0067] The user signs up for a subscription and receives the device, which they then install in an appropriate location and connect to the network and power.

[0068] Step 4:

[0069] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[0070] Step 5:

[0071] The server verifies the identity from the device and associates it with the user's account, thereby confirming that the device belongs to a particular facility.

[0072] Step 6:

[0073] The device collects audio and video data in real time, which is then immediately sent to a server.

[0074] Step 7:

[0075] The server buffers the received data and inputs it into a data analysis means, which includes an AI algorithm.

[0076] Step 8:

[0077] AI algorithms installed on the server analyze audio and video data, including detecting abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[0078] Step 9:

[0079] If an anomaly is detected, the server generates an alert, which includes details such as the type of anomaly, the time it occurred, and the location where it occurred.

[0080] Step 10:

[0081] Based on the generated alert information, the server sends a notification to the user using the specified notification method (email, SMS, or in-app notification).

[0082] Step 11:

[0083] Users can receive alert notifications and access a dedicated app or website to check detailed information, making it easier to understand the situation on-site and enable prompt action.

[0084] Step 12:

[0085] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[0086] Step 13:

[0087] The server periodically generates a report summarizing the analysis results, allowing users to understand the safety status of the entire facility and take any necessary improvement measures.

[0088] In this way, the system monitors the situation within the facility in real time, promptly notifies the user when an abnormality occurs, and ensures the safety of the facility by allowing the user to take appropriate action. This series of processing steps enables early detection of abuse and rapid response.

[0089] Example 1

[0090] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0091] Abuse in daycare centers and elderly care facilities is a major social problem, and there is a need for a rapid and accurate monitoring system to prevent it from happening. These facilities also face serious labor shortages, necessitating a highly automated system. Current monitoring systems have difficulty detecting abnormal behavior and sounds in real time and responding quickly. Therefore, there is a need for a system that utilizes AI technology to efficiently detect abnormalities while ensuring safety within the facility.

[0092] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0093] In this invention, the server includes a means for creating an account, a means for transmitting device identification information to the server and linking it to a user account, a means for collecting audio and video data, a means for analyzing the received data by time frame, a means for detecting abnormal behavior or abnormal sounds using an AI algorithm, a means for generating an alert containing abnormal information, a means for notifying the generated alert via email, text message, or app, a means for storing the audio and video data for a certain period of time, and a means for generating a report of the results of data analysis. This enables early detection of possible acts of abuse occurring within the facility and rapid response.

[0094] The "account creation means" refers to the means by which a user accesses the system's website and creates an account by entering necessary information such as the facility name, location, and contact information.

[0095] The "means of transmitting device identification information to a server and associating it with a user account" refers to a means in which the device transmits identification information to a server after startup, and the server associates the device with a specific user account based on that information.

[0096] "Means for collecting audio and video data" means means for continuously capturing audio and video data by devices installed within the facility.

[0097] The "means for analyzing received data for each time frame" refers to a means for storing data received by the server in a buffer and analyzing the data for each fixed time frame.

[0098] "Means for detecting abnormal behavior and sounds using AI algorithms" refers to a means by which the server uses AI technology to identify abnormal behavior and sounds from audio and video data.

[0099] The "means for generating an alert including abnormality information" refers to a means for generating an alert including information such as the type of abnormality, the time of occurrence, and the location of occurrence when the server detects abnormal behavior or abnormal sound.

[0100] "Means of notifying users of generated alerts via email, text message, or app" refers to means of notifying users of server-generated alerts via email, text message, or a dedicated app.

[0101] "Means for storing audio and video data for a certain period of time" refers to the means by which the server stores collected audio and video data for a certain period of time for the purpose of later verification and report creation.

[0102] The "means for generating a report of the data analysis results" refers to a means by which the server periodically generates a report summarizing the results of the data analysis and anomaly detection, and provides it to the user.

[0103] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[0104] Account Creation and Initial Setup

[0105] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and send it to the server. The server stores the user's account information in a database, allowing it to configure and manage each facility. The database uses a general-purpose database management system such as MySQL (registered trademark) or PostgreSQL.

[0106] Device Registration and Connection

[0107] When a user signs up for a subscription, they are sent a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After the devices start up, they automatically send identification information to a server. The server uses this information to associate the devices with specific user accounts. Devices are identified using MAC addresses or unique IDs.

[0108] Data Collection and Monitoring

[0109] The device continuously collects audio and video data at the installation site. The collected data is sent in real time to a server, which stores the data in a buffer and inputs it into a data analysis tool at regular intervals. Data is sent using secure HTTP or MQTT protocols.

[0110] Data analysis and anomaly detection

[0111] The server uses a generative AI model to analyze the received audio and video data. Specifically, it uses machine learning frameworks such as TENSORFLOW (registered trademark) and PyTorch to detect abnormal behavior and abnormal audio. Abnormal behavior is detected as violent acts, and abnormal audio is detected as words such as "help me."

[0112] When the server detects an anomaly, it generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time of occurrence, and the location of the anomaly. The content of the alert contains enough information for the user to take prompt action.

[0113] Alert Notifications

[0114] The server generates alerts and notifies users via email, text message, or a dedicated app, using external services such as SendGrid and Twilio APIs to send notifications.

[0115] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, allowing them to quickly grasp the situation on-site. Based on this information, they can take appropriate action.

[0116] Data storage and report generation

[0117] The server stores the collected audio and video data for a certain period of time. The data is stored in cloud storage such as Amazon S3 or Google Cloud Storage. The stored data is used for later verification and report creation.

[0118] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user in PDF or Excel format, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[0119] Specific examples

[0120] For example, consider a case where a staff member at a nursing home is verbally abusing an elderly person.

[0121] 1. The device captures audio data from within the facility and sends it to a server. Specifically, the device's microphone picks up abusive language such as "idiot" and "die."

[0122] 2. The server analyzes the received audio data and detects abusive language.

[0123] 3. The server detects the abnormality and generates detailed alert information.

[0124] 4. The server sends an alert to the facility manager. Specifically, a push notification is sent to the manager's smartphone.

[0125] 5. The user (facility manager) checks the notification, rushes to the scene, records and saves the situation, and warns the staff.

[0126] In this way, the system of the present invention makes full use of AI technology to ensure safety within the facility and enables early detection of abuse and rapid response.

[0127] Prompt Sentence Examples

[0128] Here is an example of starting the system's processing by inputting the following prompt sentence into the generative AI model:

[0129] Set up an AI device at nursing home ABC and start real-time monitoring. Set it up to send an alert to the administrator as soon as it detects abusive language or abnormal behavior by staff.

[0130] This prompt is used as an example to demonstrate system configuration and processing.

[0131] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0132] Step 1:

[0133] Users access the system's website and create an account.

[0134] Input: Facility name, address, contact information, etc.

[0135] Process: The user enters this information into the registration form and clicks the "Submit" button.

[0136] Output: The data is sent to the server.

[0137] Step 2:

[0138] The server stores user account information in a database.

[0139] Input: Information such as facility name, address, and contact details submitted by the user.

[0140] Processing: Information is stored using database software (e.g., MySQL).

[0141] Output: The user account is registered in the database and a notification is sent to the user confirming account creation.

[0142] Step 3:

[0143] Once a user signs up for a subscription, they will be sent an AI device.

[0144] Input: Subscription contract information.

[0145] Processing: After the contract is confirmed, a specified number of AI devices will be sent to the user.

[0146] Output: The AI ​​device arrives at the user's facility.

[0147] Step 4:

[0148] Users install AI devices within their facilities and connect them to a network and power source.

[0149] Inputs: AI device, network information, power outlet.

[0150] Process: Place the device in its designated location, plug it into a power outlet, and configure Wi-Fi.

[0151] Output: The AI ​​device starts up.

[0152] Step 5:

[0153] After the device starts up, it automatically sends its identification information to the server.

[0154] Input: Device identification information (MAC address, unique ID, etc.).

[0155] Processing: The device establishes a network connection and sends its identification information to the server.

[0156] Output: The identification information is sent to the server.

[0157] Step 6:

[0158] The server associates the device identity with the user account.

[0159] Input: Device identification information, user account information.

[0160] Action: Updates the database to associate the device information with the user account.

[0161] Output: The device is associated with the user account.

[0162] Step 7:

[0163] The device continuously collects audio and video data from the location where it is installed.

[0164] Input: Location audio and video data.

[0165] Processing: The device's built-in camera and microphone capture audio and video.

[0166] Output: Collected data is temporarily stored on the device.

[0167] Step 8:

[0168] The device sends the collected data to a server in real time.

[0169] Input: Collected audio and video data.

[0170] Processing: Data is sent using secure HTTP or MQTT protocols.

[0171] Output: The data is sent to the server.

[0172] Step 9:

[0173] The server stores the received data in a buffer.

[0174] Input: Audio and video data sent from the device.

[0175] Processing: Storing data in RAM and temporary storage.

[0176] Output: The data is held in a buffer.

[0177] Step 10:

[0178] The server uses AI algorithms to analyze the data held in the buffer.

[0179] Input: Buffered audio and video data.

[0180] Processing: Detect anomalous behavior and sounds using TensorFlow and PyTorch.

[0181] Output: The analysis results are obtained.

[0182] Step 11:

[0183] The server detects abnormal behavior or sounds and generates alert information.

[0184] Input: Analysis results of the AI ​​model.

[0185] Processing: Generate an alert based on information such as the type of anomaly, the time of occurrence, and the location of the anomaly.

[0186] Output: Alert information is generated.

[0187] Step 12:

[0188] The server generates alerts and notifies users via email, text message, or a dedicated app.

[0189] Input: Alert information.

[0190] Processing: Send notifications using SendGrid or Twilio APIs.

[0191] Output: The alert is notified to the user.

[0192] Step 13:

[0193] The user checks the received alert notification and obtains detailed information.

[0194] Input: Alert notification.

[0195] Action: Access the dedicated app or website to check detailed information.

[0196] Output: Information is obtained to understand the situation on site.

[0197] Step 14:

[0198] The server stores the collected audio and video data for a certain period of time.

[0199] Input: Collected audio and video data.

[0200] Processing: Store data in Amazon S3 or Google Cloud Storage.

[0201] Output: Data is saved to cloud storage.

[0202] Step 15:

[0203] The server periodically generates a report summarizing the results of the data analysis.

[0204] Input: Analysis results, past data.

[0205] Processing: Generates reports in PDF and Excel formats and provides them to the user.

[0206] Output: A report is provided to understand the status of the facility.

[0207] (Application example 1)

[0208] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0209] In modern society, early detection and prevention of abuse in facilities such as daycare centers and elderly care facilities is a major challenge. However, conventional monitoring systems have limitations in their ability to analyze massive amounts of data in real time and quickly detect abnormal behavior. This increases the time between the occurrence of an abnormality and a response, risking the spread of damage. In addition, multiple notification methods and devices are often not properly coordinated, hindering administrators' ability to respond quickly.

[0210] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0211] In this invention, the server includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, a cloud storage means, an abnormal behavior analysis means using a generative AI model, and a push notification means. This allows for real-time analysis of audio and video data collected within the facility, enabling the immediate detection of abnormal behavior or abnormal sounds. Furthermore, by using multiple notification methods, managers can quickly identify abnormalities and take necessary measures.

[0212] "Data collection means" refers to a system that uses devices and sensors to collect data such as audio and video in real time.

[0213] "Data analysis means" refers to a system that includes software and algorithms for processing collected data and detecting abnormal behavior or sounds.

[0214] An "anomaly detection means" is a system that includes functions and algorithms for automatically detecting abnormal behavior or abnormal sounds from collected and analyzed data.

[0215] An "alert notification method" is a communication method such as email, text message, in-app notification, or push notification that is used to warn an administrator when an abnormality is detected.

[0216] "Data storage means" refers to a storage system that stores collected audio and video data for a certain period of time so that it can be verified or referenced later.

[0217] "Cloud storage means" refers to a remote storage service for storing and managing data over the Internet, enabling the storage of large amounts of data without being restricted by physical devices.

[0218] "Means for analyzing abnormal behavior using generative AI models" refers to a function that uses artificial intelligence (AI) models to detect and identify unusual behavior patterns when analyzing collected data.

[0219] "Push notification" is a technology for sending instant notifications to smartphones and other devices, providing users with important information in real time.

[0220] This invention is a system for preventing abuse in facilities such as nurseries and nursing homes. The system is made up of the following main components, and its specific implementation method is described below.

[0221] 1. Data Collection Methods

[0222] The data collection methods include devices and sensors that collect audio and video data in real time within the facility. Specifically, microphones and cameras are used. These devices are installed in various locations within the facility, continuously collect data, and transmit it to a server via a network.

[0223] 2. Data analysis methods

[0224] The data analysis means includes software and algorithms for processing the collected data and detecting abnormal behavior and sounds. Specific software used is a Python program and an external analysis API. This allows data to be analyzed in real time and abnormal behavior and sounds to be detected.

[0225] 3. Anomaly detection methods

[0226] The anomaly detection method includes functions and algorithms that automatically detect abnormal behavior and sounds from collected and analyzed data. Specifically, a generative AI model is used. This AI model learns behaviors that deviate from normal behavior patterns and identifies abnormalities.

[0227] 4. Alert notification methods

[0228] Alert notification means include communication means such as email, text message, in-app notification, and push notification to alert administrators when an abnormality is detected, allowing administrators to immediately identify the abnormality and take prompt action.

[0229] 5. Data storage means

[0230] The data storage means includes a storage system that stores collected audio and video data for a certain period of time so that it can be verified and referenced later. Specifically, Firebase storage is used. This allows past data to be stored and verified when necessary.

[0231] 6. Cloud Storage Solutions

[0232] Cloud storage services are remote storage services for storing and managing data over the Internet, allowing for the storage of large amounts of data without being restricted by physical devices. Specifically, Firebase Storage is one such service.

[0233] 7. Methods for analyzing abnormal behavior using generative AI models

[0234] The abnormal behavior analysis method using a generative AI model includes a function that uses an artificial intelligence (AI) model to detect and identify abnormal behavior patterns when analyzing collected data, thereby enabling the detection of abnormal behavior with high accuracy.

[0235] 8. Push Notification Methods

[0236] Push notification is a technology for instantly sending notifications to smartphones and other devices, including a means to provide users with important information in real time, allowing administrators to quickly identify and respond to abnormalities.

[0237] Specific examples

[0238] For example, if a staff member in a nursing home is verbally abusing an elderly person, the system will function as follows:

[0239] 1. The device captures audio data within the facility and sends it to the server.

[0240] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[0241] 3. The server detects the abnormality and generates alert information.

[0242] 4. The server sends a notification to the facility manager via email or a dedicated app.

[0243] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[0244] Prompt Sentence Examples

[0245] "Please tell me how to detect abnormal sounds while working at a nursing home."

[0246] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0247] Step 1:

[0248] The user registers an account with the system along with facility information. The user enters facility information (facility name, address, contact information, etc.) and sends it to the server. This information is saved in a database by the server and linked to the user account. The input is facility information, and the output is the user profile saved in the database.

[0249] Step 2:

[0250] A user installs a device in a facility and connects it to the network and power. When the device starts up, it automatically sends its identification information to the server. The server receives this identification information and associates the device with a specific user account. The input is the device's identification information, and the output is the device information stored in a database.

[0251] Step 3:

[0252] The device collects audio and video data in the facility in real time and sends it to the server. The server stores the received data in a buffer and inputs it into the data analysis means at regular time intervals. The input is the audio and video data transmitted in real time, and the output is the data temporarily stored in the buffer.

[0253] Step 4:

[0254] The server analyzes the collected audio and video data using a generative AI model. The purpose of the analysis is to detect abnormal behavior or sounds that deviate from normal behavior patterns. The input is the audio and video data stored in the buffer, and the output is the analysis results, which are information on abnormal behavior or sounds.

[0255] Step 5:

[0256] If an anomaly is detected, the server generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time it occurred, and the location where it occurred. The input is the anomaly detection information as an analysis result, and the output is the alert information.

[0257] Step 6:

[0258] The server generates alerts and sends them to administrators via email, text message, in-app notification, or push notification. The input is the alert information, and the output is the notification received by the administrator.

[0259] Step 7:

[0260] The administrator receives a notification and accesses a dedicated app or website to check detailed information. The input is a notification of an abnormality detection, and the output is the result of checking the detailed information. Based on this information, the administrator can rush to the site and take appropriate measures.

[0261] Step 8:

[0262] The server stores the collected audio and video data for a certain period of time and uses it for later verification and report creation. The input is the collected audio and video data, and the output is the data saved in storage. This makes it possible to reference past data as needed.

[0263] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0264] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[0265] Account Creation and Initial Setup

[0266] First, users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0267] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[0268] Device Registration and Connection

[0269] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[0270] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[0271] Data Collection and Monitoring

[0272] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[0273] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[0274] Data analysis and anomaly detection

[0275] The server uses AI algorithms to analyze the collected audio and video data.

[0276] AI algorithms detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[0277] Emotion recognition by emotion engine

[0278] The emotion engine installed on the server analyzes the collected audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[0279] The server uses the emotional state recognized by the emotion engine as additional input when the anomaly detection means determines an anomaly.

[0280] Alert Notifications

[0281] If an anomaly is detected, the server generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state.

[0282] The server notifies the user of the alert via email, SMS, or a dedicated in-app notification, or a combination of these.

[0283] Data storage and report generation

[0284] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[0285] The server periodically generates a report summarizing the analysis results and provides it to the user, who can use this report to understand the safety status of the entire facility and take any necessary improvement measures.

[0286] Specific examples

[0287] For example, let's assume that a staff member at a nursing home is using abusive or insulting language towards an elderly person. In this case, the system works as follows:

[0288] 1. The device captures audio and video data from within the facility and sends it to the server.

[0289] 2. The server analyzes the received data and detects abnormal voices, including abusive or insulting language.

[0290] 3. The server uses an emotion engine to recognize the emotional state of the elderly person (e.g., fear, pain) from the video data.

[0291] 4. The server generates alert information based on the abnormality and emotion recognition results and notifies the facility manager of this information.

[0292] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (such as warning staff or implementing countermeasures).

[0293] In this way, by combining an emotion engine, the present invention provides a system that achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, which contributes to improving safety within facilities and is particularly effective in ensuring the welfare of the elderly and children.

[0294] The processing flow will be explained below.

[0295] Step 1:

[0296] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0297] Step 2:

[0298] The server receives the user's input information, saves the account information in a database, and generates authentication information (user ID and password) and sends it to the user's email address.

[0299] Step 3:

[0300] Users sign a subscription contract and receive a specified number of AI devices, which they then install in appropriate locations within their facility and connect to a network and power source.

[0301] Step 4:

[0302] When the device starts up, it sends its identification information (device ID, MAC address, etc.) to the server.

[0303] Step 5:

[0304] The server receives the identification information and associates it with a user account, thereby establishing that the device belongs to a particular facility.

[0305] Step 6:

[0306] The device continuously collects audio and video data within the facility and transmits this data to a server in real time.

[0307] Step 7:

[0308] The server stores the received data in a buffer and performs analysis using a data analysis method. An AI algorithm is used.

[0309] Step 8:

[0310] The server's AI algorithm analyzes audio and video data to detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[0311] Step 9:

[0312] An emotion engine installed on the server analyzes audio and video data and recognizes the user's emotional state (e.g., fear, pain, etc.).

[0313] Step 10:

[0314] The server uses the emotional state as additional data to improve the accuracy of anomaly detection. If the emotional state is associated with an anomaly, the server will determine the anomaly with higher sensitivity.

[0315] Step 11:

[0316] As a final analysis result, the server generates alert information if an anomaly is detected, which includes details of abnormal behavior, abnormal sound, recognized emotional state, time of occurrence, and location of occurrence.

[0317] Step 12:

[0318] Based on the alert information, the server sends a notification to the user using one or more of email, SMS, and / or a notification within a dedicated app.

[0319] Step 13:

[0320] Users receive an alert notification and can access a dedicated app or website to check detailed information, allowing them to understand the situation on-site in detail and take prompt action.

[0321] Step 14:

[0322] The server stores the collected data for a certain period of time, which allows for future reference and verification of the data.

[0323] Step 15:

[0324] The server periodically generates a report summarizing the analysis results and provides it to the user, helping them understand the safety status of the entire facility and take any necessary improvement measures.

[0325] In this way, through a series of processing steps, the system can quickly detect acts of abuse in daycare centers and elderly care facilities and take immediate action.By combining it with an emotion engine, it becomes possible to detect anomalies with high accuracy, taking into account emotional states.

[0326] Example 2

[0327] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0328] To prevent abuse in daycare centers and elderly care facilities, it is necessary to accurately detect abnormal behavior and sounds within the facility and take prompt action. However, conventional monitoring systems are limited to detecting abnormal behavior and sounds, and do not take emotional states into account when detecting abnormalities. Therefore, a system that can more reliably ensure safety within the facility is needed.

[0329] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0330] In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means, which not only detects abnormal behavior and abnormal voice from audio and video data but also recognizes emotional states, enabling more accurate detection of abnormalities and quicker response.

[0331] "Data collection means" means a device or system for continuously collecting audio and video data within a facility.

[0332] "Data analysis means" refers to software or algorithms that analyze collected audio and video data and detect abnormal behavior or abnormal sounds.

[0333] The "anomaly detection means" is an algorithm or system for identifying abnormal behavior or abnormal sounds from the data analyzed by the data analysis means and detecting abnormalities.

[0334] "Emotion recognition means" refers to software or algorithms that analyze and recognize emotional states based on collected audio and video data.

[0335] An "alert notification means" is a device or system for notifying a user of information about a detected abnormality, and includes means such as email, text message, and in-app notification.

[0336] "Data storage means" refers to a device or system that stores collected and analyzed data for a certain period of time and makes it available for later reference.

[0337] "Server" means a central processing unit for collecting, analyzing, storing, and notifying data.

[0338] An "AI algorithm" is a computational method that uses machine learning and deep learning to analyze audio and video data and detect anomalies.

[0339] A "time frame" is a unit of time for processing data by dividing it into regular time intervals.

[0340] A "buffer" is a storage area for temporarily storing data.

[0341] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an emotion recognition means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[0342] To start using the system, users first access the system's website and create an account. They enter the required information (facility name, address, contact information, etc.) and submit it. The server then receives the entered information and stores it in a database. This allows for the configuration and management of each facility.

[0343] Next, once the user completes the subscription contract, they will receive a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After booting up, the devices will send their identification information (device ID, MAC address, etc.) to the server.

[0344] The devices are installed at various locations within the facility and continuously collect audio and video data using cameras and microphones. The collected data is transmitted in real time to a server, which temporarily stores it in a buffer. The data is then input into a data analysis tool every certain time frame (e.g., 5 or 10 seconds).

[0345] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying). It also uses an emotion engine to analyze audio and video data to recognize the user's emotional state. Voice tone and facial expression data are used to analyze the emotional state.

[0346] This emotional state is important auxiliary data for the anomaly detection means to determine an abnormality. If an abnormality is detected, the server generates detailed alert information. This information includes the type of abnormality, the time of occurrence, the location of occurrence, and the recognized emotional state. The generated alert information is notified to the user via one or more of email, text message, and / or dedicated in-app notification.

[0347] Furthermore, the server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to grasp the safety status of the entire facility and take necessary improvement measures.

[0348] Specific examples

[0349] For example, consider a situation where a staff member at a nursing home uses abusive or insulting language toward an elderly person. In this case, a device captures audio and video data from within the facility and sends it to a server. The server then stores the received data in a buffer and analyzes it using an AI algorithm. Abnormal audio, including abusive or insulting language, is detected. The server then uses an emotion engine to recognize the elderly person's emotional state (e.g., fear, distress) from the video data. An alert is then generated based on the abnormal behavior and the emotion recognition results, and this information is sent to the facility manager. The user (facility manager) who receives the notification can check the details, rush to the scene, and take the necessary action (e.g., warn staff, implement immediate countermeasures).

[0350] In this way, by combining the emotion engine, this system achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, thereby improving safety within facilities and ensuring the welfare of the elderly and children.

[0351] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0352] Step 1:

[0353] The user accesses the system's website, creates an account, enters the necessary information such as the facility name, address, and contact information, and submits it.

[0354] Input: User information such as facility name, address, and contact details.

[0355] Process: Sends the information entered by the user to the server.

[0356] Output: User information is saved on the server.

[0357] Step 2:

[0358] The server receives the information sent by the user and stores it in a database, which allows for the configuration and management of each facility.

[0359] Input: User-submitted information such as facility name, address, and contact details.

[0360] Processing: The server stores the information in a database.

[0361] Output: User information stored in the database.

[0362] Step 3:

[0363] Once a user completes a subscription, a set number of AI devices will be delivered to the facility, which the user must then install and connect to the network and power source.

[0364] Input: Subscription contract information, AI device.

[0365] Processing: The AI ​​device is installed on-site and connected to a network and power source.

[0366] Output: AI device installed within the facility.

[0367] Step 4:

[0368] After the device starts up, it sends its own identification information (device ID, MAC address, etc.) to the server.

[0369] Input: Device ID, MAC Address.

[0370] Process: The device starts up and sends its identification information to the server.

[0371] Output: The device identification sent to the server.

[0372] Step 5:

[0373] The devices are installed at various locations within the facility and use cameras and microphones to continuously collect audio and video data.

[0374] Input: Location audio and video data.

[0375] Processing: Capture audio and video data in real time.

[0376] Output: Collected audio and video data.

[0377] Step 6:

[0378] The server receives the collected data in real time, temporarily stores it in a buffer, and then inputs it into the data analysis means every certain time frame (e.g., 5 seconds, 10 seconds).

[0379] Input: Collected audio and video data.

[0380] Processing: Data is held in a buffer and input to a data analysis tool for each time frame.

[0381] Output: Audio and video data input to data analysis means.

[0382] Step 7:

[0383] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[0384] Input: Input audio and video data per time frame.

[0385] Processing: Analyze the data using AI algorithms to detect abnormal behavior and audio.

[0386] Output: Detected abnormal behavior and audio.

[0387] Step 8:

[0388] The server uses an emotion engine to analyze the audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[0389] Input: Input audio and video data per time frame.

[0390] Processing: Analyze the emotional state using the emotion engine.

[0391] Output: Perceived emotional state.

[0392] Step 9:

[0393] If the server detects an anomaly, it generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state. The generated alert information is then sent to the user via email, text message, or in-app notification.

[0394] Input: Detected abnormal behavior and emotional state.

[0395] Action: Generate and notify alert information.

[0396] Output: Alert information notified to the user.

[0397] Step 10:

[0398] The server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to understand the safety status of the entire facility and take necessary improvement measures.

[0399] Input: Collected audio and video data.

[0400] Processing: The data is stored for a certain period of time and the analysis results are generated as a report.

[0401] Output: The report provided to the user.

[0402] (Application example 2)

[0403] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0404] The problem to be solved by this invention is to effectively prevent abuse in nurseries and nursing homes and to increase safety within the facilities. In particular, the object is to provide a system that can detect abuse at an early stage and take prompt and appropriate action.

[0405] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means. This allows for real-time data collection from devices installed in the facility and analysis of the collected data, enabling early detection of abusive behavior and abnormal behavior and recognition of emotional states based on voice tone and facial expression data. Furthermore, when an abnormality is detected, an alert can be quickly sent via email, text message, or a dedicated app, prompting facility managers and relatives to take immediate action. Furthermore, by storing video and audio data along with the analysis results for a certain period of time, the data can be referenced later, providing evidence of problematic behavior.

[0406] "Data collection means" refers to means for collecting audio data and video data from devices installed within the facility.

[0407] The "data analysis means" is a means for analyzing collected audio data and video data and extracting necessary information.

[0408] "Abnormality detection means" is a means for detecting abusive behavior or abnormal behaviour from information extracted through data analysis.

[0409] The "emotion recognition means" is a means for analyzing voice tone and facial expression data to recognize the emotional state of a subject.

[0410] The "alert notification means" is a means for notifying relevant users of an alert when an abnormality is detected using email, text message, and app notification.

[0411] "Data storage means" refers to a means for storing collected data and analysis results for a certain period of time so that they can be referenced later.

[0412] A "server" is a central processing unit that integrates data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means, and functions as a system.

[0413] This system includes data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means to prevent abuse in facilities such as daycare centers and elderly care facilities. The server is at the core of this system, integrating and managing each device and user device.

[0414] Installing and connecting the device

[0415] A certain number of AI devices will be installed within the facility. These devices will be connected to the network and power source, and after startup, will send their identification information (device ID, MAC address, etc.) to the server.

[0416] Data collection and analysis

[0417] The device continuously collects audio and video data at the installation location and transmits this data to the server in real time. The server stores the received data in a buffer and inputs it to the data analysis means at regular intervals. The data analysis means mainly uses the following software libraries:

[0418] Keras: A machine learning library for implementing emotion engine models.

[0419] OpenCV: A computer vision library for preprocessing image data and analyzing video data.

[0420] Anomaly detection and emotion recognition

[0421] The server uses AI algorithms to analyze the collected audio and video data. The anomaly detection means detects abnormal behavior and sounds, such as violent actions, screaming, or crying. The emotion recognition means analyzes voice tone and facial expression data to recognize the subject's emotional state (e.g., fear, pain). This improves the accuracy of anomaly detection.

[0422] Alert Notifications

[0423] If an abnormality is detected, the server generates detailed alert information and notifies facility managers and relatives of the incident. The alert notification method uses the following communication methods:

[0424] Email

[0425] Text message

[0426] In-app notifications

[0427] Data storage and reporting

[0428] The server stores the collected data and analysis results for a certain period of time, allowing the data to be referenced later and providing evidence of problematic behavior.The server also periodically generates a report summarizing the analysis results and provides it to users, making it easier to understand the safety status of the entire facility.

[0429] Specific examples

[0430] For example, if a staff member at a nursing home uses abusive or insulting language towards an elderly person, the system works as follows:

[0431] 1. The device captures audio and video data from within the facility and sends it to the server.

[0432] 2. The server analyzes the received data and detects abnormal voice and behavior, including abusive or insulting language.

[0433] 3. The emotion recognition means recognizes the elderly person's emotional state (e.g., fear, distress) from voice tone and facial expression data.

[0434] 4. Based on the results of anomaly detection and emotion recognition, the server generates alert information and notifies facility managers and relatives.

[0435] 5. Users can receive notifications, check detailed information, rush to the scene, and take necessary action (such as alerting staff or implementing countermeasures).

[0436] Example prompts for generative AI models

[0437] "We will demonstrate an anomaly detection system for abusive language and insulting behavior in a nursing home. The camera collects video and audio in real time, and if the emotion engine detects abnormal behavior, it will send an alert to the administrator."

[0438] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0439] Step 1:

[0440] Device installation and initial setup:

[0441] Users install a set number of AI devices within their facilities and connect them to a network and power source. After starting up, the devices send their identification information (device ID, MAC address, etc.) to the server, which then recognizes the devices and registers them in the system.

[0442] Input: Device identification information

[0443] Output: Device registration information on the server

[0444] Specific operations: Device network connection, sending of identification information, registration process on the server

[0445] Step 2:

[0446] Data collection:

[0447] The device installed inside the terminal continuously collects audio and video data from within the facility and transmits it to a server in real time.

[0448] Input: Audio and video data

[0449] Output: Raw data sent to the server

[0450] Specific operations: audio and video capture, real-time data transmission

[0451] Step 3:

[0452] Data Analysis:

[0453] The server stores the received audio and video data in a buffer and then analyzes the data using Keras and OpenCV.

[0454] Input: Raw audio and video data

[0455] Output: Analyzed data (e.g., specific motion or voice features)

[0456] Specific operation: Data input, preprocessing, and analysis using machine learning models

[0457] Step 4:

[0458] Anomaly detection:

[0459] Based on the results of the data analysis, the server uses anomaly detection methods to detect abusive or abnormal behavior, such as identifying specific movement features or voice patterns as abnormal.

[0460] Input: Parsed data

[0461] Output: Anomaly detection results

[0462] Specific operations: feature analysis, abnormal / normal determination

[0463] Step 5:

[0464] Emotion recognition:

[0465] The server uses an emotion recognition means to recognize the emotional state of the subject from the analyzed voice tone and facial expression data.

[0466] Input: Voice tone data and facial expression data

[0467] Output: Emotion recognition results (e.g. fear, pain)

[0468] Specific actions: Analysis of voice tone, analysis of facial expression data, estimation of emotional state

[0469] Step 6:

[0470] Alerting and Notification:

[0471] The server generates alert information based on the results of anomaly detection and emotion recognition, and notifies facility managers and relatives through alert notification methods such as email, text message, and in-app notification.

[0472] Input: Anomaly detection results and emotion recognition results

[0473] Output: Alert information, notification message

[0474] Specific actions: Generate alert information, send notification messages

[0475] Step 7:

[0476] Data storage and reporting:

[0477] The server stores the collected data and analysis results for a certain period of time, and periodically generates a report summarizing the analysis results and provides it to the user. This allows the data to be referenced later, providing evidence of problematic behavior.

[0478] Input: Collected data, analysis results

[0479] Output: Saved data, scheduled reports

[0480] Specific actions: saving data, generating reports, and providing them to users

[0481] Through the above processing steps, the present system can enhance safety within the facility and ensure the welfare of the elderly and children in particular.

[0482] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0483] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0484] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0485] [Second embodiment]

[0486] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0487] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0488] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0489] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0490] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0491] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0492] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0493] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0494] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0495] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0496] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0497] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0498] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[0499] Account Creation and Initial Setup

[0500] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and then send it to the server.

[0501] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[0502] Device Registration and Connection

[0503] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[0504] After a device boots up, it automatically sends identifying information to a server, which uses this information to associate the device with a specific user account.

[0505] Data Collection and Monitoring

[0506] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[0507] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[0508] Data analysis and anomaly detection

[0509] The server uses AI algorithms to analyze the received audio and video data, with the goal of detecting any abnormal behavior or sounds that deviate from normal patterns of behavior.

[0510] If the server detects any abnormal behavior or sound, it generates an alert containing information about the abnormality, including the type of abnormality, the time of occurrence, and the location where it occurred.

[0511] Alert Notifications

[0512] The server generates an alert and sends it via email, SMS, or a dedicated app, or a combination of these.

[0513] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, enabling them to take prompt action.

[0514] Data storage and report generation

[0515] The server stores the collected audio and video data for a certain period of time, and the stored data is used for later verification and report creation.

[0516] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[0517] Specific examples

[0518] For example, let's assume that a staff member at a nursing home is verbally abusing an elderly person. In this case, the system works as follows:

[0519] 1. The device captures audio data within the facility and sends it to the server.

[0520] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[0521] 3. The server detects the abnormality and generates alert information.

[0522] 4. The server sends a notification to the facility manager via email or a dedicated app.

[0523] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[0524] In this way, the present invention provides a system that utilizes AI technology to ensure safety within facilities and enables early detection of abuse and rapid response.

[0525] The processing flow will be explained below.

[0526] Step 1:

[0527] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0528] Step 2:

[0529] The server creates an account based on the received information, stores it in the database, and sends authentication information (user ID and password) to the user's email address.

[0530] Step 3:

[0531] The user signs up for a subscription and receives the device, which they then install in an appropriate location and connect to the network and power.

[0532] Step 4:

[0533] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[0534] Step 5:

[0535] The server verifies the identity from the device and associates it with the user's account, thereby confirming that the device belongs to a particular facility.

[0536] Step 6:

[0537] The device collects audio and video data in real time, which is then immediately sent to a server.

[0538] Step 7:

[0539] The server buffers the received data and inputs it into a data analysis means, which includes an AI algorithm.

[0540] Step 8:

[0541] AI algorithms installed on the server analyze audio and video data, including detecting abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[0542] Step 9:

[0543] If an anomaly is detected, the server generates an alert, which includes details such as the type of anomaly, the time it occurred, and the location where it occurred.

[0544] Step 10:

[0545] Based on the generated alert information, the server sends a notification to the user using the specified notification method (email, SMS, or in-app notification).

[0546] Step 11:

[0547] Users can receive alert notifications and access a dedicated app or website to check detailed information, making it easier to understand the situation on-site and enable prompt action.

[0548] Step 12:

[0549] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[0550] Step 13:

[0551] The server periodically generates a report summarizing the analysis results, allowing users to understand the safety status of the entire facility and take any necessary improvement measures.

[0552] In this way, the system monitors the situation within the facility in real time, promptly notifies the user when an abnormality occurs, and ensures the safety of the facility by allowing the user to take appropriate action. This series of processing steps enables early detection of abuse and rapid response.

[0553] Example 1

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

[0555] Abuse in daycare centers and elderly care facilities is a major social problem, and there is a need for a rapid and accurate monitoring system to prevent it from happening. These facilities also face serious labor shortages, necessitating a highly automated system. Current monitoring systems have difficulty detecting abnormal behavior and sounds in real time and responding quickly. Therefore, there is a need for a system that utilizes AI technology to efficiently detect abnormalities while ensuring safety within the facility.

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

[0557] In this invention, the server includes a means for creating an account, a means for transmitting device identification information to the server and linking it to a user account, a means for collecting audio and video data, a means for analyzing the received data by time frame, a means for detecting abnormal behavior or abnormal sounds using an AI algorithm, a means for generating an alert containing abnormal information, a means for notifying the generated alert via email, text message, or app, a means for storing the audio and video data for a certain period of time, and a means for generating a report of the results of data analysis. This enables early detection of possible acts of abuse occurring within the facility and rapid response.

[0558] The "account creation means" refers to the means by which a user accesses the system's website and creates an account by entering necessary information such as the facility name, location, and contact information.

[0559] The "means of transmitting device identification information to a server and associating it with a user account" refers to a means in which the device transmits identification information to a server after startup, and the server associates the device with a specific user account based on that information.

[0560] "Means for collecting audio and video data" means means for continuously capturing audio and video data by devices installed within the facility.

[0561] The "means for analyzing received data for each time frame" refers to a means for storing data received by the server in a buffer and analyzing the data for each fixed time frame.

[0562] "Means for detecting abnormal behavior and sounds using AI algorithms" refers to a means by which the server uses AI technology to identify abnormal behavior and sounds from audio and video data.

[0563] The "means for generating an alert including abnormality information" refers to a means for generating an alert including information such as the type of abnormality, the time of occurrence, and the location of occurrence when the server detects abnormal behavior or abnormal sound.

[0564] "Means of notifying users of generated alerts via email, text message, or app" refers to means of notifying users of server-generated alerts via email, text message, or a dedicated app.

[0565] "Means for storing audio and video data for a certain period of time" refers to the means by which the server stores collected audio and video data for a certain period of time for the purpose of later verification and report creation.

[0566] The "means for generating a report of the data analysis results" refers to a means by which the server periodically generates a report summarizing the results of the data analysis and anomaly detection, and provides it to the user.

[0567] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[0568] Account Creation and Initial Setup

[0569] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and send it to the server. The server stores the user's account information in a database, allowing it to configure and manage each facility. The database uses a general-purpose database management system such as MySQL or PostgreSQL.

[0570] Device Registration and Connection

[0571] When a user signs up for a subscription, they are sent a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After the devices start up, they automatically send identification information to a server. The server uses this information to associate the devices with specific user accounts. Devices are identified using MAC addresses or unique IDs.

[0572] Data Collection and Monitoring

[0573] The device continuously collects audio and video data at the installation site. The collected data is sent in real time to a server, which stores the data in a buffer and inputs it into a data analysis tool at regular intervals. Data is sent using secure HTTP or MQTT protocols.

[0574] Data analysis and anomaly detection

[0575] The server uses a generative AI model to analyze the received audio and video data. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to detect abnormal behavior and abnormal audio. Abnormal behavior is detected as violent acts, and abnormal audio is detected as words such as "help me."

[0576] When the server detects an anomaly, it generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time of occurrence, and the location of the anomaly. The content of the alert contains enough information for the user to take prompt action.

[0577] Alert Notifications

[0578] The server generates alerts and notifies users via email, text message, or a dedicated app, using external services such as SendGrid and Twilio APIs to send notifications.

[0579] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, allowing them to quickly grasp the situation on-site. Based on this information, they can take appropriate action.

[0580] Data storage and report generation

[0581] The server stores the collected audio and video data for a certain period of time. The data is then stored in cloud storage such as Amazon S3 or Google Cloud Storage. The stored data can be used for later verification and report creation.

[0582] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user in PDF or Excel format, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[0583] Specific examples

[0584] For example, consider a case where a staff member at a nursing home is verbally abusing an elderly person.

[0585] 1. The device captures audio data from within the facility and sends it to a server. Specifically, the device's microphone picks up abusive language such as "idiot" and "die."

[0586] 2. The server analyzes the received audio data and detects abusive language.

[0587] 3. The server detects the abnormality and generates detailed alert information.

[0588] 4. The server sends an alert to the facility manager. Specifically, a push notification is sent to the manager's smartphone.

[0589] 5. The user (facility manager) checks the notification, rushes to the scene, records and saves the situation, and warns the staff.

[0590] In this way, the system of the present invention makes full use of AI technology to ensure safety within the facility and enables early detection of abuse and rapid response.

[0591] Prompt Sentence Examples

[0592] Here is an example of starting the system's processing by inputting the following prompt sentence into the generative AI model:

[0593] Set up an AI device at nursing home ABC and start real-time monitoring. Set it up to send an alert to the administrator as soon as it detects abusive language or abnormal behavior by staff.

[0594] This prompt is used as an example to demonstrate system configuration and processing.

[0595] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0596] Step 1:

[0597] Users access the system's website and create an account.

[0598] Input: Facility name, address, contact information, etc.

[0599] Process: The user enters this information into the registration form and clicks the "Submit" button.

[0600] Output: The data is sent to the server.

[0601] Step 2:

[0602] The server stores user account information in a database.

[0603] Input: Information such as facility name, address, and contact details submitted by the user.

[0604] Processing: Information is stored using database software (e.g., MySQL).

[0605] Output: The user account is registered in the database and a notification is sent to the user confirming account creation.

[0606] Step 3:

[0607] Once a user signs up for a subscription, they will be sent an AI device.

[0608] Input: Subscription contract information.

[0609] Processing: After the contract is confirmed, a specified number of AI devices will be sent to the user.

[0610] Output: The AI ​​device arrives at the user's facility.

[0611] Step 4:

[0612] Users install AI devices within their facilities and connect them to a network and power source.

[0613] Inputs: AI device, network information, power outlet.

[0614] Process: Place the device in its designated location, plug it into a power outlet, and configure Wi-Fi.

[0615] Output: The AI ​​device starts up.

[0616] Step 5:

[0617] After the device starts up, it automatically sends its identification information to the server.

[0618] Input: Device identification information (MAC address, unique ID, etc.).

[0619] Processing: The device establishes a network connection and sends its identification information to the server.

[0620] Output: The identification information is sent to the server.

[0621] Step 6:

[0622] The server associates the device identity with the user account.

[0623] Input: Device identification information, user account information.

[0624] Action: Updates the database to associate the device information with the user account.

[0625] Output: The device is associated with the user account.

[0626] Step 7:

[0627] The device continuously collects audio and video data from the location where it is installed.

[0628] Input: Location audio and video data.

[0629] Processing: The device's built-in camera and microphone capture audio and video.

[0630] Output: Collected data is temporarily stored on the device.

[0631] Step 8:

[0632] The device sends the collected data to a server in real time.

[0633] Input: Collected audio and video data.

[0634] Processing: Data is sent using secure HTTP or MQTT protocols.

[0635] Output: The data is sent to the server.

[0636] Step 9:

[0637] The server stores the received data in a buffer.

[0638] Input: Audio and video data sent from the device.

[0639] Processing: Storing data in RAM and temporary storage.

[0640] Output: The data is held in a buffer.

[0641] Step 10:

[0642] The server uses AI algorithms to analyze the data held in the buffer.

[0643] Input: Buffered audio and video data.

[0644] Processing: Detect anomalous behavior and sounds using TensorFlow and PyTorch.

[0645] Output: The analysis results are obtained.

[0646] Step 11:

[0647] The server detects abnormal behavior or sounds and generates alert information.

[0648] Input: Analysis results of the AI ​​model.

[0649] Processing: Generate an alert based on information such as the type of anomaly, the time of occurrence, and the location of the anomaly.

[0650] Output: Alert information is generated.

[0651] Step 12:

[0652] The server generates alerts and notifies users via email, text message, or a dedicated app.

[0653] Input: Alert information.

[0654] Processing: Send notifications using SendGrid or Twilio APIs.

[0655] Output: The alert is notified to the user.

[0656] Step 13:

[0657] The user checks the received alert notification and obtains detailed information.

[0658] Input: Alert notification.

[0659] Action: Access the dedicated app or website to check detailed information.

[0660] Output: Information is obtained to understand the situation on site.

[0661] Step 14:

[0662] The server stores the collected audio and video data for a certain period of time.

[0663] Input: Collected audio and video data.

[0664] Processing: Store data in Amazon S3 or Google Cloud Storage.

[0665] Output: Data is saved to cloud storage.

[0666] Step 15:

[0667] The server periodically generates a report summarizing the results of the data analysis.

[0668] Input: Analysis results, past data.

[0669] Processing: Generates reports in PDF and Excel formats and provides them to the user.

[0670] Output: A report is provided to understand the status of the facility.

[0671] (Application example 1)

[0672] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0673] In modern society, early detection and prevention of abuse in facilities such as daycare centers and elderly care facilities is a major challenge. However, conventional monitoring systems have limitations in their ability to analyze massive amounts of data in real time and quickly detect abnormal behavior. This increases the time between the occurrence of an abnormality and a response, risking the spread of damage. In addition, multiple notification methods and devices are often not properly coordinated, hindering administrators' ability to respond quickly.

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

[0675] In this invention, the server includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, a cloud storage means, an abnormal behavior analysis means using a generative AI model, and a push notification means. This allows for real-time analysis of audio and video data collected within the facility, enabling the immediate detection of abnormal behavior or abnormal sounds. Furthermore, by using multiple notification methods, managers can quickly identify abnormalities and take necessary measures.

[0676] "Data collection means" refers to a system that uses devices and sensors to collect data such as audio and video in real time.

[0677] "Data analysis means" refers to a system that includes software and algorithms for processing collected data and detecting abnormal behavior or sounds.

[0678] An "anomaly detection means" is a system that includes functions and algorithms for automatically detecting abnormal behavior or abnormal sounds from collected and analyzed data.

[0679] An "alert notification method" is a communication method such as email, text message, in-app notification, or push notification that is used to warn an administrator when an abnormality is detected.

[0680] "Data storage means" refers to a storage system that stores collected audio and video data for a certain period of time so that it can be verified or referenced later.

[0681] "Cloud storage means" refers to a remote storage service for storing and managing data over the Internet, enabling the storage of large amounts of data without being restricted by physical devices.

[0682] "Means for analyzing abnormal behavior using generative AI models" refers to a function that uses artificial intelligence (AI) models to detect and identify unusual behavior patterns when analyzing collected data.

[0683] "Push notification" is a technology for sending instant notifications to smartphones and other devices, providing users with important information in real time.

[0684] This invention is a system for preventing abuse in facilities such as nurseries and nursing homes. The system is made up of the following main components, and its specific implementation method is described below.

[0685] 1. Data Collection Methods

[0686] The data collection methods include devices and sensors that collect audio and video data in real time within the facility. Specifically, microphones and cameras are used. These devices are installed in various locations within the facility, continuously collect data, and transmit it to a server via a network.

[0687] 2. Data analysis methods

[0688] The data analysis means includes software and algorithms for processing the collected data and detecting abnormal behavior and sounds. Specific software used is a Python program and an external analysis API. This allows data to be analyzed in real time and abnormal behavior and sounds to be detected.

[0689] 3. Anomaly detection methods

[0690] The anomaly detection method includes functions and algorithms that automatically detect abnormal behavior and sounds from collected and analyzed data. Specifically, a generative AI model is used. This AI model learns behaviors that deviate from normal behavior patterns and identifies abnormalities.

[0691] 4. Alert notification methods

[0692] Alert notification means include communication means such as email, text message, in-app notification, and push notification to alert administrators when an abnormality is detected, allowing administrators to immediately identify the abnormality and take prompt action.

[0693] 5. Data storage means

[0694] The data storage means includes a storage system that stores collected audio and video data for a certain period of time so that it can be verified and referenced later. Specifically, Firebase storage is used. This allows past data to be stored and verified when necessary.

[0695] 6. Cloud Storage Solutions

[0696] Cloud storage services are remote storage services for storing and managing data over the Internet, allowing for the storage of large amounts of data without being restricted by physical devices. Specifically, Firebase Storage is one such service.

[0697] 7. Methods for analyzing abnormal behavior using generative AI models

[0698] The abnormal behavior analysis method using a generative AI model includes a function that uses an artificial intelligence (AI) model to detect and identify abnormal behavior patterns when analyzing collected data, thereby enabling the detection of abnormal behavior with high accuracy.

[0699] 8. Push Notification Methods

[0700] Push notification is a technology for instantly sending notifications to smartphones and other devices, including a means to provide users with important information in real time, allowing administrators to quickly identify and respond to abnormalities.

[0701] Specific examples

[0702] For example, if a staff member in a nursing home is verbally abusing an elderly person, the system will function as follows:

[0703] 1. The device captures audio data within the facility and sends it to the server.

[0704] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[0705] 3. The server detects the abnormality and generates alert information.

[0706] 4. The server sends a notification to the facility manager via email or a dedicated app.

[0707] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[0708] Prompt Sentence Examples

[0709] "Please tell me how to detect abnormal sounds while working at a nursing home."

[0710] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0711] Step 1:

[0712] The user registers an account with the system along with facility information. The user enters facility information (facility name, address, contact information, etc.) and sends it to the server. This information is saved in a database by the server and linked to the user account. The input is facility information, and the output is the user profile saved in the database.

[0713] Step 2:

[0714] A user installs a device in a facility and connects it to the network and power. When the device starts up, it automatically sends its identification information to the server. The server receives this identification information and associates the device with a specific user account. The input is the device's identification information, and the output is the device information stored in a database.

[0715] Step 3:

[0716] The device collects audio and video data in the facility in real time and sends it to the server. The server stores the received data in a buffer and inputs it into the data analysis means at regular time intervals. The input is the audio and video data transmitted in real time, and the output is the data temporarily stored in the buffer.

[0717] Step 4:

[0718] The server analyzes the collected audio and video data using a generative AI model. The purpose of the analysis is to detect abnormal behavior or sounds that deviate from normal behavior patterns. The input is the audio and video data stored in the buffer, and the output is the analysis results, which are information on abnormal behavior or sounds.

[0719] Step 5:

[0720] If an anomaly is detected, the server generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time it occurred, and the location where it occurred. The input is the anomaly detection information as an analysis result, and the output is the alert information.

[0721] Step 6:

[0722] The server generates alerts and sends them to administrators via email, text message, in-app notification, or push notification. The input is the alert information, and the output is the notification received by the administrator.

[0723] Step 7:

[0724] The administrator receives a notification and accesses a dedicated app or website to check detailed information. The input is a notification of an abnormality detection, and the output is the result of checking the detailed information. Based on this information, the administrator can rush to the site and take appropriate measures.

[0725] Step 8:

[0726] The server stores the collected audio and video data for a certain period of time and uses it for later verification and report creation. The input is the collected audio and video data, and the output is the data saved in storage. This makes it possible to reference past data as needed.

[0727] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0728] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[0729] Account Creation and Initial Setup

[0730] First, users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0731] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[0732] Device Registration and Connection

[0733] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[0734] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[0735] Data Collection and Monitoring

[0736] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[0737] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[0738] Data analysis and anomaly detection

[0739] The server uses AI algorithms to analyze the collected audio and video data.

[0740] AI algorithms detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[0741] Emotion recognition by emotion engine

[0742] The emotion engine installed on the server analyzes the collected audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[0743] The server uses the emotional state recognized by the emotion engine as additional input when the anomaly detection means determines an anomaly.

[0744] Alert Notifications

[0745] If an anomaly is detected, the server generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state.

[0746] The server notifies the user of the alert via email, SMS, or a dedicated in-app notification, or a combination of these.

[0747] Data storage and report generation

[0748] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[0749] The server periodically generates a report summarizing the analysis results and provides it to the user, who can use this report to understand the safety status of the entire facility and take any necessary improvement measures.

[0750] Specific examples

[0751] For example, let's assume that a staff member at a nursing home is using abusive or insulting language towards an elderly person. In this case, the system works as follows:

[0752] 1. The device captures audio and video data from within the facility and sends it to the server.

[0753] 2. The server analyzes the received data and detects abnormal voices, including abusive or insulting language.

[0754] 3. The server uses an emotion engine to recognize the emotional state of the elderly person (e.g., fear, pain) from the video data.

[0755] 4. The server generates alert information based on the abnormality and emotion recognition results and notifies the facility manager of this information.

[0756] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (such as warning staff or implementing countermeasures).

[0757] In this way, by combining an emotion engine, the present invention provides a system that achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, which contributes to improving safety within facilities and is particularly effective in ensuring the welfare of the elderly and children.

[0758] The processing flow will be explained below.

[0759] Step 1:

[0760] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0761] Step 2:

[0762] The server receives the user's input information, saves the account information in a database, and generates authentication information (user ID and password) and sends it to the user's email address.

[0763] Step 3:

[0764] Users sign a subscription contract and receive a specified number of AI devices, which they then install in appropriate locations within their facility and connect to a network and power source.

[0765] Step 4:

[0766] When the device starts up, it sends its identification information (device ID, MAC address, etc.) to the server.

[0767] Step 5:

[0768] The server receives the identification information and associates it with a user account, thereby establishing that the device belongs to a particular facility.

[0769] Step 6:

[0770] The device continuously collects audio and video data within the facility and transmits this data to a server in real time.

[0771] Step 7:

[0772] The server stores the received data in a buffer and performs analysis using a data analysis method. An AI algorithm is used.

[0773] Step 8:

[0774] The server's AI algorithm analyzes audio and video data to detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[0775] Step 9:

[0776] An emotion engine installed on the server analyzes audio and video data and recognizes the user's emotional state (e.g., fear, pain, etc.).

[0777] Step 10:

[0778] The server uses the emotional state as additional data to improve the accuracy of anomaly detection. If the emotional state is associated with an anomaly, the server will determine the anomaly with higher sensitivity.

[0779] Step 11:

[0780] As a final analysis result, the server generates alert information if an anomaly is detected, which includes details of abnormal behavior, abnormal sound, recognized emotional state, time of occurrence, and location of occurrence.

[0781] Step 12:

[0782] Based on the alert information, the server sends a notification to the user using one or more of email, SMS, and / or a notification within a dedicated app.

[0783] Step 13:

[0784] Users receive an alert notification and can access a dedicated app or website to check detailed information, allowing them to understand the situation on-site in detail and take prompt action.

[0785] Step 14:

[0786] The server stores the collected data for a certain period of time, which allows for future reference and verification of the data.

[0787] Step 15:

[0788] The server periodically generates a report summarizing the analysis results and provides it to the user, helping them understand the safety status of the entire facility and take any necessary improvement measures.

[0789] In this way, through a series of processing steps, the system can quickly detect acts of abuse in daycare centers and elderly care facilities and take immediate action.By combining it with an emotion engine, it becomes possible to detect anomalies with high accuracy, taking into account emotional states.

[0790] Example 2

[0791] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0792] To prevent abuse in daycare centers and elderly care facilities, it is necessary to accurately detect abnormal behavior and sounds within the facility and take prompt action. However, conventional monitoring systems are limited to detecting abnormal behavior and sounds, and do not take emotional states into account when detecting abnormalities. Therefore, a system that can more reliably ensure safety within the facility is needed.

[0793] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0794] In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means, which not only detects abnormal behavior and abnormal voice from audio and video data but also recognizes emotional states, enabling more accurate detection of abnormalities and quicker response.

[0795] "Data collection means" means a device or system for continuously collecting audio and video data within a facility.

[0796] "Data analysis means" refers to software or algorithms that analyze collected audio and video data and detect abnormal behavior or abnormal sounds.

[0797] The "anomaly detection means" is an algorithm or system for identifying abnormal behavior or abnormal sounds from the data analyzed by the data analysis means and detecting abnormalities.

[0798] "Emotion recognition means" refers to software or algorithms that analyze and recognize emotional states based on collected audio and video data.

[0799] An "alert notification means" is a device or system for notifying a user of information about a detected abnormality, and includes means such as email, text message, and in-app notification.

[0800] "Data storage means" refers to a device or system that stores collected and analyzed data for a certain period of time and makes it available for later reference.

[0801] "Server" means a central processing unit for collecting, analyzing, storing, and notifying data.

[0802] An "AI algorithm" is a computational method that uses machine learning and deep learning to analyze audio and video data and detect anomalies.

[0803] A "time frame" is a unit of time for processing data by dividing it into regular time intervals.

[0804] A "buffer" is a storage area for temporarily storing data.

[0805] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an emotion recognition means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[0806] To start using the system, users first access the system's website and create an account. They enter the required information (facility name, address, contact information, etc.) and submit it. The server then receives the entered information and stores it in a database. This allows for the configuration and management of each facility.

[0807] Next, once the user completes the subscription contract, they will receive a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After booting up, the devices will send their identification information (device ID, MAC address, etc.) to the server.

[0808] The devices are installed at various locations within the facility and continuously collect audio and video data using cameras and microphones. The collected data is transmitted in real time to a server, which temporarily stores it in a buffer. The data is then input into a data analysis tool every certain time frame (e.g., 5 or 10 seconds).

[0809] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying). It also uses an emotion engine to analyze audio and video data to recognize the user's emotional state. Voice tone and facial expression data are used to analyze the emotional state.

[0810] This emotional state is important auxiliary data for the anomaly detection means to determine an abnormality. If an abnormality is detected, the server generates detailed alert information. This information includes the type of abnormality, the time of occurrence, the location of occurrence, and the recognized emotional state. The generated alert information is notified to the user via one or more of email, text message, and / or dedicated in-app notification.

[0811] Furthermore, the server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to grasp the safety status of the entire facility and take necessary improvement measures.

[0812] Specific examples

[0813] For example, consider a situation where a staff member at a nursing home uses abusive or insulting language toward an elderly person. In this case, a device captures audio and video data from within the facility and sends it to a server. The server then stores the received data in a buffer and analyzes it using an AI algorithm. Abnormal audio, including abusive or insulting language, is detected. The server then uses an emotion engine to recognize the elderly person's emotional state (e.g., fear, distress) from the video data. An alert is then generated based on the abnormal behavior and the emotion recognition results, and this information is sent to the facility manager. The user (facility manager) who receives the notification can check the details, rush to the scene, and take the necessary action (e.g., warn staff, implement immediate countermeasures).

[0814] In this way, by combining the emotion engine, this system achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, thereby improving safety within facilities and ensuring the welfare of the elderly and children.

[0815] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0816] Step 1:

[0817] The user accesses the system's website, creates an account, enters the necessary information such as the facility name, address, and contact information, and submits it.

[0818] Input: User information such as facility name, address, and contact details.

[0819] Process: Sends the information entered by the user to the server.

[0820] Output: User information is saved on the server.

[0821] Step 2:

[0822] The server receives the information sent by the user and stores it in a database, which allows for the configuration and management of each facility.

[0823] Input: User-submitted information such as facility name, address, and contact details.

[0824] Processing: The server stores the information in a database.

[0825] Output: User information stored in the database.

[0826] Step 3:

[0827] Once a user completes a subscription, a set number of AI devices will be delivered to the facility, which the user must then install and connect to the network and power source.

[0828] Input: Subscription contract information, AI device.

[0829] Processing: The AI ​​device is installed on-site and connected to a network and power source.

[0830] Output: AI device installed within the facility.

[0831] Step 4:

[0832] After the device starts up, it sends its own identification information (device ID, MAC address, etc.) to the server.

[0833] Input: Device ID, MAC Address.

[0834] Process: The device starts up and sends its identification information to the server.

[0835] Output: The device identification sent to the server.

[0836] Step 5:

[0837] The devices are installed at various locations within the facility and use cameras and microphones to continuously collect audio and video data.

[0838] Input: Location audio and video data.

[0839] Processing: Capture audio and video data in real time.

[0840] Output: Collected audio and video data.

[0841] Step 6:

[0842] The server receives the collected data in real time, temporarily stores it in a buffer, and then inputs it into the data analysis means every certain time frame (e.g., 5 seconds, 10 seconds).

[0843] Input: Collected audio and video data.

[0844] Processing: Data is held in a buffer and input to a data analysis tool for each time frame.

[0845] Output: Audio and video data input to data analysis means.

[0846] Step 7:

[0847] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[0848] Input: Input audio and video data per time frame.

[0849] Processing: Analyze the data using AI algorithms to detect abnormal behavior and audio.

[0850] Output: Detected abnormal behavior and audio.

[0851] Step 8:

[0852] The server uses an emotion engine to analyze the audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[0853] Input: Input audio and video data per time frame.

[0854] Processing: Analyze the emotional state using the emotion engine.

[0855] Output: Perceived emotional state.

[0856] Step 9:

[0857] If the server detects an anomaly, it generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state. The generated alert information is then sent to the user via email, text message, or in-app notification.

[0858] Input: Detected abnormal behavior and emotional state.

[0859] Action: Generate and notify alert information.

[0860] Output: Alert information notified to the user.

[0861] Step 10:

[0862] The server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to understand the safety status of the entire facility and take necessary improvement measures.

[0863] Input: Collected audio and video data.

[0864] Processing: The data is stored for a certain period of time and the analysis results are generated as a report.

[0865] Output: The report provided to the user.

[0866] (Application example 2)

[0867] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0868] The problem to be solved by this invention is to effectively prevent abuse in nurseries and nursing homes and to increase safety within the facilities. In particular, the object is to provide a system that can detect abuse at an early stage and take prompt and appropriate action.

[0869] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means. This allows for real-time data collection from devices installed in the facility and analysis of the collected data, enabling early detection of abusive behavior and abnormal behavior and recognition of emotional states based on voice tone and facial expression data. Furthermore, when an abnormality is detected, an alert can be quickly sent via email, text message, or a dedicated app, prompting facility managers and relatives to take immediate action. Furthermore, by storing video and audio data along with the analysis results for a certain period of time, the data can be referenced later, providing evidence of problematic behavior.

[0870] "Data collection means" refers to means for collecting audio data and video data from devices installed within the facility.

[0871] The "data analysis means" is a means for analyzing collected audio data and video data and extracting necessary information.

[0872] "Abnormality detection means" is a means for detecting abusive behavior or abnormal behaviour from information extracted through data analysis.

[0873] The "emotion recognition means" is a means for analyzing voice tone and facial expression data to recognize the emotional state of a subject.

[0874] The "alert notification means" is a means for notifying relevant users of an alert when an abnormality is detected using email, text message, and app notification.

[0875] "Data storage means" refers to a means for storing collected data and analysis results for a certain period of time so that they can be referenced later.

[0876] A "server" is a central processing unit that integrates data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means, and functions as a system.

[0877] This system includes data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means to prevent abuse in facilities such as daycare centers and elderly care facilities. The server is at the core of this system, integrating and managing each device and user device.

[0878] Installing and connecting the device

[0879] A certain number of AI devices will be installed within the facility. These devices will be connected to the network and power source, and after startup, will send their identification information (device ID, MAC address, etc.) to the server.

[0880] Data collection and analysis

[0881] The device continuously collects audio and video data at the installation location and transmits this data to the server in real time. The server stores the received data in a buffer and inputs it to the data analysis means at regular intervals. The data analysis means mainly uses the following software libraries:

[0882] Keras: A machine learning library for implementing emotion engine models.

[0883] OpenCV: A computer vision library for preprocessing image data and analyzing video data.

[0884] Anomaly detection and emotion recognition

[0885] The server uses AI algorithms to analyze the collected audio and video data. The anomaly detection means detects abnormal behavior and sounds, such as violent actions, screaming, or crying. The emotion recognition means analyzes voice tone and facial expression data to recognize the subject's emotional state (e.g., fear, pain). This improves the accuracy of anomaly detection.

[0886] Alert Notifications

[0887] If an abnormality is detected, the server generates detailed alert information and notifies facility managers and relatives of the incident. The alert notification method uses the following communication methods:

[0888] Email

[0889] Text message

[0890] In-app notifications

[0891] Data storage and reporting

[0892] The server stores the collected data and analysis results for a certain period of time, allowing the data to be referenced later and providing evidence of problematic behavior.The server also periodically generates a report summarizing the analysis results and provides it to users, making it easier to understand the safety status of the entire facility.

[0893] Specific examples

[0894] For example, if a staff member at a nursing home uses abusive or insulting language towards an elderly person, the system works as follows:

[0895] 1. The device captures audio and video data from within the facility and sends it to the server.

[0896] 2. The server analyzes the received data and detects abnormal voice and behavior, including abusive or insulting language.

[0897] 3. The emotion recognition means recognizes the elderly person's emotional state (e.g., fear, distress) from voice tone and facial expression data.

[0898] 4. Based on the results of anomaly detection and emotion recognition, the server generates alert information and notifies facility managers and relatives.

[0899] 5. Users can receive notifications, check detailed information, rush to the scene, and take necessary action (such as alerting staff or implementing countermeasures).

[0900] Example prompts for generative AI models

[0901] "We will demonstrate an anomaly detection system for abusive language and insulting behavior in a nursing home. The camera collects video and audio in real time, and if the emotion engine detects abnormal behavior, it will send an alert to the administrator."

[0902] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0903] Step 1:

[0904] Device installation and initial setup:

[0905] Users install a set number of AI devices within their facilities and connect them to a network and power source. After starting up, the devices send their identification information (device ID, MAC address, etc.) to the server, which then recognizes the devices and registers them in the system.

[0906] Input: Device identification information

[0907] Output: Device registration information on the server

[0908] Specific operations: Device network connection, sending of identification information, registration process on the server

[0909] Step 2:

[0910] Data collection:

[0911] The device installed inside the terminal continuously collects audio and video data from within the facility and transmits it to a server in real time.

[0912] Input: Audio and video data

[0913] Output: Raw data sent to the server

[0914] Specific operations: audio and video capture, real-time data transmission

[0915] Step 3:

[0916] Data Analysis:

[0917] The server stores the received audio and video data in a buffer and then analyzes the data using Keras and OpenCV.

[0918] Input: Raw audio and video data

[0919] Output: Analyzed data (e.g., specific motion or voice features)

[0920] Specific operation: Data input, preprocessing, and analysis using machine learning models

[0921] Step 4:

[0922] Anomaly detection:

[0923] Based on the results of the data analysis, the server uses anomaly detection methods to detect abusive or abnormal behavior, such as identifying specific movement features or voice patterns as abnormal.

[0924] Input: Parsed data

[0925] Output: Anomaly detection results

[0926] Specific operations: feature analysis, abnormal / normal determination

[0927] Step 5:

[0928] Emotion recognition:

[0929] The server uses an emotion recognition means to recognize the emotional state of the subject from the analyzed voice tone and facial expression data.

[0930] Input: Voice tone data and facial expression data

[0931] Output: Emotion recognition results (e.g. fear, pain)

[0932] Specific actions: Analysis of voice tone, analysis of facial expression data, estimation of emotional state

[0933] Step 6:

[0934] Alerting and Notification:

[0935] The server generates alert information based on the results of anomaly detection and emotion recognition, and notifies facility managers and relatives through alert notification methods such as email, text message, and in-app notification.

[0936] Input: Anomaly detection results and emotion recognition results

[0937] Output: Alert information, notification message

[0938] Specific actions: Generate alert information, send notification messages

[0939] Step 7:

[0940] Data storage and reporting:

[0941] The server stores the collected data and analysis results for a certain period of time, and periodically generates a report summarizing the analysis results and provides it to the user. This allows the data to be referenced later, providing evidence of problematic behavior.

[0942] Input: Collected data, analysis results

[0943] Output: Saved data, scheduled reports

[0944] Specific actions: saving data, generating reports, and providing them to users

[0945] Through the above processing steps, the present system can enhance safety within the facility and ensure the welfare of the elderly and children in particular.

[0946] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0947] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0948] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0949] [Third embodiment]

[0950] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0951] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0952] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0953] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0954] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0955] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0956] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0957] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0958] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0959] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0960] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0961] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0962] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[0963] Account Creation and Initial Setup

[0964] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and then send it to the server.

[0965] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[0966] Device Registration and Connection

[0967] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[0968] After a device boots up, it automatically sends identifying information to a server, which uses this information to associate the device with a specific user account.

[0969] Data Collection and Monitoring

[0970] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[0971] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[0972] Data analysis and anomaly detection

[0973] The server uses AI algorithms to analyze the received audio and video data, with the goal of detecting any abnormal behavior or sounds that deviate from normal patterns of behavior.

[0974] If the server detects any abnormal behavior or sound, it generates an alert containing information about the abnormality, including the type of abnormality, the time of occurrence, and the location where it occurred.

[0975] Alert Notifications

[0976] The server generates an alert and sends it via email, SMS, or a dedicated app, or a combination of these.

[0977] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, enabling them to take prompt action.

[0978] Data storage and report generation

[0979] The server stores the collected audio and video data for a certain period of time, and the stored data is used for later verification and report creation.

[0980] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[0981] Specific examples

[0982] For example, let's assume that a staff member at a nursing home is verbally abusing an elderly person. In this case, the system works as follows:

[0983] 1. The device captures audio data within the facility and sends it to the server.

[0984] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[0985] 3. The server detects the abnormality and generates alert information.

[0986] 4. The server sends a notification to the facility manager via email or a dedicated app.

[0987] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[0988] In this way, the present invention provides a system that utilizes AI technology to ensure safety within facilities and enables early detection of abuse and rapid response.

[0989] The processing flow will be explained below.

[0990] Step 1:

[0991] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[0992] Step 2:

[0993] The server creates an account based on the received information, stores it in the database, and sends authentication information (user ID and password) to the user's email address.

[0994] Step 3:

[0995] The user signs up for a subscription and receives the device, which they then install in an appropriate location and connect to the network and power.

[0996] Step 4:

[0997] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[0998] Step 5:

[0999] The server verifies the identity from the device and associates it with the user's account, thereby confirming that the device belongs to a particular facility.

[1000] Step 6:

[1001] The device collects audio and video data in real time, which is then immediately sent to a server.

[1002] Step 7:

[1003] The server buffers the received data and inputs it into a data analysis means, which includes an AI algorithm.

[1004] Step 8:

[1005] AI algorithms installed on the server analyze audio and video data, including detecting abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[1006] Step 9:

[1007] If an anomaly is detected, the server generates an alert, which includes details such as the type of anomaly, the time it occurred, and the location where it occurred.

[1008] Step 10:

[1009] Based on the generated alert information, the server sends a notification to the user using the specified notification method (email, SMS, or in-app notification).

[1010] Step 11:

[1011] Users can receive alert notifications and access a dedicated app or website to check detailed information, making it easier to understand the situation on-site and enable prompt action.

[1012] Step 12:

[1013] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[1014] Step 13:

[1015] The server periodically generates a report summarizing the analysis results, allowing users to understand the safety status of the entire facility and take any necessary improvement measures.

[1016] In this way, the system monitors the situation within the facility in real time, promptly notifies the user when an abnormality occurs, and ensures the safety of the facility by allowing the user to take appropriate action. This series of processing steps enables early detection of abuse and rapid response.

[1017] Example 1

[1018] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1019] Abuse in daycare centers and elderly care facilities is a major social problem, and there is a need for a rapid and accurate monitoring system to prevent it from happening. These facilities also face serious labor shortages, necessitating a highly automated system. Current monitoring systems have difficulty detecting abnormal behavior and sounds in real time and responding quickly. Therefore, there is a need for a system that utilizes AI technology to efficiently detect abnormalities while ensuring safety within the facility.

[1020] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1021] In this invention, the server includes a means for creating an account, a means for transmitting device identification information to the server and linking it to a user account, a means for collecting audio and video data, a means for analyzing the received data by time frame, a means for detecting abnormal behavior or abnormal sounds using an AI algorithm, a means for generating an alert containing abnormal information, a means for notifying the generated alert via email, text message, or app, a means for storing the audio and video data for a certain period of time, and a means for generating a report of the results of data analysis. This enables early detection of possible acts of abuse occurring within the facility and rapid response.

[1022] The "account creation means" refers to the means by which a user accesses the system's website and creates an account by entering necessary information such as the facility name, location, and contact information.

[1023] The "means of transmitting device identification information to a server and associating it with a user account" refers to a means in which the device transmits identification information to a server after startup, and the server associates the device with a specific user account based on that information.

[1024] "Means for collecting audio and video data" means means for continuously capturing audio and video data by devices installed within the facility.

[1025] The "means for analyzing received data for each time frame" refers to a means for storing data received by the server in a buffer and analyzing the data for each fixed time frame.

[1026] "Means for detecting abnormal behavior and sounds using AI algorithms" refers to a means by which the server uses AI technology to identify abnormal behavior and sounds from audio and video data.

[1027] The "means for generating an alert including abnormality information" refers to a means for generating an alert including information such as the type of abnormality, the time of occurrence, and the location of occurrence when the server detects abnormal behavior or abnormal sound.

[1028] "Means of notifying users of generated alerts via email, text message, or app" refers to means of notifying users of server-generated alerts via email, text message, or a dedicated app.

[1029] "Means for storing audio and video data for a certain period of time" refers to the means by which the server stores collected audio and video data for a certain period of time for the purpose of later verification and report creation.

[1030] The "means for generating a report of the data analysis results" refers to a means by which the server periodically generates a report summarizing the results of the data analysis and anomaly detection, and provides it to the user.

[1031] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[1032] Account Creation and Initial Setup

[1033] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and send it to the server. The server stores the user's account information in a database, allowing it to configure and manage each facility. The database uses a general-purpose database management system such as MySQL or PostgreSQL.

[1034] Device Registration and Connection

[1035] When a user signs up for a subscription, they are sent a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After the devices start up, they automatically send identification information to a server. The server uses this information to associate the devices with specific user accounts. Devices are identified using MAC addresses or unique IDs.

[1036] Data Collection and Monitoring

[1037] The device continuously collects audio and video data at the installation site. The collected data is sent in real time to a server, which stores the data in a buffer and inputs it into a data analysis tool at regular intervals. Data is sent using secure HTTP or MQTT protocols.

[1038] Data analysis and anomaly detection

[1039] The server uses a generative AI model to analyze the received audio and video data. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to detect abnormal behavior and abnormal audio. Abnormal behavior is detected as violent acts, and abnormal audio is detected as words such as "help me."

[1040] When the server detects an anomaly, it generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time of occurrence, and the location of the anomaly. The content of the alert contains enough information for the user to take prompt action.

[1041] Alert Notifications

[1042] The server generates alerts and notifies users via email, text message, or a dedicated app, using external services such as SendGrid and Twilio APIs to send notifications.

[1043] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, allowing them to quickly grasp the situation on-site. Based on this information, they can take appropriate action.

[1044] Data storage and report generation

[1045] The server stores the collected audio and video data for a certain period of time. The data is then stored in cloud storage such as Amazon S3 or Google Cloud Storage. The stored data can be used for later verification and report creation.

[1046] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user in PDF or Excel format, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[1047] Specific examples

[1048] For example, consider a case where a staff member at a nursing home is verbally abusing an elderly person.

[1049] 1. The device captures audio data from within the facility and sends it to a server. Specifically, the device's microphone picks up abusive language such as "idiot" and "die."

[1050] 2. The server analyzes the received audio data and detects abusive language.

[1051] 3. The server detects the abnormality and generates detailed alert information.

[1052] 4. The server sends an alert to the facility manager. Specifically, a push notification is sent to the manager's smartphone.

[1053] 5. The user (facility manager) checks the notification, rushes to the scene, records and saves the situation, and warns the staff.

[1054] In this way, the system of the present invention makes full use of AI technology to ensure safety within the facility and enables early detection of abuse and rapid response.

[1055] Prompt Sentence Examples

[1056] Here is an example of starting the system's processing by inputting the following prompt sentence into the generative AI model:

[1057] Set up an AI device at nursing home ABC and start real-time monitoring. Set it up to send an alert to the administrator as soon as it detects abusive language or abnormal behavior by staff.

[1058] This prompt is used as an example to demonstrate system configuration and processing.

[1059] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1060] Step 1:

[1061] Users access the system's website and create an account.

[1062] Input: Facility name, address, contact information, etc.

[1063] Process: The user enters this information into the registration form and clicks the "Submit" button.

[1064] Output: The data is sent to the server.

[1065] Step 2:

[1066] The server stores user account information in a database.

[1067] Input: Information such as facility name, address, and contact details submitted by the user.

[1068] Processing: Information is stored using database software (e.g., MySQL).

[1069] Output: The user account is registered in the database and a notification is sent to the user confirming account creation.

[1070] Step 3:

[1071] Once a user signs up for a subscription, they will be sent an AI device.

[1072] Input: Subscription contract information.

[1073] Processing: After the contract is confirmed, a specified number of AI devices will be sent to the user.

[1074] Output: The AI ​​device arrives at the user's facility.

[1075] Step 4:

[1076] Users install AI devices within their facilities and connect them to a network and power source.

[1077] Inputs: AI device, network information, power outlet.

[1078] Process: Place the device in its designated location, plug it into a power outlet, and configure Wi-Fi.

[1079] Output: The AI ​​device starts up.

[1080] Step 5:

[1081] After the device starts up, it automatically sends its identification information to the server.

[1082] Input: Device identification information (MAC address, unique ID, etc.).

[1083] Processing: The device establishes a network connection and sends its identification information to the server.

[1084] Output: The identification information is sent to the server.

[1085] Step 6:

[1086] The server associates the device identity with the user account.

[1087] Input: Device identification information, user account information.

[1088] Action: Updates the database to associate the device information with the user account.

[1089] Output: The device is associated with the user account.

[1090] Step 7:

[1091] The device continuously collects audio and video data from the location where it is installed.

[1092] Input: Location audio and video data.

[1093] Processing: The device's built-in camera and microphone capture audio and video.

[1094] Output: Collected data is temporarily stored on the device.

[1095] Step 8:

[1096] The device sends the collected data to a server in real time.

[1097] Input: Collected audio and video data.

[1098] Processing: Data is sent using secure HTTP or MQTT protocols.

[1099] Output: The data is sent to the server.

[1100] Step 9:

[1101] The server stores the received data in a buffer.

[1102] Input: Audio and video data sent from the device.

[1103] Processing: Storing data in RAM and temporary storage.

[1104] Output: The data is held in a buffer.

[1105] Step 10:

[1106] The server uses AI algorithms to analyze the data held in the buffer.

[1107] Input: Buffered audio and video data.

[1108] Processing: Detect anomalous behavior and sounds using TensorFlow and PyTorch.

[1109] Output: The analysis results are obtained.

[1110] Step 11:

[1111] The server detects abnormal behavior or sounds and generates alert information.

[1112] Input: Analysis results of the AI ​​model.

[1113] Processing: Generate an alert based on information such as the type of anomaly, the time of occurrence, and the location of the anomaly.

[1114] Output: Alert information is generated.

[1115] Step 12:

[1116] The server generates alerts and notifies users via email, text message, or a dedicated app.

[1117] Input: Alert information.

[1118] Processing: Send notifications using SendGrid or Twilio APIs.

[1119] Output: The alert is notified to the user.

[1120] Step 13:

[1121] The user checks the received alert notification and obtains detailed information.

[1122] Input: Alert notification.

[1123] Action: Access the dedicated app or website to check detailed information.

[1124] Output: Information is obtained to understand the situation on site.

[1125] Step 14:

[1126] The server stores the collected audio and video data for a certain period of time.

[1127] Input: Collected audio and video data.

[1128] Processing: Store data in Amazon S3 or Google Cloud Storage.

[1129] Output: Data is saved to cloud storage.

[1130] Step 15:

[1131] The server periodically generates a report summarizing the results of the data analysis.

[1132] Input: Analysis results, past data.

[1133] Processing: Generates reports in PDF and Excel formats and provides them to the user.

[1134] Output: A report is provided to understand the status of the facility.

[1135] (Application example 1)

[1136] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1137] In modern society, early detection and prevention of abuse in facilities such as daycare centers and elderly care facilities is a major challenge. However, conventional monitoring systems have limitations in their ability to analyze massive amounts of data in real time and quickly detect abnormal behavior. This increases the time between the occurrence of an abnormality and a response, risking the spread of damage. In addition, multiple notification methods and devices are often not properly coordinated, hindering administrators' ability to respond quickly.

[1138] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1139] In this invention, the server includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, a cloud storage means, an abnormal behavior analysis means using a generative AI model, and a push notification means. This allows for real-time analysis of audio and video data collected within the facility, enabling the immediate detection of abnormal behavior or abnormal sounds. Furthermore, by using multiple notification methods, managers can quickly identify abnormalities and take necessary measures.

[1140] "Data collection means" refers to a system that uses devices and sensors to collect data such as audio and video in real time.

[1141] "Data analysis means" refers to a system that includes software and algorithms for processing collected data and detecting abnormal behavior or sounds.

[1142] An "anomaly detection means" is a system that includes functions and algorithms for automatically detecting abnormal behavior or abnormal sounds from collected and analyzed data.

[1143] An "alert notification method" is a communication method such as email, text message, in-app notification, or push notification that is used to warn an administrator when an abnormality is detected.

[1144] "Data storage means" refers to a storage system that stores collected audio and video data for a certain period of time so that it can be verified or referenced later.

[1145] "Cloud storage means" refers to a remote storage service for storing and managing data over the Internet, enabling the storage of large amounts of data without being restricted by physical devices.

[1146] "Means for analyzing abnormal behavior using generative AI models" refers to a function that uses artificial intelligence (AI) models to detect and identify unusual behavior patterns when analyzing collected data.

[1147] "Push notification" is a technology for sending instant notifications to smartphones and other devices, providing users with important information in real time.

[1148] This invention is a system for preventing abuse in facilities such as nurseries and nursing homes. The system is made up of the following main components, and its specific implementation method is described below.

[1149] 1. Data Collection Methods

[1150] The data collection methods include devices and sensors that collect audio and video data in real time within the facility. Specifically, microphones and cameras are used. These devices are installed in various locations within the facility, continuously collect data, and transmit it to a server via a network.

[1151] 2. Data analysis methods

[1152] The data analysis means includes software and algorithms for processing the collected data and detecting abnormal behavior and sounds. Specific software used is a Python program and an external analysis API. This allows data to be analyzed in real time and abnormal behavior and sounds to be detected.

[1153] 3. Anomaly detection methods

[1154] The anomaly detection method includes functions and algorithms that automatically detect abnormal behavior and sounds from collected and analyzed data. Specifically, a generative AI model is used. This AI model learns behaviors that deviate from normal behavior patterns and identifies abnormalities.

[1155] 4. Alert notification methods

[1156] Alert notification means include communication means such as email, text message, in-app notification, and push notification to alert administrators when an abnormality is detected, allowing administrators to immediately identify the abnormality and take prompt action.

[1157] 5. Data storage means

[1158] The data storage means includes a storage system that stores collected audio and video data for a certain period of time so that it can be verified and referenced later. Specifically, Firebase storage is used. This allows past data to be stored and verified when necessary.

[1159] 6. Cloud Storage Solutions

[1160] Cloud storage services are remote storage services for storing and managing data over the Internet, allowing for the storage of large amounts of data without being restricted by physical devices. Specifically, Firebase Storage is one such service.

[1161] 7. Methods for analyzing abnormal behavior using generative AI models

[1162] The abnormal behavior analysis method using a generative AI model includes a function that uses an artificial intelligence (AI) model to detect and identify abnormal behavior patterns when analyzing collected data, thereby enabling the detection of abnormal behavior with high accuracy.

[1163] 8. Push Notification Methods

[1164] Push notification is a technology for instantly sending notifications to smartphones and other devices, including a means to provide users with important information in real time, allowing administrators to quickly identify and respond to abnormalities.

[1165] Specific examples

[1166] For example, if a staff member in a nursing home is verbally abusing an elderly person, the system will function as follows:

[1167] 1. The device captures audio data within the facility and sends it to the server.

[1168] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[1169] 3. The server detects the abnormality and generates alert information.

[1170] 4. The server sends a notification to the facility manager via email or a dedicated app.

[1171] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[1172] Prompt Sentence Examples

[1173] "Please tell me how to detect abnormal sounds while working at a nursing home."

[1174] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1175] Step 1:

[1176] The user registers an account with the system along with facility information. The user enters facility information (facility name, address, contact information, etc.) and sends it to the server. This information is saved in a database by the server and linked to the user account. The input is facility information, and the output is the user profile saved in the database.

[1177] Step 2:

[1178] A user installs a device in a facility and connects it to the network and power. When the device starts up, it automatically sends its identification information to the server. The server receives this identification information and associates the device with a specific user account. The input is the device's identification information, and the output is the device information stored in a database.

[1179] Step 3:

[1180] The device collects audio and video data in the facility in real time and sends it to the server. The server stores the received data in a buffer and inputs it into the data analysis means at regular time intervals. The input is the audio and video data transmitted in real time, and the output is the data temporarily stored in the buffer.

[1181] Step 4:

[1182] The server analyzes the collected audio and video data using a generative AI model. The purpose of the analysis is to detect abnormal behavior or sounds that deviate from normal behavior patterns. The input is the audio and video data stored in the buffer, and the output is the analysis results, which are information on abnormal behavior or sounds.

[1183] Step 5:

[1184] If an anomaly is detected, the server generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time it occurred, and the location where it occurred. The input is the anomaly detection information as an analysis result, and the output is the alert information.

[1185] Step 6:

[1186] The server generates alerts and sends them to administrators via email, text message, in-app notification, or push notification. The input is the alert information, and the output is the notification received by the administrator.

[1187] Step 7:

[1188] The administrator receives a notification and accesses a dedicated app or website to check detailed information. The input is a notification of an abnormality detection, and the output is the result of checking the detailed information. Based on this information, the administrator can rush to the site and take appropriate measures.

[1189] Step 8:

[1190] The server stores the collected audio and video data for a certain period of time and uses it for later verification and report creation. The input is the collected audio and video data, and the output is the data saved in storage. This makes it possible to reference past data as needed.

[1191] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1192] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[1193] Account Creation and Initial Setup

[1194] First, users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[1195] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[1196] Device Registration and Connection

[1197] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[1198] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[1199] Data Collection and Monitoring

[1200] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[1201] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[1202] Data analysis and anomaly detection

[1203] The server uses AI algorithms to analyze the collected audio and video data.

[1204] AI algorithms detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[1205] Emotion recognition by emotion engine

[1206] The emotion engine installed on the server analyzes the collected audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[1207] The server uses the emotional state recognized by the emotion engine as additional input when the anomaly detection means determines an anomaly.

[1208] Alert Notifications

[1209] If an anomaly is detected, the server generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state.

[1210] The server notifies the user of the alert via email, SMS, or a dedicated in-app notification, or a combination of these.

[1211] Data storage and report generation

[1212] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[1213] The server periodically generates a report summarizing the analysis results and provides it to the user, who can use this report to understand the safety status of the entire facility and take any necessary improvement measures.

[1214] Specific examples

[1215] For example, let's assume that a staff member at a nursing home is using abusive or insulting language towards an elderly person. In this case, the system works as follows:

[1216] 1. The device captures audio and video data from within the facility and sends it to the server.

[1217] 2. The server analyzes the received data and detects abnormal voices, including abusive or insulting language.

[1218] 3. The server uses an emotion engine to recognize the emotional state of the elderly person (e.g., fear, pain) from the video data.

[1219] 4. The server generates alert information based on the abnormality and emotion recognition results and notifies the facility manager of this information.

[1220] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (such as warning staff or implementing countermeasures).

[1221] In this way, by combining an emotion engine, the present invention provides a system that achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, which contributes to improving safety within facilities and is particularly effective in ensuring the welfare of the elderly and children.

[1222] The processing flow will be explained below.

[1223] Step 1:

[1224] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[1225] Step 2:

[1226] The server receives the user's input information, saves the account information in a database, and generates authentication information (user ID and password) and sends it to the user's email address.

[1227] Step 3:

[1228] Users sign a subscription contract and receive a specified number of AI devices, which they then install in appropriate locations within their facility and connect to a network and power source.

[1229] Step 4:

[1230] When the device starts up, it sends its identification information (device ID, MAC address, etc.) to the server.

[1231] Step 5:

[1232] The server receives the identification information and associates it with a user account, thereby establishing that the device belongs to a particular facility.

[1233] Step 6:

[1234] The device continuously collects audio and video data within the facility and transmits this data to a server in real time.

[1235] Step 7:

[1236] The server stores the received data in a buffer and performs analysis using a data analysis method. An AI algorithm is used.

[1237] Step 8:

[1238] The server's AI algorithm analyzes audio and video data to detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[1239] Step 9:

[1240] An emotion engine installed on the server analyzes audio and video data and recognizes the user's emotional state (e.g., fear, pain, etc.).

[1241] Step 10:

[1242] The server uses the emotional state as additional data to improve the accuracy of anomaly detection. If the emotional state is associated with an anomaly, the server will determine the anomaly with higher sensitivity.

[1243] Step 11:

[1244] As a final analysis result, the server generates alert information if an anomaly is detected, which includes details of abnormal behavior, abnormal sound, recognized emotional state, time of occurrence, and location of occurrence.

[1245] Step 12:

[1246] Based on the alert information, the server sends a notification to the user using one or more of email, SMS, and / or a notification within a dedicated app.

[1247] Step 13:

[1248] Users receive an alert notification and can access a dedicated app or website to check detailed information, allowing them to understand the situation on-site in detail and take prompt action.

[1249] Step 14:

[1250] The server stores the collected data for a certain period of time, which allows for future reference and verification of the data.

[1251] Step 15:

[1252] The server periodically generates a report summarizing the analysis results and provides it to the user, helping them understand the safety status of the entire facility and take any necessary improvement measures.

[1253] In this way, through a series of processing steps, the system can quickly detect acts of abuse in daycare centers and elderly care facilities and take immediate action.By combining it with an emotion engine, it becomes possible to detect anomalies with high accuracy, taking into account emotional states.

[1254] Example 2

[1255] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1256] To prevent abuse in daycare centers and elderly care facilities, it is necessary to accurately detect abnormal behavior and sounds within the facility and take prompt action. However, conventional monitoring systems are limited to detecting abnormal behavior and sounds, and do not take emotional states into account when detecting abnormalities. Therefore, a system that can more reliably ensure safety within the facility is needed.

[1257] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1258] In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means, which not only detects abnormal behavior and abnormal voice from audio and video data but also recognizes emotional states, enabling more accurate detection of abnormalities and quicker response.

[1259] "Data collection means" means a device or system for continuously collecting audio and video data within a facility.

[1260] "Data analysis means" refers to software or algorithms that analyze collected audio and video data and detect abnormal behavior or abnormal sounds.

[1261] The "anomaly detection means" is an algorithm or system for identifying abnormal behavior or abnormal sounds from the data analyzed by the data analysis means and detecting abnormalities.

[1262] "Emotion recognition means" refers to software or algorithms that analyze and recognize emotional states based on collected audio and video data.

[1263] An "alert notification means" is a device or system for notifying a user of information about a detected abnormality, and includes means such as email, text message, and in-app notification.

[1264] "Data storage means" refers to a device or system that stores collected and analyzed data for a certain period of time and makes it available for later reference.

[1265] "Server" means a central processing unit for collecting, analyzing, storing, and notifying data.

[1266] An "AI algorithm" is a computational method that uses machine learning and deep learning to analyze audio and video data and detect anomalies.

[1267] A "time frame" is a unit of time for processing data by dividing it into regular time intervals.

[1268] A "buffer" is a storage area for temporarily storing data.

[1269] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an emotion recognition means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[1270] To start using the system, users first access the system's website and create an account. They enter the required information (facility name, address, contact information, etc.) and submit it. The server then receives the entered information and stores it in a database. This allows for the configuration and management of each facility.

[1271] Next, once the user completes the subscription contract, they will receive a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After booting up, the devices will send their identification information (device ID, MAC address, etc.) to the server.

[1272] The devices are installed at various locations within the facility and continuously collect audio and video data using cameras and microphones. The collected data is transmitted in real time to a server, which temporarily stores it in a buffer. The data is then input into a data analysis tool every certain time frame (e.g., 5 or 10 seconds).

[1273] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying). It also uses an emotion engine to analyze audio and video data to recognize the user's emotional state. Voice tone and facial expression data are used to analyze the emotional state.

[1274] This emotional state is important auxiliary data for the anomaly detection means to determine an abnormality. If an abnormality is detected, the server generates detailed alert information. This information includes the type of abnormality, the time of occurrence, the location of occurrence, and the recognized emotional state. The generated alert information is notified to the user via one or more of email, text message, and / or dedicated in-app notification.

[1275] Furthermore, the server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to grasp the safety status of the entire facility and take necessary improvement measures.

[1276] Specific examples

[1277] For example, consider a situation where a staff member at a nursing home uses abusive or insulting language toward an elderly person. In this case, a device captures audio and video data from within the facility and sends it to a server. The server then stores the received data in a buffer and analyzes it using an AI algorithm. Abnormal audio, including abusive or insulting language, is detected. The server then uses an emotion engine to recognize the elderly person's emotional state (e.g., fear, distress) from the video data. An alert is then generated based on the abnormal behavior and the emotion recognition results, and this information is sent to the facility manager. The user (facility manager) who receives the notification can check the details, rush to the scene, and take the necessary action (e.g., warn staff, implement immediate countermeasures).

[1278] In this way, by combining the emotion engine, this system achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, thereby improving safety within facilities and ensuring the welfare of the elderly and children.

[1279] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1280] Step 1:

[1281] The user accesses the system's website, creates an account, enters the necessary information such as the facility name, address, and contact information, and submits it.

[1282] Input: User information such as facility name, address, and contact details.

[1283] Process: Sends the information entered by the user to the server.

[1284] Output: User information is saved on the server.

[1285] Step 2:

[1286] The server receives the information sent by the user and stores it in a database, which allows for the configuration and management of each facility.

[1287] Input: User-submitted information such as facility name, address, and contact details.

[1288] Processing: The server stores the information in a database.

[1289] Output: User information stored in the database.

[1290] Step 3:

[1291] Once a user completes a subscription, a set number of AI devices will be delivered to the facility, which the user must then install and connect to the network and power source.

[1292] Input: Subscription contract information, AI device.

[1293] Processing: The AI ​​device is installed on-site and connected to a network and power source.

[1294] Output: AI device installed within the facility.

[1295] Step 4:

[1296] After the device starts up, it sends its own identification information (device ID, MAC address, etc.) to the server.

[1297] Input: Device ID, MAC Address.

[1298] Process: The device starts up and sends its identification information to the server.

[1299] Output: The device identification sent to the server.

[1300] Step 5:

[1301] The devices are installed at various locations within the facility and use cameras and microphones to continuously collect audio and video data.

[1302] Input: Location audio and video data.

[1303] Processing: Capture audio and video data in real time.

[1304] Output: Collected audio and video data.

[1305] Step 6:

[1306] The server receives the collected data in real time, temporarily stores it in a buffer, and then inputs it into the data analysis means every certain time frame (e.g., 5 seconds, 10 seconds).

[1307] Input: Collected audio and video data.

[1308] Processing: Data is held in a buffer and input to a data analysis tool for each time frame.

[1309] Output: Audio and video data input to data analysis means.

[1310] Step 7:

[1311] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[1312] Input: Input audio and video data per time frame.

[1313] Processing: Analyze the data using AI algorithms to detect abnormal behavior and audio.

[1314] Output: Detected abnormal behavior and audio.

[1315] Step 8:

[1316] The server uses an emotion engine to analyze the audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[1317] Input: Input audio and video data per time frame.

[1318] Processing: Analyze the emotional state using the emotion engine.

[1319] Output: Perceived emotional state.

[1320] Step 9:

[1321] If the server detects an anomaly, it generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state. The generated alert information is then sent to the user via email, text message, or in-app notification.

[1322] Input: Detected abnormal behavior and emotional state.

[1323] Action: Generate and notify alert information.

[1324] Output: Alert information notified to the user.

[1325] Step 10:

[1326] The server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to understand the safety status of the entire facility and take necessary improvement measures.

[1327] Input: Collected audio and video data.

[1328] Processing: The data is stored for a certain period of time and the analysis results are generated as a report.

[1329] Output: The report provided to the user.

[1330] (Application example 2)

[1331] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1332] The problem to be solved by this invention is to effectively prevent abuse in nurseries and nursing homes and to increase safety within the facilities. In particular, the object is to provide a system that can detect abuse at an early stage and take prompt and appropriate action.

[1333] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means. This allows for real-time data collection from devices installed in the facility and analysis of the collected data, enabling early detection of abusive behavior and abnormal behavior and recognition of emotional states based on voice tone and facial expression data. Furthermore, when an abnormality is detected, an alert can be quickly sent via email, text message, or a dedicated app, prompting facility managers and relatives to take immediate action. Furthermore, by storing video and audio data along with the analysis results for a certain period of time, the data can be referenced later, providing evidence of problematic behavior.

[1334] "Data collection means" refers to means for collecting audio data and video data from devices installed within the facility.

[1335] The "data analysis means" is a means for analyzing collected audio data and video data and extracting necessary information.

[1336] "Abnormality detection means" is a means for detecting abusive behavior or abnormal behaviour from information extracted through data analysis.

[1337] The "emotion recognition means" is a means for analyzing voice tone and facial expression data to recognize the emotional state of a subject.

[1338] The "alert notification means" is a means for notifying relevant users of an alert when an abnormality is detected using email, text message, and app notification.

[1339] "Data storage means" refers to a means for storing collected data and analysis results for a certain period of time so that they can be referenced later.

[1340] A "server" is a central processing unit that integrates data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means, and functions as a system.

[1341] This system includes data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means to prevent abuse in facilities such as daycare centers and elderly care facilities. The server is at the core of this system, integrating and managing each device and user device.

[1342] Installing and connecting the device

[1343] A certain number of AI devices will be installed within the facility. These devices will be connected to the network and power source, and after startup, will send their identification information (device ID, MAC address, etc.) to the server.

[1344] Data collection and analysis

[1345] The device continuously collects audio and video data at the installation location and transmits this data to the server in real time. The server stores the received data in a buffer and inputs it to the data analysis means at regular intervals. The data analysis means mainly uses the following software libraries:

[1346] Keras: A machine learning library for implementing emotion engine models.

[1347] OpenCV: A computer vision library for preprocessing image data and analyzing video data.

[1348] Anomaly detection and emotion recognition

[1349] The server uses AI algorithms to analyze the collected audio and video data. The anomaly detection means detects abnormal behavior and sounds, such as violent actions, screaming, or crying. The emotion recognition means analyzes voice tone and facial expression data to recognize the subject's emotional state (e.g., fear, pain). This improves the accuracy of anomaly detection.

[1350] Alert Notifications

[1351] If an abnormality is detected, the server generates detailed alert information and notifies facility managers and relatives of the incident. The alert notification method uses the following communication methods:

[1352] Email

[1353] Text message

[1354] In-app notifications

[1355] Data storage and reporting

[1356] The server stores the collected data and analysis results for a certain period of time, allowing the data to be referenced later and providing evidence of problematic behavior.The server also periodically generates a report summarizing the analysis results and provides it to users, making it easier to understand the safety status of the entire facility.

[1357] Specific examples

[1358] For example, if a staff member at a nursing home uses abusive or insulting language towards an elderly person, the system works as follows:

[1359] 1. The device captures audio and video data from within the facility and sends it to the server.

[1360] 2. The server analyzes the received data and detects abnormal voice and behavior, including abusive or insulting language.

[1361] 3. The emotion recognition means recognizes the elderly person's emotional state (e.g., fear, distress) from voice tone and facial expression data.

[1362] 4. Based on the results of anomaly detection and emotion recognition, the server generates alert information and notifies facility managers and relatives.

[1363] 5. Users can receive notifications, check detailed information, rush to the scene, and take necessary action (such as alerting staff or implementing countermeasures).

[1364] Example prompts for generative AI models

[1365] "We will demonstrate an anomaly detection system for abusive language and insulting behavior in a nursing home. The camera collects video and audio in real time, and if the emotion engine detects abnormal behavior, it will send an alert to the administrator."

[1366] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1367] Step 1:

[1368] Device installation and initial setup:

[1369] Users install a set number of AI devices within their facilities and connect them to a network and power source. After starting up, the devices send their identification information (device ID, MAC address, etc.) to the server, which then recognizes the devices and registers them in the system.

[1370] Input: Device identification information

[1371] Output: Device registration information on the server

[1372] Specific operations: Device network connection, sending of identification information, registration process on the server

[1373] Step 2:

[1374] Data collection:

[1375] The device installed inside the terminal continuously collects audio and video data from within the facility and transmits it to a server in real time.

[1376] Input: Audio and video data

[1377] Output: Raw data sent to the server

[1378] Specific operations: audio and video capture, real-time data transmission

[1379] Step 3:

[1380] Data Analysis:

[1381] The server stores the received audio and video data in a buffer and then analyzes the data using Keras and OpenCV.

[1382] Input: Raw audio and video data

[1383] Output: Analyzed data (e.g., specific motion or voice features)

[1384] Specific operation: Data input, preprocessing, and analysis using machine learning models

[1385] Step 4:

[1386] Anomaly detection:

[1387] Based on the results of the data analysis, the server uses anomaly detection methods to detect abusive or abnormal behavior, such as identifying specific movement features or voice patterns as abnormal.

[1388] Input: Parsed data

[1389] Output: Anomaly detection results

[1390] Specific operations: feature analysis, abnormal / normal determination

[1391] Step 5:

[1392] Emotion recognition:

[1393] The server uses an emotion recognition means to recognize the emotional state of the subject from the analyzed voice tone and facial expression data.

[1394] Input: Voice tone data and facial expression data

[1395] Output: Emotion recognition results (e.g. fear, pain)

[1396] Specific actions: Analysis of voice tone, analysis of facial expression data, estimation of emotional state

[1397] Step 6:

[1398] Alerting and Notification:

[1399] The server generates alert information based on the results of anomaly detection and emotion recognition, and notifies facility managers and relatives through alert notification methods such as email, text message, and in-app notification.

[1400] Input: Anomaly detection results and emotion recognition results

[1401] Output: Alert information, notification message

[1402] Specific actions: Generate alert information, send notification messages

[1403] Step 7:

[1404] Data storage and reporting:

[1405] The server stores the collected data and analysis results for a certain period of time, and periodically generates a report summarizing the analysis results and provides it to the user. This allows the data to be referenced later, providing evidence of problematic behavior.

[1406] Input: Collected data, analysis results

[1407] Output: Saved data, scheduled reports

[1408] Specific actions: saving data, generating reports, and providing them to users

[1409] Through the above processing steps, the present system can enhance safety within the facility and ensure the welfare of the elderly and children in particular.

[1410] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1411] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1412] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1413] [Fourth embodiment]

[1414] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1415] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1416] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1417] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1418] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1419] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1420] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1421] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1422] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1423] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1424] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1425] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1426] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1427] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[1428] Account Creation and Initial Setup

[1429] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and then send it to the server.

[1430] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[1431] Device Registration and Connection

[1432] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[1433] After a device boots up, it automatically sends identifying information to a server, which uses this information to associate the device with a specific user account.

[1434] Data Collection and Monitoring

[1435] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[1436] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[1437] Data analysis and anomaly detection

[1438] The server uses AI algorithms to analyze the received audio and video data, with the goal of detecting any abnormal behavior or sounds that deviate from normal patterns of behavior.

[1439] If the server detects any abnormal behavior or sound, it generates an alert containing information about the abnormality, including the type of abnormality, the time of occurrence, and the location where it occurred.

[1440] Alert Notifications

[1441] The server generates an alert and sends it via email, SMS, or a dedicated app, or a combination of these.

[1442] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, enabling them to take prompt action.

[1443] Data storage and report generation

[1444] The server stores the collected audio and video data for a certain period of time, and the stored data is used for later verification and report creation.

[1445] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[1446] Specific examples

[1447] For example, let's assume that a staff member at a nursing home is verbally abusing an elderly person. In this case, the system works as follows:

[1448] 1. The device captures audio data within the facility and sends it to the server.

[1449] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[1450] 3. The server detects the abnormality and generates alert information.

[1451] 4. The server sends a notification to the facility manager via email or a dedicated app.

[1452] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[1453] In this way, the present invention provides a system that utilizes AI technology to ensure safety within facilities and enables early detection of abuse and rapid response.

[1454] The processing flow will be explained below.

[1455] Step 1:

[1456] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[1457] Step 2:

[1458] The server creates an account based on the received information, stores it in the database, and sends authentication information (user ID and password) to the user's email address.

[1459] Step 3:

[1460] The user signs up for a subscription and receives the device, which they then install in an appropriate location and connect to the network and power.

[1461] Step 4:

[1462] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[1463] Step 5:

[1464] The server verifies the identity from the device and associates it with the user's account, thereby confirming that the device belongs to a particular facility.

[1465] Step 6:

[1466] The device collects audio and video data in real time, which is then immediately sent to a server.

[1467] Step 7:

[1468] The server buffers the received data and inputs it into a data analysis means, which includes an AI algorithm.

[1469] Step 8:

[1470] AI algorithms installed on the server analyze audio and video data, including detecting abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[1471] Step 9:

[1472] If an anomaly is detected, the server generates an alert, which includes details such as the type of anomaly, the time it occurred, and the location where it occurred.

[1473] Step 10:

[1474] Based on the generated alert information, the server sends a notification to the user using the specified notification method (email, SMS, or in-app notification).

[1475] Step 11:

[1476] Users can receive alert notifications and access a dedicated app or website to check detailed information, making it easier to understand the situation on-site and enable prompt action.

[1477] Step 12:

[1478] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[1479] Step 13:

[1480] The server periodically generates a report summarizing the analysis results, allowing users to understand the safety status of the entire facility and take any necessary improvement measures.

[1481] In this way, the system monitors the situation within the facility in real time, promptly notifies the user when an abnormality occurs, and ensures the safety of the facility by allowing the user to take appropriate action. This series of processing steps enables early detection of abuse and rapid response.

[1482] Example 1

[1483] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1484] Abuse in daycare centers and elderly care facilities is a major social problem, and there is a need for a rapid and accurate monitoring system to prevent it from happening. These facilities also face serious labor shortages, necessitating a highly automated system. Current monitoring systems have difficulty detecting abnormal behavior and sounds in real time and responding quickly. Therefore, there is a need for a system that utilizes AI technology to efficiently detect abnormalities while ensuring safety within the facility.

[1485] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1486] In this invention, the server includes a means for creating an account, a means for transmitting device identification information to the server and linking it to a user account, a means for collecting audio and video data, a means for analyzing the received data by time frame, a means for detecting abnormal behavior or abnormal sounds using an AI algorithm, a means for generating an alert containing abnormal information, a means for notifying the generated alert via email, text message, or app, a means for storing the audio and video data for a certain period of time, and a means for generating a report of the results of data analysis. This enables early detection of possible acts of abuse occurring within the facility and rapid response.

[1487] The "account creation means" refers to the means by which a user accesses the system's website and creates an account by entering necessary information such as the facility name, location, and contact information.

[1488] The "means of transmitting device identification information to a server and associating it with a user account" refers to a means in which the device transmits identification information to a server after startup, and the server associates the device with a specific user account based on that information.

[1489] "Means for collecting audio and video data" means means for continuously capturing audio and video data by devices installed within the facility.

[1490] The "means for analyzing received data for each time frame" refers to a means for storing data received by the server in a buffer and analyzing the data for each fixed time frame.

[1491] "Means for detecting abnormal behavior and sounds using AI algorithms" refers to a means by which the server uses AI technology to identify abnormal behavior and sounds from audio and video data.

[1492] The "means for generating an alert including abnormality information" refers to a means for generating an alert including information such as the type of abnormality, the time of occurrence, and the location of occurrence when the server detects abnormal behavior or abnormal sound.

[1493] "Means of notifying users of generated alerts via email, text message, or app" refers to means of notifying users of server-generated alerts via email, text message, or a dedicated app.

[1494] "Means for storing audio and video data for a certain period of time" refers to the means by which the server stores collected audio and video data for a certain period of time for the purpose of later verification and report creation.

[1495] The "means for generating a report of the data analysis results" refers to a means by which the server periodically generates a report summarizing the results of the data analysis and anomaly detection, and provides it to the user.

[1496] This invention relates to a system that uses AI devices to prevent abuse in daycare centers and elderly care facilities. The system includes data collection means, data analysis means, anomaly detection means, alert notification means, and data storage means, thereby ensuring safety within the facility.

[1497] Account Creation and Initial Setup

[1498] First, users access the system's website and create an account. They enter the necessary information, such as the facility name, location, and contact information, and send it to the server. The server stores the user's account information in a database, allowing it to configure and manage each facility. The database uses a general-purpose database management system such as MySQL or PostgreSQL.

[1499] Device Registration and Connection

[1500] When a user signs up for a subscription, they are sent a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After the devices start up, they automatically send identification information to a server. The server uses this information to associate the devices with specific user accounts. Devices are identified using MAC addresses or unique IDs.

[1501] Data Collection and Monitoring

[1502] The device continuously collects audio and video data at the installation site. The collected data is sent in real time to a server, which stores the data in a buffer and inputs it into a data analysis tool at regular intervals. Data is sent using secure HTTP or MQTT protocols.

[1503] Data analysis and anomaly detection

[1504] The server uses a generative AI model to analyze the received audio and video data. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to detect abnormal behavior and abnormal audio. Abnormal behavior is detected as violent acts, and abnormal audio is detected as words such as "help me."

[1505] When the server detects an anomaly, it generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time of occurrence, and the location of the anomaly. The content of the alert contains enough information for the user to take prompt action.

[1506] Alert Notifications

[1507] The server generates alerts and notifies users via email, text message, or a dedicated app, using external services such as SendGrid and Twilio APIs to send notifications.

[1508] Users can check the alert notification they receive and access a dedicated app or website to obtain detailed information, allowing them to quickly grasp the situation on-site. Based on this information, they can take appropriate action.

[1509] Data storage and report generation

[1510] The server stores the collected audio and video data for a certain period of time. The data is then stored in cloud storage such as Amazon S3 or Google Cloud Storage. The stored data can be used for later verification and report creation.

[1511] The server periodically generates a report summarizing the results of the analysis and anomaly detection and provides it to the user in PDF or Excel format, allowing the user to grasp the overall status of the facility and take any necessary improvement measures.

[1512] Specific examples

[1513] For example, consider a case where a staff member at a nursing home is verbally abusing an elderly person.

[1514] 1. The device captures audio data from within the facility and sends it to a server. Specifically, the device's microphone picks up abusive language such as "idiot" and "die."

[1515] 2. The server analyzes the received audio data and detects abusive language.

[1516] 3. The server detects the abnormality and generates detailed alert information.

[1517] 4. The server sends an alert to the facility manager. Specifically, a push notification is sent to the manager's smartphone.

[1518] 5. The user (facility manager) checks the notification, rushes to the scene, records and saves the situation, and warns the staff.

[1519] In this way, the system of the present invention makes full use of AI technology to ensure safety within the facility and enables early detection of abuse and rapid response.

[1520] Prompt Sentence Examples

[1521] Here is an example of starting the system's processing by inputting the following prompt sentence into the generative AI model:

[1522] Set up an AI device at nursing home ABC and start real-time monitoring. Set it up to send an alert to the administrator as soon as it detects abusive language or abnormal behavior by staff.

[1523] This prompt is used as an example to demonstrate system configuration and processing.

[1524] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1525] Step 1:

[1526] Users access the system's website and create an account.

[1527] Input: Facility name, address, contact information, etc.

[1528] Process: The user enters this information into the registration form and clicks the "Submit" button.

[1529] Output: The data is sent to the server.

[1530] Step 2:

[1531] The server stores user account information in a database.

[1532] Input: Information such as facility name, address, and contact details submitted by the user.

[1533] Processing: Information is stored using database software (e.g., MySQL).

[1534] Output: The user account is registered in the database and a notification is sent to the user confirming account creation.

[1535] Step 3:

[1536] Once a user signs up for a subscription, they will be sent an AI device.

[1537] Input: Subscription contract information.

[1538] Processing: After the contract is confirmed, a specified number of AI devices will be sent to the user.

[1539] Output: The AI ​​device arrives at the user's facility.

[1540] Step 4:

[1541] Users install AI devices within their facilities and connect them to a network and power source.

[1542] Inputs: AI device, network information, power outlet.

[1543] Process: Place the device in its designated location, plug it into a power outlet, and configure Wi-Fi.

[1544] Output: The AI ​​device starts up.

[1545] Step 5:

[1546] After the device starts up, it automatically sends its identification information to the server.

[1547] Input: Device identification information (MAC address, unique ID, etc.).

[1548] Processing: The device establishes a network connection and sends its identification information to the server.

[1549] Output: The identification information is sent to the server.

[1550] Step 6:

[1551] The server associates the device identity with the user account.

[1552] Input: Device identification information, user account information.

[1553] Action: Updates the database to associate the device information with the user account.

[1554] Output: The device is associated with the user account.

[1555] Step 7:

[1556] The device continuously collects audio and video data from the location where it is installed.

[1557] Input: Location audio and video data.

[1558] Processing: The device's built-in camera and microphone capture audio and video.

[1559] Output: Collected data is temporarily stored on the device.

[1560] Step 8:

[1561] The device sends the collected data to a server in real time.

[1562] Input: Collected audio and video data.

[1563] Processing: Data is sent using secure HTTP or MQTT protocols.

[1564] Output: The data is sent to the server.

[1565] Step 9:

[1566] The server stores the received data in a buffer.

[1567] Input: Audio and video data sent from the device.

[1568] Processing: Storing data in RAM and temporary storage.

[1569] Output: The data is held in a buffer.

[1570] Step 10:

[1571] The server uses AI algorithms to analyze the data held in the buffer.

[1572] Input: Buffered audio and video data.

[1573] Processing: Detect anomalous behavior and sounds using TensorFlow and PyTorch.

[1574] Output: The analysis results are obtained.

[1575] Step 11:

[1576] The server detects abnormal behavior or sounds and generates alert information.

[1577] Input: Analysis results of the AI ​​model.

[1578] Processing: Generate an alert based on information such as the type of anomaly, the time of occurrence, and the location of the anomaly.

[1579] Output: Alert information is generated.

[1580] Step 12:

[1581] The server generates alerts and notifies users via email, text message, or a dedicated app.

[1582] Input: Alert information.

[1583] Processing: Send notifications using SendGrid or Twilio APIs.

[1584] Output: The alert is notified to the user.

[1585] Step 13:

[1586] The user checks the received alert notification and obtains detailed information.

[1587] Input: Alert notification.

[1588] Action: Access the dedicated app or website to check detailed information.

[1589] Output: Information is obtained to understand the situation on site.

[1590] Step 14:

[1591] The server stores the collected audio and video data for a certain period of time.

[1592] Input: Collected audio and video data.

[1593] Processing: Store data in Amazon S3 or Google Cloud Storage.

[1594] Output: Data is saved to cloud storage.

[1595] Step 15:

[1596] The server periodically generates a report summarizing the results of the data analysis.

[1597] Input: Analysis results, past data.

[1598] Processing: Generates reports in PDF and Excel formats and provides them to the user.

[1599] Output: A report is provided to understand the status of the facility.

[1600] (Application example 1)

[1601] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1602] In modern society, early detection and prevention of abuse in facilities such as daycare centers and elderly care facilities is a major challenge. However, conventional monitoring systems have limitations in their ability to analyze massive amounts of data in real time and quickly detect abnormal behavior. This increases the time between the occurrence of an abnormality and a response, risking the spread of damage. In addition, multiple notification methods and devices are often not properly coordinated, hindering administrators' ability to respond quickly.

[1603] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1604] In this invention, the server includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, a cloud storage means, an abnormal behavior analysis means using a generative AI model, and a push notification means. This allows for real-time analysis of audio and video data collected within the facility, enabling the immediate detection of abnormal behavior or abnormal sounds. Furthermore, by using multiple notification methods, managers can quickly identify abnormalities and take necessary measures.

[1605] "Data collection means" refers to a system that uses devices and sensors to collect data such as audio and video in real time.

[1606] "Data analysis means" refers to a system that includes software and algorithms for processing collected data and detecting abnormal behavior or sounds.

[1607] An "anomaly detection means" is a system that includes functions and algorithms for automatically detecting abnormal behavior or abnormal sounds from collected and analyzed data.

[1608] An "alert notification method" is a communication method such as email, text message, in-app notification, or push notification that is used to warn an administrator when an abnormality is detected.

[1609] "Data storage means" refers to a storage system that stores collected audio and video data for a certain period of time so that it can be verified or referenced later.

[1610] "Cloud storage means" refers to a remote storage service for storing and managing data over the Internet, enabling the storage of large amounts of data without being restricted by physical devices.

[1611] "Means for analyzing abnormal behavior using generative AI models" refers to a function that uses artificial intelligence (AI) models to detect and identify unusual behavior patterns when analyzing collected data.

[1612] "Push notification" is a technology for sending instant notifications to smartphones and other devices, providing users with important information in real time.

[1613] This invention is a system for preventing abuse in facilities such as nurseries and nursing homes. The system is made up of the following main components, and its specific implementation method is described below.

[1614] 1. Data Collection Methods

[1615] The data collection methods include devices and sensors that collect audio and video data in real time within the facility. Specifically, microphones and cameras are used. These devices are installed in various locations within the facility, continuously collect data, and transmit it to a server via a network.

[1616] 2. Data analysis methods

[1617] The data analysis means includes software and algorithms for processing the collected data and detecting abnormal behavior and sounds. Specific software used is a Python program and an external analysis API. This allows data to be analyzed in real time and abnormal behavior and sounds to be detected.

[1618] 3. Anomaly detection methods

[1619] The anomaly detection method includes functions and algorithms that automatically detect abnormal behavior and sounds from collected and analyzed data. Specifically, a generative AI model is used. This AI model learns behaviors that deviate from normal behavior patterns and identifies abnormalities.

[1620] 4. Alert notification methods

[1621] Alert notification means include communication means such as email, text message, in-app notification, and push notification to alert administrators when an abnormality is detected, allowing administrators to immediately identify the abnormality and take prompt action.

[1622] 5. Data storage means

[1623] The data storage means includes a storage system that stores collected audio and video data for a certain period of time so that it can be verified and referenced later. Specifically, Firebase storage is used. This allows past data to be stored and verified when necessary.

[1624] 6. Cloud Storage Solutions

[1625] Cloud storage services are remote storage services for storing and managing data over the Internet, allowing for the storage of large amounts of data without being restricted by physical devices. Specifically, Firebase Storage is one such service.

[1626] 7. Methods for analyzing abnormal behavior using generative AI models

[1627] The abnormal behavior analysis method using a generative AI model includes a function that uses an artificial intelligence (AI) model to detect and identify abnormal behavior patterns when analyzing collected data, thereby enabling the detection of abnormal behavior with high accuracy.

[1628] 8. Push Notification Methods

[1629] Push notification is a technology for instantly sending notifications to smartphones and other devices, including a means to provide users with important information in real time, allowing administrators to quickly identify and respond to abnormalities.

[1630] Specific examples

[1631] For example, if a staff member in a nursing home is verbally abusing an elderly person, the system will function as follows:

[1632] 1. The device captures audio data within the facility and sends it to the server.

[1633] 2. The server analyzes the received audio data and detects abnormal audio such as abusive language.

[1634] 3. The server detects the abnormality and generates alert information.

[1635] 4. The server sends a notification to the facility manager via email or a dedicated app.

[1636] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (warning staff and recording the situation).

[1637] Prompt Sentence Examples

[1638] "Please tell me how to detect abnormal sounds while working at a nursing home."

[1639] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1640] Step 1:

[1641] The user registers an account with the system along with facility information. The user enters facility information (facility name, address, contact information, etc.) and sends it to the server. This information is saved in a database by the server and linked to the user account. The input is facility information, and the output is the user profile saved in the database.

[1642] Step 2:

[1643] A user installs a device in a facility and connects it to the network and power. When the device starts up, it automatically sends its identification information to the server. The server receives this identification information and associates the device with a specific user account. The input is the device's identification information, and the output is the device information stored in a database.

[1644] Step 3:

[1645] The device collects audio and video data in the facility in real time and sends it to the server. The server stores the received data in a buffer and inputs it into the data analysis means at regular time intervals. The input is the audio and video data transmitted in real time, and the output is the data temporarily stored in the buffer.

[1646] Step 4:

[1647] The server analyzes the collected audio and video data using a generative AI model. The purpose of the analysis is to detect abnormal behavior or sounds that deviate from normal behavior patterns. The input is the audio and video data stored in the buffer, and the output is the analysis results, which are information on abnormal behavior or sounds.

[1648] Step 5:

[1649] If an anomaly is detected, the server generates an alert containing information about the anomaly. This alert includes the type of anomaly, the time it occurred, and the location where it occurred. The input is the anomaly detection information as an analysis result, and the output is the alert information.

[1650] Step 6:

[1651] The server generates alerts and sends them to administrators via email, text message, in-app notification, or push notification. The input is the alert information, and the output is the notification received by the administrator.

[1652] Step 7:

[1653] The administrator receives a notification and accesses a dedicated app or website to check detailed information. The input is a notification of an abnormality detection, and the output is the result of checking the detailed information. Based on this information, the administrator can rush to the site and take appropriate measures.

[1654] Step 8:

[1655] The server stores the collected audio and video data for a certain period of time and uses it for later verification and report creation. The input is the collected audio and video data, and the output is the data saved in storage. This makes it possible to reference past data as needed.

[1656] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1657] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[1658] Account Creation and Initial Setup

[1659] First, users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[1660] The server stores user account information in a database, allowing users to configure and manage settings for each facility.

[1661] Device Registration and Connection

[1662] When a user signs up for a subscription, they will receive a set number of AI devices, which they then install in their facility and connect to the network and power source.

[1663] After starting up, the device sends its own identification information (device ID, MAC address, etc.) to the server.

[1664] Data Collection and Monitoring

[1665] The device continuously collects audio and video data at the installation site, and transmits the collected data to a server in real time.

[1666] The server stores the received data in a buffer and inputs it to the data analysis means at regular time intervals.

[1667] Data analysis and anomaly detection

[1668] The server uses AI algorithms to analyze the collected audio and video data.

[1669] AI algorithms detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying).

[1670] Emotion recognition by emotion engine

[1671] The emotion engine installed on the server analyzes the collected audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[1672] The server uses the emotional state recognized by the emotion engine as additional input when the anomaly detection means determines an anomaly.

[1673] Alert Notifications

[1674] If an anomaly is detected, the server generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state.

[1675] The server notifies the user of the alert via email, SMS, or a dedicated in-app notification, or a combination of these.

[1676] Data storage and report generation

[1677] The server stores the collected data for a certain period of time, which allows you to refer to the data later.

[1678] The server periodically generates a report summarizing the analysis results and provides it to the user, who can use this report to understand the safety status of the entire facility and take any necessary improvement measures.

[1679] Specific examples

[1680] For example, let's assume that a staff member at a nursing home is using abusive or insulting language towards an elderly person. In this case, the system works as follows:

[1681] 1. The device captures audio and video data from within the facility and sends it to the server.

[1682] 2. The server analyzes the received data and detects abnormal voices, including abusive or insulting language.

[1683] 3. The server uses an emotion engine to recognize the emotional state of the elderly person (e.g., fear, pain) from the video data.

[1684] 4. The server generates alert information based on the abnormality and emotion recognition results and notifies the facility manager of this information.

[1685] 5. The user (facility manager) receives the notification, checks the detailed information, rushes to the scene, and takes the necessary action (such as warning staff or implementing countermeasures).

[1686] In this way, by combining an emotion engine, the present invention provides a system that achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, which contributes to improving safety within facilities and is particularly effective in ensuring the welfare of the elderly and children.

[1687] The processing flow will be explained below.

[1688] Step 1:

[1689] Users access the system's website, create an account, enter the required information (facility name, address, contact information, etc.), and submit.

[1690] Step 2:

[1691] The server receives the user's input information, saves the account information in a database, and generates authentication information (user ID and password) and sends it to the user's email address.

[1692] Step 3:

[1693] Users sign a subscription contract and receive a specified number of AI devices, which they then install in appropriate locations within their facility and connect to a network and power source.

[1694] Step 4:

[1695] When the device starts up, it sends its identification information (device ID, MAC address, etc.) to the server.

[1696] Step 5:

[1697] The server receives the identification information and associates it with a user account, thereby establishing that the device belongs to a particular facility.

[1698] Step 6:

[1699] The device continuously collects audio and video data within the facility and transmits this data to a server in real time.

[1700] Step 7:

[1701] The server stores the received data in a buffer and performs analysis using a data analysis method. An AI algorithm is used.

[1702] Step 8:

[1703] The server's AI algorithm analyzes audio and video data to detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[1704] Step 9:

[1705] An emotion engine installed on the server analyzes audio and video data and recognizes the user's emotional state (e.g., fear, pain, etc.).

[1706] Step 10:

[1707] The server uses the emotional state as additional data to improve the accuracy of anomaly detection. If the emotional state is associated with an anomaly, the server will determine the anomaly with higher sensitivity.

[1708] Step 11:

[1709] As a final analysis result, the server generates alert information if an anomaly is detected, which includes details of abnormal behavior, abnormal sound, recognized emotional state, time of occurrence, and location of occurrence.

[1710] Step 12:

[1711] Based on the alert information, the server sends a notification to the user using one or more of email, SMS, and / or a notification within a dedicated app.

[1712] Step 13:

[1713] Users receive an alert notification and can access a dedicated app or website to check detailed information, allowing them to understand the situation on-site in detail and take prompt action.

[1714] Step 14:

[1715] The server stores the collected data for a certain period of time, which allows for future reference and verification of the data.

[1716] Step 15:

[1717] The server periodically generates a report summarizing the analysis results and provides it to the user, helping them understand the safety status of the entire facility and take any necessary improvement measures.

[1718] In this way, through a series of processing steps, the system can quickly detect acts of abuse in daycare centers and elderly care facilities and take immediate action.By combining it with an emotion engine, it becomes possible to detect anomalies with high accuracy, taking into account emotional states.

[1719] Example 2

[1720] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1721] To prevent abuse in daycare centers and elderly care facilities, it is necessary to accurately detect abnormal behavior and sounds within the facility and take prompt action. However, conventional monitoring systems are limited to detecting abnormal behavior and sounds, and do not take emotional states into account when detecting abnormalities. Therefore, a system that can more reliably ensure safety within the facility is needed.

[1722] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1723] In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means, which not only detects abnormal behavior and abnormal voice from audio and video data but also recognizes emotional states, enabling more accurate detection of abnormalities and quicker response.

[1724] "Data collection means" means a device or system for continuously collecting audio and video data within a facility.

[1725] "Data analysis means" refers to software or algorithms that analyze collected audio and video data and detect abnormal behavior or abnormal sounds.

[1726] The "anomaly detection means" is an algorithm or system for identifying abnormal behavior or abnormal sounds from the data analyzed by the data analysis means and detecting abnormalities.

[1727] "Emotion recognition means" refers to software or algorithms that analyze and recognize emotional states based on collected audio and video data.

[1728] An "alert notification means" is a device or system for notifying a user of information about a detected abnormality, and includes means such as email, text message, and in-app notification.

[1729] "Data storage means" refers to a device or system that stores collected and analyzed data for a certain period of time and makes it available for later reference.

[1730] "Server" means a central processing unit for collecting, analyzing, storing, and notifying data.

[1731] An "AI algorithm" is a computational method that uses machine learning and deep learning to analyze audio and video data and detect anomalies.

[1732] A "time frame" is a unit of time for processing data by dividing it into regular time intervals.

[1733] A "buffer" is a storage area for temporarily storing data.

[1734] This invention relates to a system that uses an AI device combined with an emotion engine to prevent abuse in daycare centers and elderly care facilities. The system includes a data collection means, a data analysis means, an anomaly detection means, an emotion recognition means, an alert notification means, a data storage means, and an emotion engine, thereby ensuring safety within the facility.

[1735] To start using the system, users first access the system's website and create an account. They enter the required information (facility name, address, contact information, etc.) and submit it. The server then receives the entered information and stores it in a database. This allows for the configuration and management of each facility.

[1736] Next, once the user completes the subscription contract, they will receive a set number of AI devices. The user installs these devices in their facility and connects them to the network and power source. After booting up, the devices will send their identification information (device ID, MAC address, etc.) to the server.

[1737] The devices are installed at various locations within the facility and continuously collect audio and video data using cameras and microphones. The collected data is transmitted in real time to a server, which temporarily stores it in a buffer. The data is then input into a data analysis tool every certain time frame (e.g., 5 or 10 seconds).

[1738] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent movements) and abnormal sounds (such as screaming or crying). It also uses an emotion engine to analyze audio and video data to recognize the user's emotional state. Voice tone and facial expression data are used to analyze the emotional state.

[1739] This emotional state is important auxiliary data for the anomaly detection means to determine an abnormality. If an abnormality is detected, the server generates detailed alert information. This information includes the type of abnormality, the time of occurrence, the location of occurrence, and the recognized emotional state. The generated alert information is notified to the user via one or more of email, text message, and / or dedicated in-app notification.

[1740] Furthermore, the server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to grasp the safety status of the entire facility and take necessary improvement measures.

[1741] Specific examples

[1742] For example, consider a situation where a staff member at a nursing home uses abusive or insulting language toward an elderly person. In this case, a device captures audio and video data from within the facility and sends it to a server. The server then stores the received data in a buffer and analyzes it using an AI algorithm. Abnormal audio, including abusive or insulting language, is detected. The server then uses an emotion engine to recognize the elderly person's emotional state (e.g., fear, distress) from the video data. An alert is then generated based on the abnormal behavior and the emotion recognition results, and this information is sent to the facility manager. The user (facility manager) who receives the notification can check the details, rush to the scene, and take the necessary action (e.g., warn staff, implement immediate countermeasures).

[1743] In this way, by combining the emotion engine, this system achieves more accurate anomaly detection and faster response than conventional anomaly detection systems, thereby improving safety within facilities and ensuring the welfare of the elderly and children.

[1744] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1745] Step 1:

[1746] The user accesses the system's website, creates an account, enters the necessary information such as the facility name, address, and contact information, and submits it.

[1747] Input: User information such as facility name, address, and contact details.

[1748] Process: Sends the information entered by the user to the server.

[1749] Output: User information is saved on the server.

[1750] Step 2:

[1751] The server receives the information sent by the user and stores it in a database, which allows for the configuration and management of each facility.

[1752] Input: User-submitted information such as facility name, address, and contact details.

[1753] Processing: The server stores the information in a database.

[1754] Output: User information stored in the database.

[1755] Step 3:

[1756] Once a user completes a subscription, a set number of AI devices will be delivered to the facility, which the user must then install and connect to the network and power source.

[1757] Input: Subscription contract information, AI device.

[1758] Processing: The AI ​​device is installed on-site and connected to a network and power source.

[1759] Output: AI device installed within the facility.

[1760] Step 4:

[1761] After the device starts up, it sends its own identification information (device ID, MAC address, etc.) to the server.

[1762] Input: Device ID, MAC Address.

[1763] Process: The device starts up and sends its identification information to the server.

[1764] Output: The device identification sent to the server.

[1765] Step 5:

[1766] The devices are installed at various locations within the facility and use cameras and microphones to continuously collect audio and video data.

[1767] Input: Location audio and video data.

[1768] Processing: Capture audio and video data in real time.

[1769] Output: Collected audio and video data.

[1770] Step 6:

[1771] The server receives the collected data in real time, temporarily stores it in a buffer, and then inputs it into the data analysis means every certain time frame (e.g., 5 seconds, 10 seconds).

[1772] Input: Collected audio and video data.

[1773] Processing: Data is held in a buffer and input to a data analysis tool for each time frame.

[1774] Output: Audio and video data input to data analysis means.

[1775] Step 7:

[1776] The server uses AI algorithms (e.g., TensorFlow, PyTorch) to analyze the collected data and detect abnormal behavior (such as violent actions) and abnormal sounds (such as screaming or crying).

[1777] Input: Input audio and video data per time frame.

[1778] Processing: Analyze the data using AI algorithms to detect abnormal behavior and audio.

[1779] Output: Detected abnormal behavior and audio.

[1780] Step 8:

[1781] The server uses an emotion engine to analyze the audio and video data to recognize the user's emotional state, using voice tone and facial expression data.

[1782] Input: Input audio and video data per time frame.

[1783] Processing: Analyze the emotional state using the emotion engine.

[1784] Output: Perceived emotional state.

[1785] Step 9:

[1786] If the server detects an anomaly, it generates detailed alert information, including the type of anomaly, the time and location of the anomaly, and the perceived emotional state. The generated alert information is then sent to the user via email, text message, or in-app notification.

[1787] Input: Detected abnormal behavior and emotional state.

[1788] Action: Generate and notify alert information.

[1789] Output: Alert information notified to the user.

[1790] Step 10:

[1791] The server has the function of storing the collected data for a certain period of time and periodically generates reports on the results and provides them to users, allowing them to understand the safety status of the entire facility and take necessary improvement measures.

[1792] Input: Collected audio and video data.

[1793] Processing: The data is stored for a certain period of time and the analysis results are generated as a report.

[1794] Output: The report provided to the user.

[1795] (Application example 2)

[1796] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1797] The problem to be solved by this invention is to effectively prevent abuse in nurseries and nursing homes and to increase safety within the facilities. In particular, the object is to provide a system that can detect abuse at an early stage and take prompt and appropriate action.

[1798] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a data collection means, a data analysis means, an abnormality detection means, an emotion recognition means, an alert notification means, and a data storage means. This allows for real-time data collection from devices installed in the facility and analysis of the collected data, enabling early detection of abusive behavior and abnormal behavior and recognition of emotional states based on voice tone and facial expression data. Furthermore, when an abnormality is detected, an alert can be quickly sent via email, text message, or a dedicated app, prompting facility managers and relatives to take immediate action. Furthermore, by storing video and audio data along with the analysis results for a certain period of time, the data can be referenced later, providing evidence of problematic behavior.

[1799] "Data collection means" refers to means for collecting audio data and video data from devices installed within the facility.

[1800] The "data analysis means" is a means for analyzing collected audio data and video data and extracting necessary information.

[1801] "Abnormality detection means" is a means for detecting abusive behavior or abnormal behaviour from information extracted through data analysis.

[1802] The "emotion recognition means" is a means for analyzing voice tone and facial expression data to recognize the emotional state of a subject.

[1803] The "alert notification means" is a means for notifying relevant users of an alert when an abnormality is detected using email, text message, and app notification.

[1804] "Data storage means" refers to a means for storing collected data and analysis results for a certain period of time so that they can be referenced later.

[1805] A "server" is a central processing unit that integrates data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means, and functions as a system.

[1806] This system includes data collection means, data analysis means, anomaly detection means, emotion recognition means, alert notification means, and data storage means to prevent abuse in facilities such as daycare centers and elderly care facilities. The server is at the core of this system, integrating and managing each device and user device.

[1807] Installing and connecting the device

[1808] A certain number of AI devices will be installed within the facility. These devices will be connected to the network and power source, and after startup, will send their identification information (device ID, MAC address, etc.) to the server.

[1809] Data collection and analysis

[1810] The device continuously collects audio and video data at the installation location and transmits this data to the server in real time. The server stores the received data in a buffer and inputs it to the data analysis means at regular intervals. The data analysis means mainly uses the following software libraries:

[1811] Keras: A machine learning library for implementing emotion engine models.

[1812] OpenCV: A computer vision library for preprocessing image data and analyzing video data.

[1813] Anomaly detection and emotion recognition

[1814] The server uses AI algorithms to analyze the collected audio and video data. The anomaly detection means detects abnormal behavior and sounds, such as violent actions, screaming, or crying. The emotion recognition means analyzes voice tone and facial expression data to recognize the subject's emotional state (e.g., fear, pain). This improves the accuracy of anomaly detection.

[1815] Alert Notifications

[1816] If an abnormality is detected, the server generates detailed alert information and notifies facility managers and relatives of the incident. The alert notification method uses the following communication methods:

[1817] Email

[1818] Text message

[1819] In-app notifications

[1820] Data storage and reporting

[1821] The server stores the collected data and analysis results for a certain period of time, allowing the data to be referenced later and providing evidence of problematic behavior.The server also periodically generates a report summarizing the analysis results and provides it to users, making it easier to understand the safety status of the entire facility.

[1822] Specific examples

[1823] For example, if a staff member at a nursing home uses abusive or insulting language towards an elderly person, the system works as follows:

[1824] 1. The device captures audio and video data from within the facility and sends it to the server.

[1825] 2. The server analyzes the received data and detects abnormal voice and behavior, including abusive or insulting language.

[1826] 3. The emotion recognition means recognizes the elderly person's emotional state (e.g., fear, distress) from voice tone and facial expression data.

[1827] 4. Based on the results of anomaly detection and emotion recognition, the server generates alert information and notifies facility managers and relatives.

[1828] 5. Users can receive notifications, check detailed information, rush to the scene, and take necessary action (such as alerting staff or implementing countermeasures).

[1829] Example prompts for generative AI models

[1830] "We will demonstrate an anomaly detection system for abusive language and insulting behavior in a nursing home. The camera collects video and audio in real time, and if the emotion engine detects abnormal behavior, it will send an alert to the administrator."

[1831] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1832] Step 1:

[1833] Device installation and initial setup:

[1834] Users install a set number of AI devices within their facilities and connect them to a network and power source. After starting up, the devices send their identification information (device ID, MAC address, etc.) to the server, which then recognizes the devices and registers them in the system.

[1835] Input: Device identification information

[1836] Output: Device registration information on the server

[1837] Specific operations: Device network connection, sending of identification information, registration process on the server

[1838] Step 2:

[1839] Data collection:

[1840] The device installed inside the terminal continuously collects audio and video data from within the facility and transmits it to a server in real time.

[1841] Input: Audio and video data

[1842] Output: Raw data sent to the server

[1843] Specific operations: audio and video capture, real-time data transmission

[1844] Step 3:

[1845] Data Analysis:

[1846] The server stores the received audio and video data in a buffer and then analyzes the data using Keras and OpenCV.

[1847] Input: Raw audio and video data

[1848] Output: Analyzed data (e.g., specific motion or voice features)

[1849] Specific operation: Data input, preprocessing, and analysis using machine learning models

[1850] Step 4:

[1851] Anomaly detection:

[1852] Based on the results of the data analysis, the server uses anomaly detection methods to detect abusive or abnormal behavior, such as identifying specific movement features or voice patterns as abnormal.

[1853] Input: Parsed data

[1854] Output: Anomaly detection results

[1855] Specific operations: feature analysis, abnormal / normal determination

[1856] Step 5:

[1857] Emotion recognition:

[1858] The server uses an emotion recognition means to recognize the emotional state of the subject from the analyzed voice tone and facial expression data.

[1859] Input: Voice tone data and facial expression data

[1860] Output: Emotion recognition results (e.g. fear, pain)

[1861] Specific actions: Analysis of voice tone, analysis of facial expression data, estimation of emotional state

[1862] Step 6:

[1863] Alerting and Notification:

[1864] The server generates alert information based on the results of anomaly detection and emotion recognition, and notifies facility managers and relatives through alert notification methods such as email, text message, and in-app notification.

[1865] Input: Anomaly detection results and emotion recognition results

[1866] Output: Alert information, notification message

[1867] Specific actions: Generate alert information, send notification messages

[1868] Step 7:

[1869] Data storage and reporting:

[1870] The server stores the collected data and analysis results for a certain period of time, and periodically generates a report summarizing the analysis results and provides it to the user. This allows the data to be referenced later, providing evidence of problematic behavior.

[1871] Input: Collected data, analysis results

[1872] Output: Saved data, scheduled reports

[1873] Specific actions: saving data, generating reports, and providing them to users

[1874] Through the above processing steps, the present system can enhance safety within the facility and ensure the welfare of the elderly and children in particular.

[1875] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1876] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1877] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1878] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1879] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1880] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1881] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1882] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1883] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1884] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1885] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1886] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1887] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1888] 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.

[1889] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1890] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1891] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1892] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1893] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1894] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1895] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1896] The following is further disclosed regarding the above embodiment.

[1897] (Claim 1)

[1898] data collection means;

[1899] data analysis means;

[1900] An anomaly detection means;

[1901] an alert notification means;

[1902] Data storage means

[1903] A system including:

[1904] (Claim 2)

[1905] 10. The system of claim 1, wherein the anomaly detection means comprises an algorithm for detecting abnormal sounds in the audio data.

[1906] (Claim 3)

[1907] 10. The system of claim 1, wherein the alert notification means uses at least one of an email, a text message, and an in-app notification.

[1908] "Example 1"

[1909] (Claim 1)

[1910] How to create an account;

[1911] means for transmitting device identification information to a server and associating the device with a user account;

[1912] means for collecting audio and video data;

[1913] a means for analyzing the received data by time frame;

[1914] A method for detecting abnormal behavior and abnormal sounds using AI algorithms,

[1915] means for generating an alert containing anomaly information;

[1916] A means to notify you of generated alerts via email, text message, or app;

[1917] a means for storing the audio and video data for a certain period of time;

[1918] A means of generating reports of data analysis results

[1919] A system including:

[1920] (Claim 2)

[1921] 10. The system of claim 1, wherein the anomaly detection means comprises an algorithm for detecting abnormal sounds in the audio data.

[1922] (Claim 3)

[1923] 10. The system of claim 1, wherein the alert notification means uses at least one of an email, a text message, and an in-app notification.

[1924] "Application Example 1"

[1925] (Claim 1)

[1926] data collection means;

[1927] data analysis means;

[1928] An anomaly detection means;

[1929] an alert notification means;

[1930] A data storage means;

[1931] a cloud storage means;

[1932] Abnormal behavior analysis method using generative AI models;

[1933] Push notification methods

[1934] A system including:

[1935] (Claim 2)

[1936] 10. The system of claim 1, wherein the anomaly detection means includes an algorithm for detecting abnormal sounds in audio data and a generative AI model for detecting abnormal behavior in video data.

[1937] (Claim 3)

[1938] 10. The system of claim 1, wherein the alert notification means uses at least one of email, text message, in-app notification, and push notification.

[1939] "Example 2: Combining Emotion Engines"

[1940] (Claim 1)

[1941] data collection means;

[1942] data analysis means;

[1943] An anomaly detection means;

[1944] An emotion recognition means;

[1945] an alert notification means;

[1946] Data storage means

[1947] A system including:

[1948] (Claim 2)

[1949] 10. The system of claim 1, wherein the anomaly detection means comprises an algorithm for detecting abnormal sounds in the audio data.

[1950] (Claim 3)

[1951] 10. The system of claim 1, wherein the alert notification means uses at least one of an email, a text message, and an in-app notification.

[1952] (Claim 4)

[1953] 10. The system of claim 1, wherein the emotion recognition means comprises an algorithm that analyzes audio and video data to recognize an emotional state.

[1954] (Claim 5)

[1955] 2. The system according to claim 1, wherein the data analysis means has a function of processing the data for each fixed time frame.

[1956] (Claim 6)

[1957] 2. The system according to claim 1, wherein the data storage means has a function of storing the collected data for a certain period of time.

[1958] (Claim 7)

[1959] 2. The system of claim 1, wherein the emotion recognition means has a function of providing additional input to the anomaly detection means' anomaly determination.

[1960] "Application example 2 when combining emotion engines"

[1961] (Claim 1)

[1962] data collection means;

[1963] data analysis means;

[1964] An anomaly detection means;

[1965] An emotion recognition means;

[1966] an alert notification means;

[1967] Data storage means

[1968] A system including:

[1969] (Claim 2)

[1970] 2. The system of claim 1, wherein the anomaly detection means includes an algorithm for detecting abnormal sounds in audio data and abnormal behavior in video data.

[1971] (Claim 3)

[1972] 2. The system of claim 1, wherein the emotion recognition means includes an algorithm that analyzes voice tone and facial expression data to recognize an emotional state.

[1973] (Claim 4)

[1974] 10. The system of claim 1, wherein the alert notification means uses at least one of an email, a text message, and an in-app notification.

[1975] (Claim 5)

[1976] 2. The system according to claim 1, wherein the data storage means stores the video and audio data together with the analysis results for a certain period of time so that the data can be referenced later. [Explanation of symbols]

[1977] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. data collection means; data analysis means; An anomaly detection means; an alert notification means; Data storage means A system including:

2. 2. The system of claim 1, wherein the anomaly detection means includes an algorithm for detecting abnormal sounds in the audio data.

3. The system of claim 1 , wherein the alert notification means uses at least one of an email, a text message, and an in-app notification.

Images