System

Abstract

A system is provided.SOLUTION: Means for acquiring video data from a plurality of video capturing devices installed in a nursery facility; means for transmitting the video data to a server device in real time; A system comprising: means for identifying a child's behavior pattern; means for detecting an abnormal behavior or a dangerous situation by the server device; means for issuing an alarm to a parent or a caregiver when the abnormal behavior or the dangerous situation is detected; means for continuously monitoring the video data 24 hours a day and 365 days a year by the server device; and means for providing information for optimizing a childcare plan based on the analysis result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In childcare facilities, constant monitoring of multiple children places a heavy burden on caregivers. Rapid response is required, particularly in the event of abnormal behavior or an emergency, but keeping an eye on all children is difficult. Caregivers also need to concentrate to provide high-quality care to each child. Conventional monitoring systems have problems with detecting abnormal behavior in real time and optimizing care plans based on long-term data accumulation. Therefore, there is a need for an efficient system that reduces the burden on caregivers and provides higher-quality care while ensuring the safety of children in childcare facilities. [Means for solving the problem]

[0005] This invention provides a means for acquiring video data from multiple video capture devices installed within a childcare facility and transmitting it to a server device in real time. The server device also includes a means for analyzing the video data to identify children's behavioral patterns and detect abnormal behavior or dangerous situations. If abnormal behavior or dangerous situations are detected, an alert is sent immediately to parents and caregivers, encouraging prompt action. This system also provides continuous monitoring 24 hours a day, 365 days a year, and includes a means for providing information to support the optimization of childcare plans through long-term data accumulation and analysis. This makes it possible to simultaneously ensure the safety of children in childcare facilities and reduce the burden on caregivers.

[0006] A "video recording device" is a camera device that is installed in rooms, play areas, etc. within a childcare facility to capture the daily activities of children.

[0007] "Video data" refers to digital data containing video information captured by a video capture device, and is used for analysis and monitoring.

[0008] A "server device" is a computer system that receives video data from multiple video capture devices via a network, and performs analysis and issues warnings.

[0009] "Transmitting in real time" means transmitting video data captured by a video shooting device to a server device immediately without delay using a communication protocol.

[0010] "Behavioral patterns" are the results of analyzing a series of children's movements and activities and classifying them into specific categories or patterns.

[0011] "Abnormal behavior" refers to behavior that is out of the ordinary or indicates an emergency, and includes falls, sudden bumps, and excessive excitement.

[0012] "Sending an alarm" means issuing a warning to parents or caregivers by text message, voice notification, or other means when abnormal behavior or a dangerous situation is detected.

[0013] "Continuous monitoring 24 hours a day, 365 days a year" means monitoring children's behavior day and night without interruption for a year.

[0014] "Optimizing childcare plans" involves analyzing accumulated behavioral data and improving plans and educational policies for more effective and safe childcare.

[0015] A "deep learning model" is a machine learning algorithm that uses neural networks to extract features from large amounts of data and classify behavioral patterns.

[0016] The "Real-time Streaming Protocol" is a communications protocol for continuously transmitting video data in real time, and is a technology for minimizing video delays. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

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

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

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

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0032] 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.

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

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected.

[0039] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[0040] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[0041] The server device detects abnormal behavior based on the results of behavioral pattern analysis. Abnormal behavior includes falls, sudden collisions, and excessive excitement. If abnormal behavior is detected, the server device immediately generates an alert and sends it to the device of the parent or caregiver. The alert is sent in the form of a push notification, SMS, email, or other format, allowing caregivers to respond quickly.

[0042] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral data, it can provide information to help optimize childcare plans. This allows childcare workers to make data-based decisions and improve the quality of childcare.

[0043] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server device instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker will immediately receive the alert and be able to respond promptly.

[0044] In this way, this system provides an effective means of ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[0045] The processing flow will be explained below.

[0046] Step 1:

[0047] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[0048] Step 2:

[0049] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[0050] Step 3:

[0051] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[0052] Step 4:

[0053] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[0054] Step 5:

[0055] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[0056] Step 6:

[0057] If an abnormal behavior is detected, the server generates an alert message including a timestamp and location information, as well as the specific details of the abnormal behavior and its urgency.

[0058] Step 7:

[0059] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[0060] Step 8:

[0061] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[0062] Step 9:

[0063] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[0064] Step 10:

[0065] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns and abnormal behavior. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[0066] The above are the specific processing steps and details of each operation of the AI ​​support and care system. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers.

[0067] Example 1

[0068] 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."

[0069] Conventional safety management systems in childcare facilities lacked the means to monitor children's daily behavior 24 hours a day, 365 days a year. Furthermore, alarm systems to quickly detect and respond to abnormal behavior or dangerous situations were also ineffective. This increased the burden on childcare workers, and led to problems such as insufficient efforts to ensure the safety of children.

[0070] 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.

[0071] In this invention, the server includes a means for temporarily storing video data in a buffer and checking for missing data and compression status, a means for identifying children's behavioral patterns from the video data after preprocessing using a deep learning model, and a means for comparing the identified behavioral patterns with set abnormal behavior conditions to detect abnormal behavior or dangerous situations. This enables real-time monitoring of children's behavior 24 hours a day, 365 days a year, quickly detecting abnormal behavior or dangerous situations, and issuing an alarm. This allows parents and caregivers to respond quickly, effectively ensuring children's safety and reducing the burden on caregivers. It can also provide information to support the optimization of childcare plans based on the accumulated behavioral data.

[0072] A "video recording device" is a device installed in a childcare facility to capture the daily activities of children.

[0073] "Video data" refers to data that includes video information of children's daily activities captured by a video camera.

[0074] The "server device" is a central processing unit for receiving and analyzing video data transmitted from multiple video capture devices installed in the childcare facility.

[0075] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time, and is a technology that minimizes data delays.

[0076] The "deep learning model" is an advanced machine learning model used to identify children's behavioral patterns from received video data and detect abnormal behavior.

[0077] A "behavioral pattern" is a sequence of specific movements or activities that children exhibit within a childcare facility.

[0078] "Abnormal behavior" refers to behavior that deviates from children's normal patterns of behavior and may lead directly to danger or crisis.

[0079] An "alert" is a warning message sent to parents or caregivers when abnormal behavior or a dangerous situation is detected.

[0080] A "push notification" is a real-time alert notification message sent directly from a server device to the device of a parent or caregiver.

[0081] "Optimizing childcare plans" refers to analyzing accumulated behavioral data to improve the quality of childcare and provide information to plan and execute more efficient childcare activities.

[0082] The present invention is a system for ensuring the safety of children in childcare facilities and reducing the burden on caregivers, and is composed of multiple video capture devices, a server device, and a notification system. Specific embodiments for carrying out the invention are described below.

[0083] Installation of video recording equipment and data acquisition

[0084] Device:

[0085] Video recording devices are installed in each room and play area within the childcare facility. 360-degree cameras or high-resolution cameras can be used for these. The video recording devices capture the children's daily activities 24 hours a day, 365 days a year, and transmit the video data to a server device in real time. A real-time streaming protocol (such as RTSP) is used to transmit the video data.

[0086] Server device data reception and analysis

[0087] server:

[0088] The server temporarily stores the received video data in a buffer and then analyzes it using a deep learning model, performing preprocessing such as color space conversion and frame resizing.

[0089] As a specific example, the server uses deep learning models such as ResNet and YOLO to identify children's behavioral patterns, including "running," "playing," and "resting."

[0090] Detecting abnormal behavior and generating alerts

[0091] server:

[0092] The server device compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior or dangerous situations, such as when a child falls off playground equipment or has a sudden collision.

[0093] When abnormal behavior is detected, an alert is immediately generated, containing information about the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[0094] Notification to parents and caregivers

[0095] server:

[0096] Alerts are sent to parents' and caregivers' devices in the form of push notifications, SMS, emails, etc. For example, an alert such as "A child has fallen off play equipment in the playroom" is immediately sent to a caregiver's smartphone.

[0097] Data accumulation and support for optimizing childcare plans

[0098] server:

[0099] The server device stores daily captured video data and analysis results over the long term, making it possible to track long-term changes in behavioral patterns. Based on the stored data, it can provide information to help optimize childcare plans. For example, it can provide statistical information on the activities of children at specific times of the day.

[0100] Specific examples

[0101] At 9:00 a.m., a video camera installed in the playroom of a childcare facility captures children playing and sends the video data to a server in real time. The server analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[0102] Prompt Sentence Examples

[0103] "Please explain the surveillance system in your childcare facility. This system uses video cameras to monitor children's behavior in real time and issues an alert if it detects abnormal behavior, such as falls or sudden bumps. The system also has 24 / 7 monitoring capabilities and analyzes the accumulated data to help optimize childcare planning."

[0104] In this way, the present invention provides a specific system for ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Program processing steps

[0107] Step 1:

[0108] Video data capture and transmission

[0109] Devices: Video cameras installed in each room and play area of ​​the childcare facility capture the children's daily activities 24 hours a day, 365 days a year. For example, they capture footage of children playing in the playroom at 9:00 AM.

[0110] Input: Children's daily activities

[0111] Output: Real-time video data

[0112] How it works: The video capture device is equipped with a 360-degree camera that captures children's movements from multiple angles. The captured video data is encrypted and sent to a server using a real-time streaming protocol (such as RTSP).

[0113] Step 2:

[0114] Data reception and primary analysis by the server

[0115] Server: The server temporarily stores the received video data in a buffer and checks for any data loss or compression issues.

[0116] Input: Real-time video data

[0117] Output: Pre-processed video data

[0118] Specific operation: The server checks the frame rate and resolution, checks for missing data, and then performs preprocessing such as color space conversion and frame resizing.

[0119] Step 3:

[0120] Behavioral pattern analysis

[0121] Server: The preprocessed video data is input into a deep learning model to identify children's behavioral patterns. Deep learning models such as ResNet and YOLO are used.

[0122] Input: Preprocessed video data

[0123] Output: Identified behavioral pattern data

[0124] Specific actions: The deep learning model in the server sequentially analyzes actions such as "running," "playing," and "resting" from video data.

[0125] Step 4:

[0126] Abnormal behavior detection

[0127] Server: Compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior and dangerous situations.

[0128] Input: Identified behavioral pattern data

[0129] Output: Abnormal behavior detection alert

[0130] Specific behavior: For example, the server detects a child falling off a playground equipment and flags it as abnormal behavior.

[0131] Step 5:

[0132] Alert generation and notification

[0133] Server: When abnormal behavior is detected, an alert is immediately generated and sent to the device of the parent or caregiver. The alert includes information such as the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[0134] Input: Abnormal behavior detection alert

[0135] Output: Warning message

[0136] Specific operation: The server generates an alert message stating, "A child has fallen off the playground equipment in the playroom," and sends it to the caregiver's smartphone in the form of a push notification, SMS, email, etc.

[0137] Step 6:

[0138] Data accumulation and support for optimizing childcare plans

[0139] Server: Stores daily captured video data and analysis results over the long term. Based on the stored data, it provides information to help optimize childcare plans.

[0140] Input: Identified behavioral pattern data and alarm data

[0141] Output: Statistics for optimizing childcare plans

[0142] Specific operation: The server generates statistics on the activities of children during specific time periods, which is used to review and improve childcare plans.

[0143] The above is the specific flow of program processing for this system.

[0144] (Application example 1)

[0145] 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."

[0146] While there are systems in childcare facilities that safely monitor children's behavior and quickly detect abnormal behavior or dangerous situations and notify parents and caregivers, similar functions are required in home and small childcare environments. There is also a need to provide a simple means of ensuring children's safety using the smartphones of parents and caregivers.

[0147] 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.

[0148] In this invention, the server includes means for acquiring video data from multiple video capture devices, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for the server device to detect abnormal behavior or dangerous situations, means for issuing an alert to parents or caregivers when the abnormal behavior or dangerous situation is detected, means for the server device to continuously monitor the video data 24 hours a day, 365 days a year, means for providing information to optimize childcare plans based on the analysis results, and means for detecting abnormal behavior and sending push notifications to smartphone applications used by families and certified childcare providers. This makes it possible to ensure the safety of children even in home or small childcare environments.

[0149] A "server device" is a central processing unit that analyzes video data over a network and issues an alarm based on the results.

[0150] A "video recording device" is a camera device that is installed in a facility or home to capture the behavior of children.

[0151] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time without delay.

[0152] A "deep learning model" is a machine learning model that analyzes children's behavioral patterns and detects abnormal behavior.

[0153] "Abnormal behavior" refers to behavior that is dangerous for children, such as falling or sudden collisions.

[0154] "Push notifications" are real-time notifications sent to users' smartphones when abnormal behavior is detected.

[0155] A "childcare plan" is a plan for childcare activities that is optimized based on children's behavioral data.

[0156] The "smartphone application used by families and certified childcare mothers" is a mobile application that monitors children's behavior at home and receives notifications when abnormal behavior is detected.

[0157] An "alert" is an alert issued when abnormal behavior or a dangerous situation is detected.

[0158] "Means for continuously monitoring the video data 24 hours a day, 365 days a year" refers to a function that continuously monitors children's behavior, regardless of day or night.

[0159] The present invention is a system for monitoring the behavior of children in childcare facilities and at home to ensure their safety. The system is configured as follows.

[0160] Hardware and software configuration

[0161] Hardware

[0162] Video recording equipment: Cameras installed in childcare facilities and homes monitor children's behavior 24 hours a day, 365 days a year.

[0163] Server device: A central processing unit that receives and analyzes video data. The server runs at all times and continuously monitors data.

[0164] Smartphone: A mobile device used by parents and certified childcare providers. Push notifications are sent when abnormal behavior is detected.

[0165] software

[0166] OpenCV: A library for preprocessing video data and image processing.

[0167] TensorFlow: A library that uses deep learning to analyze children's behavioral patterns and detect abnormal behavior.

[0168] smtplib: A Python module for sending email notifications when anomalous behavior is detected.

[0169] Data processing and calculation

[0170] 1. Camera input: The video camera continuously captures the children's behavior and transmits the video data to the server in real time using the real-time streaming protocol.

[0171] 2. Image preprocessing: The server preprocesses the received video frames to make them easier for the deep learning model to process. OpenCV is used for this preprocessing.

[0172] 3. Behavioral Analysis: Using a trained deep learning model, the server analyzes children's behavioral patterns and uses TensorFlow to identify specific movements and activities.

[0173] 4. Abnormal behavior detection: Abnormal behavior is detected based on the behavioral pattern analysis. If abnormal behavior is detected, an alert is sent to the parents using smtplib.

[0174] 5. Push notification: When abnormal behavior is detected, a push notification will be sent to the smartphone, allowing parents or certified childcare providers to take immediate action.

[0175] 6. Optimization of childcare plans: Through long-term data accumulation, the server provides information that is useful for optimizing childcare plans.

[0176] Specific examples

[0177] For example, if a camera installed in a home monitors a child's behavior and the child suddenly runs off and falls, the server will instantly detect this abnormal behavior and send a push notification to the parent's smartphone, allowing the parent to quickly rush to the scene.

[0178] Prompt Sentence Examples

[0179] "I would like to develop a real-time behavior analysis system using a camera. I need a function that detects abnormal behavior in children and notifies them by email. Please provide me with Python code that uses OpenCV and TensorFlow."

[0180] As described above, the present invention is a system that ensures the safety of children in childcare facilities and homes and reduces the burden on parents and caregivers.

[0181] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0182] Step 1:

[0183] Video data acquisition and transmission:

[0184] The video camera is installed in a childcare facility or a home and captures children's behavior 24 hours a day, 365 days a year. The captured video data is sent to a server using a real-time streaming protocol. The input is a video frame, and the output is real-time video data sent to the server.

[0185] Step 2:

[0186] Image preprocessing:

[0187] The server preprocesses the received video frames to make them easier for the deep learning model to process. Specifically, it resizes and normalizes the video frames using OpenCV. The input is real-time video data, and the output is preprocessed image data.

[0188] Step 3:

[0189] Behavioral pattern analysis:

[0190] Using the preprocessed image data, the server uses a deep learning model to analyze the children's behavioral patterns. It uses TensorFlow to identify specific movements and activities. The input is the preprocessed image data, and the output is the behavioral pattern data resulting from the analysis.

[0191] Step 4:

[0192] Anomalous behavior detection:

[0193] Based on the results of the behavioral pattern analysis, the server detects abnormal behavior. If the behavioral pattern predicted by the deep learning model is determined to be abnormal, it flags the behavior as abnormal. The input is behavioral pattern data, and the output is an abnormal behavior flag.

[0194] Step 5:

[0195] Sending an alert:

[0196] When abnormal behavior is detected, the server uses smtplib to send an alert to parents or caregivers. Specifically, it sends an alert message to a specified email address. It also uses a push notification service to send notifications in real time to smartphones. The input is an abnormal behavior flag, and the output is an alert message and a push notification.

[0197] Step 6:

[0198] Optimize childcare planning:

[0199] The server analyzes behavioral data accumulated over the long term and provides information useful for optimizing childcare plans. Specifically, it statistically processes behavioral patterns and the frequency of abnormal behavior, and provides feedback to caregivers. The input is the accumulated behavioral data, and the output is information about an optimized childcare plan.

[0200] Through these steps, this system can ensure the safety of children in childcare facilities and at home, and reduce the burden on parents and caregivers.

[0201] 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.

[0202] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected. In addition, by incorporating an emotion engine, the system recognizes user emotions and improves the accuracy of abnormal behavior detection.

[0203] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[0204] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[0205] The server device applies an algorithm to detect abnormal behavior based on the results of behavioral pattern analysis. For example, abnormal behavior such as falls, sudden collisions, and excessive excitement can be detected in real time. Furthermore, the server device is equipped with an emotion engine that can recognize the user's emotions from video and audio data. This emotion data is used as feedback to improve the accuracy of abnormal behavior detection.

[0206] When abnormal behavior is detected, the server device generates an alert message containing a timestamp and location information. The alert message includes specific details of the abnormal behavior and the level of urgency. The generated alert message is sent to the device of the parent or caregiver. The message is sent via push notification, SMS, email, etc., allowing caregivers to respond quickly.

[0207] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral and emotional data, it can provide information to help optimize childcare plans. This allows caregivers to make data-based decisions and improve the quality of childcare.

[0208] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one child falls off the playground equipment, the server device instantly detects this abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the comprehensive analysis results, an alert is sent to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[0209] In this way, this system provides an effective means of reducing the burden on caregivers and improving the quality of childcare while ensuring the safety of children. The introduction of an emotion engine further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[0210] The processing flow will be explained below.

[0211] Step 1:

[0212] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[0213] Step 2:

[0214] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[0215] Step 3:

[0216] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[0217] Step 4:

[0218] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[0219] Step 5:

[0220] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[0221] Step 6:

[0222] When abnormal behavior is detected, the server activates the emotion engine, which analyzes the facial expressions and voices of the children and caregivers around the child to assess the level of urgency.

[0223] Step 7:

[0224] Based on the evaluation results from the emotion engine, the server generates an alert message, which includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[0225] Step 8:

[0226] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[0227] Step 9:

[0228] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[0229] Step 10:

[0230] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[0231] Step 11:

[0232] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns, abnormal behavior, and emotional data. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[0233] The above are the specific processing steps and details of each operation of the AI ​​support and care system incorporating an emotion engine. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers. The introduction of the emotion engine also further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[0234] Example 2

[0235] 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."

[0236] While there are systems in childcare facilities that ensure the safety of children while reducing the burden on caregivers, conventional systems lack sufficient accuracy in detecting abnormal behavior or dangerous situations. In particular, they are unable to detect abnormal behavior while taking into account the children's emotions, making it difficult to respond quickly and appropriately. Furthermore, a system that can monitor 24 hours a day, 365 days a year is required, but achieving this requires a high-performance monitoring system, which poses challenges in terms of cost and operation.

[0237] 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.

[0238] In this invention, the server includes means for acquiring data from multiple image capture devices installed in the childcare facility, means for transmitting the data to a processing device in real time, means for the processing device to analyze the data and identify the behavioral patterns of the monitored person, means for the processing device to detect abnormal behavior or dangerous situations, means for issuing an alert to the monitored person when the abnormal behavior or dangerous situation is detected, means for the processing device to continuously monitor the data 24 hours a day, 365 days a year, means for providing information to optimize the monitoring target's plan based on the analysis results, and means for identifying the monitored person's emotions from the data and improving the accuracy of detecting abnormal behavior. This makes it possible to ensure the safety of children, reduce the burden on caregivers, and improve the accuracy of detecting abnormal behavior.

[0239] A "camera" is a device that is installed in each room and play area within a childcare facility to capture the behavior of the person being monitored.

[0240] "Data" refers to the video and audio information captured by the imaging device.

[0241] A "processing device" is a device for receiving and analyzing data and detecting anomalous behavior.

[0242] The "Real Time Streaming Protocol" is a protocol for transmitting data with minimal delay.

[0243] "Behavioral patterns" are mental categories that refer to specific movements or activities of the monitored person (e.g., running, playing, resting).

[0244] "Abnormal behavior" refers to movements or activities that deviate from normal patterns of behavior, such as sudden movements or excessive excitement.

[0245] An "alert" is an emergency message sent to a target when abnormal behavior or a dangerous situation is detected.

[0246] "Information for optimizing childcare plans" is information that includes suggestions and methods for improving the quality of childcare, based on the results of continuous data analysis.

[0247] "Emotion" refers to the psychological state of the monitored person as identified from the video and audio data.

[0248] "Continuous monitoring 24 hours a day, 365 days a year" means constantly monitoring and analyzing data 365 days a year.

[0249] MODE FOR CARRYING OUT THE INVENTION

[0250] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. This system uses multiple image capture devices to monitor children's behavior and analyzes the data with a processing device to detect abnormal behavior and dangerous situations in real time. In addition, the system improves the accuracy of abnormal behavior detection by identifying emotional data.

[0251] Hardware and Software Configuration

[0252] Imaging device

[0253] The devices used are multiple camera devices installed in each room and play area of ​​the childcare facility. Specifically, network cameras are an example. These camera devices capture children's behavior 24 hours a day, 365 days a year, and transmit the data to a processing device in real time.

[0254] Processing equipment

[0255] The server is equipped with a high-performance processor to receive and analyze data sent in real time, such as a cloud-based server (e.g., AWS EC2 instance) or an on-premise high-performance server (e.g., Dell PowerEdge series).

[0256] Data analysis

[0257] Deep learning models based on TensorFlow and PyTorch are used for data analysis. These models run on a server and identify children's behavioral patterns. Microsoft's Emotion API is used as the emotion engine, implementing algorithms to recognize user emotions from video and audio data.

[0258] Acquiring and Sending Data

[0259] The camera records the children's behavior and transmits the data to the processing unit using a real-time streaming protocol, which minimizes data latency.

[0260] Data analysis

[0261] The server analyzes the received data in real time and uses deep learning models and an emotion engine to identify children's behavioral patterns and emotions, instantly detecting abnormal behavior and dangerous situations.

[0262] Sending an alert

[0263] If abnormal behavior is detected, the server generates an alert message containing a timestamp and location information and sends it to the parent or caregiver's device via push notification, SMS, email, etc.

[0264] Optimizing childcare plans

[0265] The server accumulates data over the long term and provides information to help optimize childcare plans based on the analysis results, allowing childcare workers to make data-based decisions and improve the quality of childcare.

[0266] Specific examples

[0267] At 9:00 AM, a network camera installed in the playroom of a childcare facility captures children playing and sends the data to a server in real time. The server analyzes the data using a TensorFlow-based deep learning model to ensure that the children are playing normally. However, if one child falls off the playground equipment, the server instantly detects this abnormal behavior. It also uses the Emotion API to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the results of this analysis, an alert is sent to the childcare worker's smartphone via push notification. The childcare worker receives the alert and can respond promptly.

[0268] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0269] Step 1:

[0270] Video data capture

[0271] Terminals: Terminals are multiple camera devices installed in each room and play area of ​​the childcare facility. The terminals capture children's behavior 24 hours a day, 365 days a year. The inputs include real-time video and audio of the children. The terminals capture this video and audio data and store it in digital format. The output is the captured video and audio data.

[0272] Specific operation: The camera records video and audio from within the room in real time.

[0273] Step 2:

[0274] Video data transmission

[0275] Terminal: The terminal transmits the captured video and audio data to the processing unit in real time. The input is the digital video and audio data generated in step 1. Using the Real Time Streaming Protocol (RTSP), a live data stream is generated that is sent to the processing unit as output.

[0276] Specific operation: The terminal transmits real-time streaming data to the processing device.

[0277] Step 3:

[0278] Receiving data

[0279] Server: The server receives the video and audio data sent from the device in real time. The input is the live data stream sent in step 2. The server receives this data and temporarily stores it in memory. The output is the video and audio data stored in memory.

[0280] Specific operation: The server receives the data stream in real time and stores it in memory.

[0281] Step 4:

[0282] Behavioral pattern analysis

[0283] Server: The server uses a deep learning model to analyze the video and audio data stored in memory. The input is the data stored in step 3. The deep learning model uses TensorFlow and PyTorch to identify children's behavioral patterns from the data. The output is the identified behavioral patterns, such as "running," "playing," and "resting."

[0284] Specific operation: The server applies a deep learning model to analyze behavioral patterns.

[0285] Step 5:

[0286] Emotional Data Analysis

[0287] Server: The server uses an emotion engine to identify the user's emotions from the video and audio data. The input is the data saved in step 3. The emotion engine can be, for example, Microsoft's Azure Emotion API. The output is the identified emotion data.

[0288] Specific operation: The server uses the emotion engine to identify the user's emotion.

[0289] Step 6:

[0290] Abnormal behavior detection

[0291] Server: The server detects abnormal behaviors and dangerous situations based on behavioral patterns and emotional data. The inputs are the behavioral patterns identified in step 4 and the emotional data identified in step 5. Using a specific algorithm, it identifies sudden movements, excessive excitement, etc. as abnormal behaviors. The output is information about the detected abnormal behaviors.

[0292] Specific operation: The server integrates the analysis results and detects abnormal behavior.

[0293] Step 7:

[0294] Generate and send alerts

[0295] Server: The server generates an alert message containing a timestamp and location information based on the detected abnormal behavior information. The input is the abnormal behavior information obtained in step 6. The alert message contains the details of the abnormal behavior and its urgency. This is sent to the parent or caregiver's device via push notification, SMS, email, etc. The output is the sent alert message.

[0296] Specific operation: The server generates an alert message and sends it to the target device.

[0297] (Application example 2)

[0298] 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."

[0299] Childcare facilities are required to ensure the safety of children while reducing the burden on caregivers. However, conventional systems have low accuracy in detecting abnormal behavior and dangerous situations, making it difficult to respond quickly. In addition, because changes in emotions are not reflected in behavior analysis, there is a delay in determining the level of urgency. A system that can solve these problems, detect abnormal behavior with higher accuracy, and respond quickly is needed.

[0300] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0301] In this invention, the server includes means for acquiring video data from multiple video capture devices installed in the childcare facility, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for issuing an alert to parents or caregivers when abnormal behavior or a dangerous situation is detected, means for continuously monitoring the video data 24 hours a day, 365 days a year, means for recognizing user emotions using an emotion engine when detecting abnormal behavior and improving the accuracy of abnormal behavior detection, and means for providing information for optimizing childcare plans based on the analysis results, thereby enabling abnormal behavior to be detected with high accuracy and responding promptly and appropriately.

[0302] A "video capture device" is a camera device installed to capture visual information.

[0303] "Video data" is a collection of visual information recorded by a video capture device.

[0304] A "server device" is a computer system that processes and stores data over a network and communicates with multiple devices.

[0305] "Means for transmitting in real time" refers to means that has the function of transferring data sequentially with minimal delay.

[0306] A "behavioral pattern" refers to a consistent pattern or tendency of behavior of a particular person or group.

[0307] "Abnormal behavior" refers to movements or actions that are different from normal, such as sudden movements or falls.

[0308] A "hazardous situation" is a condition or situation that presents a high risk of accident or injury.

[0309] An "alert" is an emergency notification issued to warn of an abnormality or danger.

[0310] The "emotion engine" is an algorithm that analyzes video and audio data to identify a person's emotions.

[0311] A "care plan" is a set of activities and schedules designed to ensure the safe and proper development of children in a childcare facility.

[0312] MODE FOR CARRYING OUT THE INVENTION

[0313] This invention provides a system that ensures the safety of children and customers in childcare facilities and brick-and-mortar stores and reduces the burden on managers and caregivers. This system starts by acquiring video data from multiple video capture devices and transmitting it to a server device in real time.

[0314] The server device is implemented using a programming language such as Python and uses deep learning libraries such as TensorFlow and Keras to analyze the video data. This analysis uses a deep learning model to detect abnormal behavior and dangerous situations. This model is capable of identifying behavioral patterns obtained from the video and detecting abnormal behavior.

[0315] The server device is equipped with an emotion engine that can recognize the user's emotions from the video and audio data. This emotion engine distinguishes between emotions such as surprise, sadness, and excitement, improving the accuracy of detecting abnormal behavior.

[0316] Furthermore, if abnormal behavior or a dangerous situation is detected, the server device will issue an alert. This alert will be sent to parents or caregivers via push notification, SMS, or email. The server will include the details of the abnormal behavior, the urgency level, a timestamp, and location information in the alert message, enabling a prompt response.

[0317] This system monitors video data 24 hours a day, 365 days a year, providing data that can be used to optimize childcare plans and store management. For example, the accumulation and analysis of long-term behavioral and emotional data is expected to improve the operation of childcare facilities and increase customer satisfaction.

[0318] Specific examples

[0319] Here is a specific example. First, a video camera installed in a childcare facility's playroom captures children playing and sends the video data in real time to a server device. The server device analyzes this video data to confirm that the children are playing normally. However, if one child falls off the playground equipment, the server device immediately detects the abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. An alarm is then sent to the childcare worker's smartphone, allowing the childcare worker to respond immediately.

[0320] As an example of application in a physical store, consider the case where a customer suddenly falls in the store. At that time, a surveillance camera captures the customer's behavior, and the video data is sent to a server in real time. If the deep learning model analyzes the behavior and determines that there is an abnormality, an alert is immediately sent to the store manager based on that information. Furthermore, the emotion engine analyzes the facial expressions and voices of other customers in the store to assess the urgency of the situation.

[0321] Prompt Sentence Examples

[0322] Below is an example of a prompt sentence to input to the generative AI model.

[0323] Your goal is to create a system that detects abnormal behavior in real time from in-store surveillance footage and sends immediate alerts. The technology stack used is TensorFlow, Keras, OpenCV, and Python. Please also implement a specific alert mechanism.

[0324]

[0325] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0326] Step 1:

[0327] A video capture device acquires multiple visual information: the input is the real-time video data captured by the camera, and the output is the raw video data sent from the camera.

[0328] Step 2:

[0329] Video data captured by a video capture device is sent to a server device in real time. The input is video data from the video capture device, and the output is a video stream sent to the server via a network. The Real Time Streaming Protocol (RTSP) is used to minimize delays.

[0330] Step 3:

[0331] The server device receives the video data and begins analysis. The input is the video data sent to the server, and the output is the analyzed behavioral pattern data. The behavioral patterns are identified using deep learning models (TensorFlow, Keras).

[0332] Step 4:

[0333] The server device detects abnormal behavior and dangerous situations based on the identified behavior patterns. The input is behavior pattern data, and the output is a judgment result of abnormal or normal behavior. The analysis is performed using a deep learning algorithm.

[0334] Step 5:

[0335] The emotion engine analyzes video and audio data to recognize the user's emotions. The input is video and audio data, and the output is a judgment result of the user's emotional state (surprise, sadness, excitement, etc.). This improves the accuracy of detecting abnormal behavior.

[0336] Step 6:

[0337] When the server device detects abnormal behavior or a dangerous situation, it generates an alert. The input is the abnormal behavior judgment result and the emotional state, and the output is an alert message. The alert message includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[0338] Step 7:

[0339] Sends alerts to parents' or caregivers' devices. The input is the alert message, and the output is a notification sent in the form of push notification, SMS, email, etc. This allows parents or caregivers to respond quickly.

[0340] Step 8:

[0341] The server device continuously monitors video data 24 hours a day, 365 days a year, and stores the necessary data. The input is real-time video data, and the output is a database of long-term behavioral patterns and emotional states.

[0342] Step 9:

[0343] The accumulated data is analyzed to provide information for optimizing childcare plans and store management. The input is the accumulated behavioral patterns and emotional state data, and the output is an optimized childcare plan and store management proposal.

[0344] 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.

[0345] 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.

[0346] 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.

[0347] [Second embodiment]

[0348] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0349] 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.

[0350] 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).

[0351] 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.

[0352] 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.

[0353] 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).

[0354] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0355] 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.

[0356] 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.

[0357] 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.

[0358] 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.

[0359] 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."

[0360] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected.

[0361] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[0362] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[0363] The server device detects abnormal behavior based on the results of behavioral pattern analysis. Abnormal behavior includes falls, sudden collisions, and excessive excitement. If abnormal behavior is detected, the server device immediately generates an alert and sends it to the device of the parent or caregiver. The alert is sent in the form of a push notification, SMS, email, or other format, allowing caregivers to respond quickly.

[0364] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral data, it can provide information to help optimize childcare plans. This allows childcare workers to make data-based decisions and improve the quality of childcare.

[0365] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server device instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker will immediately receive the alert and be able to respond promptly.

[0366] In this way, this system provides an effective means of ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[0367] The processing flow will be explained below.

[0368] Step 1:

[0369] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[0370] Step 2:

[0371] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[0372] Step 3:

[0373] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[0374] Step 4:

[0375] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[0376] Step 5:

[0377] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[0378] Step 6:

[0379] If an abnormal behavior is detected, the server generates an alert message including a timestamp and location information, as well as the specific details of the abnormal behavior and its urgency.

[0380] Step 7:

[0381] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[0382] Step 8:

[0383] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[0384] Step 9:

[0385] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[0386] Step 10:

[0387] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns and abnormal behavior. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[0388] The above are the specific processing steps and details of each operation of the AI ​​support and care system. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers.

[0389] Example 1

[0390] 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."

[0391] Conventional safety management systems in childcare facilities lacked the means to monitor children's daily behavior 24 hours a day, 365 days a year. Furthermore, alarm systems to quickly detect and respond to abnormal behavior or dangerous situations were also ineffective. This increased the burden on childcare workers, and led to problems such as insufficient efforts to ensure the safety of children.

[0392] 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.

[0393] In this invention, the server includes a means for temporarily storing video data in a buffer and checking for missing data and compression status, a means for identifying children's behavioral patterns from the video data after preprocessing using a deep learning model, and a means for comparing the identified behavioral patterns with set abnormal behavior conditions to detect abnormal behavior or dangerous situations. This enables real-time monitoring of children's behavior 24 hours a day, 365 days a year, quickly detecting abnormal behavior or dangerous situations, and issuing an alarm. This allows parents and caregivers to respond quickly, effectively ensuring children's safety and reducing the burden on caregivers. It can also provide information to support the optimization of childcare plans based on the accumulated behavioral data.

[0394] A "video recording device" is a device installed in a childcare facility to capture the daily activities of children.

[0395] "Video data" refers to data that includes video information of children's daily activities captured by a video camera.

[0396] The "server device" is a central processing unit for receiving and analyzing video data transmitted from multiple video capture devices installed in the childcare facility.

[0397] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time, and is a technology that minimizes data delays.

[0398] The "deep learning model" is an advanced machine learning model used to identify children's behavioral patterns from received video data and detect abnormal behavior.

[0399] A "behavioral pattern" is a sequence of specific movements or activities that children exhibit within a childcare facility.

[0400] "Abnormal behavior" refers to behavior that deviates from children's normal patterns of behavior and may lead directly to danger or crisis.

[0401] An "alert" is a warning message sent to parents or caregivers when abnormal behavior or a dangerous situation is detected.

[0402] A "push notification" is a real-time alert notification message sent directly from a server device to the device of a parent or caregiver.

[0403] "Optimizing childcare plans" refers to analyzing accumulated behavioral data to improve the quality of childcare and provide information to plan and execute more efficient childcare activities.

[0404] The present invention is a system for ensuring the safety of children in childcare facilities and reducing the burden on caregivers, and is composed of multiple video capture devices, a server device, and a notification system. Specific embodiments for carrying out the invention are described below.

[0405] Installation of video recording equipment and data acquisition

[0406] Device:

[0407] Video recording devices are installed in each room and play area within the childcare facility. 360-degree cameras or high-resolution cameras can be used for these. The video recording devices capture the children's daily activities 24 hours a day, 365 days a year, and transmit the video data to a server device in real time. A real-time streaming protocol (such as RTSP) is used to transmit the video data.

[0408] Server device data reception and analysis

[0409] server:

[0410] The server temporarily stores the received video data in a buffer and then analyzes it using a deep learning model, performing preprocessing such as color space conversion and frame resizing.

[0411] As a specific example, the server uses deep learning models such as ResNet and YOLO to identify children's behavioral patterns, including "running," "playing," and "resting."

[0412] Detecting abnormal behavior and generating alerts

[0413] server:

[0414] The server device compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior or dangerous situations, such as when a child falls off playground equipment or has a sudden collision.

[0415] When abnormal behavior is detected, an alert is immediately generated, containing information about the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[0416] Notification to parents and caregivers

[0417] server:

[0418] Alerts are sent to parents' and caregivers' devices in the form of push notifications, SMS, emails, etc. For example, an alert such as "A child has fallen off play equipment in the playroom" is immediately sent to a caregiver's smartphone.

[0419] Data accumulation and support for optimizing childcare plans

[0420] server:

[0421] The server device stores daily captured video data and analysis results over the long term, making it possible to track long-term changes in behavioral patterns. Based on the stored data, it can provide information to help optimize childcare plans. For example, it can provide statistical information on the activities of children at specific times of the day.

[0422] Specific examples

[0423] At 9:00 a.m., a video camera installed in the playroom of a childcare facility captures children playing and sends the video data to a server in real time. The server analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[0424] Prompt Sentence Examples

[0425] "Please explain the surveillance system in your childcare facility. This system uses video cameras to monitor children's behavior in real time and issues an alert if it detects abnormal behavior, such as falls or sudden bumps. The system also has 24 / 7 monitoring capabilities and analyzes the accumulated data to help optimize childcare planning."

[0426] In this way, the present invention provides a specific system for ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[0427] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0428] Program processing steps

[0429] Step 1:

[0430] Video data capture and transmission

[0431] Devices: Video cameras installed in each room and play area of ​​the childcare facility capture the children's daily activities 24 hours a day, 365 days a year. For example, they capture footage of children playing in the playroom at 9:00 AM.

[0432] Input: Children's daily activities

[0433] Output: Real-time video data

[0434] How it works: The video capture device is equipped with a 360-degree camera that captures children's movements from multiple angles. The captured video data is encrypted and sent to a server using a real-time streaming protocol (such as RTSP).

[0435] Step 2:

[0436] Data reception and primary analysis by the server

[0437] Server: The server temporarily stores the received video data in a buffer and checks for any data loss or compression issues.

[0438] Input: Real-time video data

[0439] Output: Pre-processed video data

[0440] Specific operation: The server checks the frame rate and resolution, checks for missing data, and then performs preprocessing such as color space conversion and frame resizing.

[0441] Step 3:

[0442] Behavioral pattern analysis

[0443] Server: The preprocessed video data is input into a deep learning model to identify children's behavioral patterns. Deep learning models such as ResNet and YOLO are used.

[0444] Input: Preprocessed video data

[0445] Output: Identified behavioral pattern data

[0446] Specific actions: The deep learning model in the server sequentially analyzes actions such as "running," "playing," and "resting" from video data.

[0447] Step 4:

[0448] Abnormal behavior detection

[0449] Server: Compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior and dangerous situations.

[0450] Input: Identified behavioral pattern data

[0451] Output: Abnormal behavior detection alert

[0452] Specific behavior: For example, the server detects a child falling off a playground equipment and flags it as abnormal behavior.

[0453] Step 5:

[0454] Alert generation and notification

[0455] Server: When abnormal behavior is detected, an alert is immediately generated and sent to the device of the parent or caregiver. The alert includes information such as the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[0456] Input: Abnormal behavior detection alert

[0457] Output: Warning message

[0458] Specific operation: The server generates an alert message stating, "A child has fallen off the playground equipment in the playroom," and sends it to the caregiver's smartphone in the form of a push notification, SMS, email, etc.

[0459] Step 6:

[0460] Data accumulation and support for optimizing childcare plans

[0461] Server: Stores daily captured video data and analysis results over the long term. Based on the stored data, it provides information to help optimize childcare plans.

[0462] Input: Identified behavioral pattern data and alarm data

[0463] Output: Statistics for optimizing childcare plans

[0464] Specific operation: The server generates statistics on the activities of children during specific time periods, which is used to review and improve childcare plans.

[0465] The above is the specific flow of program processing for this system.

[0466] (Application example 1)

[0467] 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."

[0468] While there are systems in childcare facilities that safely monitor children's behavior and quickly detect abnormal behavior or dangerous situations and notify parents and caregivers, similar functions are required in home and small childcare environments. There is also a need to provide a simple means of ensuring children's safety using the smartphones of parents and caregivers.

[0469] 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.

[0470] In this invention, the server includes means for acquiring video data from multiple video capture devices, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for the server device to detect abnormal behavior or dangerous situations, means for issuing an alert to parents or caregivers when the abnormal behavior or dangerous situation is detected, means for the server device to continuously monitor the video data 24 hours a day, 365 days a year, means for providing information to optimize childcare plans based on the analysis results, and means for detecting abnormal behavior and sending push notifications to smartphone applications used by families and certified childcare providers. This makes it possible to ensure the safety of children even in home or small childcare environments.

[0471] A "server device" is a central processing unit that analyzes video data over a network and issues an alarm based on the results.

[0472] A "video recording device" is a camera device that is installed in a facility or home to capture the behavior of children.

[0473] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time without delay.

[0474] A "deep learning model" is a machine learning model that analyzes children's behavioral patterns and detects abnormal behavior.

[0475] "Abnormal behavior" refers to behavior that is dangerous for children, such as falling or sudden collisions.

[0476] "Push notifications" are real-time notifications sent to users' smartphones when abnormal behavior is detected.

[0477] A "childcare plan" is a plan for childcare activities that is optimized based on children's behavioral data.

[0478] The "smartphone application used by families and certified childcare mothers" is a mobile application that monitors children's behavior at home and receives notifications when abnormal behavior is detected.

[0479] An "alert" is an alert issued when abnormal behavior or a dangerous situation is detected.

[0480] "Means for continuously monitoring the video data 24 hours a day, 365 days a year" refers to a function that continuously monitors children's behavior, regardless of day or night.

[0481] The present invention is a system for monitoring the behavior of children in childcare facilities and at home to ensure their safety. The system is configured as follows.

[0482] Hardware and software configuration

[0483] Hardware

[0484] Video recording equipment: Cameras installed in childcare facilities and homes monitor children's behavior 24 hours a day, 365 days a year.

[0485] Server device: A central processing unit that receives and analyzes video data. The server runs at all times and continuously monitors data.

[0486] Smartphone: A mobile device used by parents and certified childcare providers. Push notifications are sent when abnormal behavior is detected.

[0487] software

[0488] OpenCV: A library for preprocessing video data and image processing.

[0489] TensorFlow: A library that uses deep learning to analyze children's behavioral patterns and detect abnormal behavior.

[0490] smtplib: A Python module for sending email notifications when anomalous behavior is detected.

[0491] Data processing and calculation

[0492] 1. Camera input: The video camera continuously captures the children's behavior and transmits the video data to the server in real time using the real-time streaming protocol.

[0493] 2. Image preprocessing: The server preprocesses the received video frames to make them easier for the deep learning model to process. OpenCV is used for this preprocessing.

[0494] 3. Behavioral Analysis: Using a trained deep learning model, the server analyzes children's behavioral patterns and uses TensorFlow to identify specific movements and activities.

[0495] 4. Abnormal behavior detection: Abnormal behavior is detected based on the behavioral pattern analysis. If abnormal behavior is detected, an alert is sent to the parents using smtplib.

[0496] 5. Push notification: When abnormal behavior is detected, a push notification will be sent to the smartphone, allowing parents or certified childcare providers to take immediate action.

[0497] 6. Optimization of childcare plans: Through long-term data accumulation, the server provides information that is useful for optimizing childcare plans.

[0498] Specific examples

[0499] For example, if a camera installed in a home monitors a child's behavior and the child suddenly runs off and falls, the server will instantly detect this abnormal behavior and send a push notification to the parent's smartphone, allowing the parent to quickly rush to the scene.

[0500] Prompt Sentence Examples

[0501] "I would like to develop a real-time behavior analysis system using a camera. I need a function that detects abnormal behavior in children and notifies them by email. Please provide me with Python code that uses OpenCV and TensorFlow."

[0502] As described above, the present invention is a system that ensures the safety of children in childcare facilities and homes and reduces the burden on parents and caregivers.

[0503] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0504] Step 1:

[0505] Video data acquisition and transmission:

[0506] The video camera is installed in a childcare facility or a home and captures children's behavior 24 hours a day, 365 days a year. The captured video data is sent to a server using a real-time streaming protocol. The input is a video frame, and the output is real-time video data sent to the server.

[0507] Step 2:

[0508] Image preprocessing:

[0509] The server preprocesses the received video frames to make them easier for the deep learning model to process. Specifically, it resizes and normalizes the video frames using OpenCV. The input is real-time video data, and the output is preprocessed image data.

[0510] Step 3:

[0511] Behavioral pattern analysis:

[0512] Using the preprocessed image data, the server uses a deep learning model to analyze the children's behavioral patterns. It uses TensorFlow to identify specific movements and activities. The input is the preprocessed image data, and the output is the behavioral pattern data resulting from the analysis.

[0513] Step 4:

[0514] Anomalous behavior detection:

[0515] Based on the results of the behavioral pattern analysis, the server detects abnormal behavior. If the behavioral pattern predicted by the deep learning model is determined to be abnormal, it flags the behavior as abnormal. The input is behavioral pattern data, and the output is an abnormal behavior flag.

[0516] Step 5:

[0517] Sending an alert:

[0518] When abnormal behavior is detected, the server uses smtplib to send an alert to parents or caregivers. Specifically, it sends an alert message to a specified email address. It also uses a push notification service to send notifications in real time to smartphones. The input is an abnormal behavior flag, and the output is an alert message and a push notification.

[0519] Step 6:

[0520] Optimize childcare planning:

[0521] The server analyzes behavioral data accumulated over the long term and provides information useful for optimizing childcare plans. Specifically, it statistically processes behavioral patterns and the frequency of abnormal behavior, and provides feedback to caregivers. The input is the accumulated behavioral data, and the output is information about an optimized childcare plan.

[0522] Through these steps, this system can ensure the safety of children in childcare facilities and at home, and reduce the burden on parents and caregivers.

[0523] 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.

[0524] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected. In addition, by incorporating an emotion engine, the system recognizes user emotions and improves the accuracy of abnormal behavior detection.

[0525] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[0526] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[0527] The server device applies an algorithm to detect abnormal behavior based on the results of behavioral pattern analysis. For example, abnormal behavior such as falls, sudden collisions, and excessive excitement can be detected in real time. Furthermore, the server device is equipped with an emotion engine that can recognize the user's emotions from video and audio data. This emotion data is used as feedback to improve the accuracy of abnormal behavior detection.

[0528] When abnormal behavior is detected, the server device generates an alert message containing a timestamp and location information. The alert message includes specific details of the abnormal behavior and the level of urgency. The generated alert message is sent to the device of the parent or caregiver. The message is sent via push notification, SMS, email, etc., allowing caregivers to respond quickly.

[0529] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral and emotional data, it can provide information to help optimize childcare plans. This allows caregivers to make data-based decisions and improve the quality of childcare.

[0530] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one child falls off the playground equipment, the server device instantly detects this abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the comprehensive analysis results, an alert is sent to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[0531] In this way, this system provides an effective means of reducing the burden on caregivers and improving the quality of childcare while ensuring the safety of children. The introduction of an emotion engine further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[0532] The processing flow will be explained below.

[0533] Step 1:

[0534] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[0535] Step 2:

[0536] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[0537] Step 3:

[0538] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[0539] Step 4:

[0540] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[0541] Step 5:

[0542] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[0543] Step 6:

[0544] When abnormal behavior is detected, the server activates the emotion engine, which analyzes the facial expressions and voices of the children and caregivers around the child to assess the level of urgency.

[0545] Step 7:

[0546] Based on the evaluation results from the emotion engine, the server generates an alert message, which includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[0547] Step 8:

[0548] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[0549] Step 9:

[0550] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[0551] Step 10:

[0552] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[0553] Step 11:

[0554] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns, abnormal behavior, and emotional data. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[0555] The above are the specific processing steps and details of each operation of the AI ​​support and care system incorporating an emotion engine. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers. The introduction of the emotion engine also further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[0556] Example 2

[0557] 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."

[0558] While there are systems in childcare facilities that ensure the safety of children while reducing the burden on caregivers, conventional systems lack sufficient accuracy in detecting abnormal behavior or dangerous situations. In particular, they are unable to detect abnormal behavior while taking into account the children's emotions, making it difficult to respond quickly and appropriately. Furthermore, a system that can monitor 24 hours a day, 365 days a year is required, but achieving this requires a high-performance monitoring system, which poses challenges in terms of cost and operation.

[0559] 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.

[0560] In this invention, the server includes means for acquiring data from multiple image capture devices installed in the childcare facility, means for transmitting the data to a processing device in real time, means for the processing device to analyze the data and identify the behavioral patterns of the monitored person, means for the processing device to detect abnormal behavior or dangerous situations, means for issuing an alert to the monitored person when the abnormal behavior or dangerous situation is detected, means for the processing device to continuously monitor the data 24 hours a day, 365 days a year, means for providing information to optimize the monitoring target's plan based on the analysis results, and means for identifying the monitored person's emotions from the data and improving the accuracy of detecting abnormal behavior. This makes it possible to ensure the safety of children, reduce the burden on caregivers, and improve the accuracy of detecting abnormal behavior.

[0561] A "camera" is a device that is installed in each room and play area within a childcare facility to capture the behavior of the person being monitored.

[0562] "Data" refers to the video and audio information captured by the imaging device.

[0563] A "processing device" is a device for receiving and analyzing data and detecting anomalous behavior.

[0564] The "Real Time Streaming Protocol" is a protocol for transmitting data with minimal delay.

[0565] "Behavioral patterns" are mental categories that refer to specific movements or activities of the monitored person (e.g., running, playing, resting).

[0566] "Abnormal behavior" refers to movements or activities that deviate from normal patterns of behavior, such as sudden movements or excessive excitement.

[0567] An "alert" is an emergency message sent to a target when abnormal behavior or a dangerous situation is detected.

[0568] "Information for optimizing childcare plans" is information that includes suggestions and methods for improving the quality of childcare, based on the results of continuous data analysis.

[0569] "Emotion" refers to the psychological state of the monitored person as identified from the video and audio data.

[0570] "Continuous monitoring 24 hours a day, 365 days a year" means constantly monitoring and analyzing data 365 days a year.

[0571] MODE FOR CARRYING OUT THE INVENTION

[0572] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. This system uses multiple image capture devices to monitor children's behavior and analyzes the data with a processing device to detect abnormal behavior and dangerous situations in real time. In addition, the system improves the accuracy of abnormal behavior detection by identifying emotional data.

[0573] Hardware and Software Configuration

[0574] Imaging device

[0575] The devices used are multiple camera devices installed in each room and play area of ​​the childcare facility. Specifically, network cameras are an example. These camera devices capture children's behavior 24 hours a day, 365 days a year, and transmit the data to a processing device in real time.

[0576] Processing equipment

[0577] The server is equipped with a high-performance processor to receive and analyze data sent in real time, such as a cloud-based server (e.g., AWS EC2 instance) or an on-premise high-performance server (e.g., Dell PowerEdge series).

[0578] Data analysis

[0579] Deep learning models based on TensorFlow and PyTorch are used for data analysis. These models run on a server and identify children's behavioral patterns. Microsoft's Emotion API is used as the emotion engine, implementing algorithms to recognize user emotions from video and audio data.

[0580] Acquiring and Sending Data

[0581] The camera records the children's behavior and transmits the data to the processing unit using a real-time streaming protocol, which minimizes data latency.

[0582] Data analysis

[0583] The server analyzes the received data in real time and uses deep learning models and an emotion engine to identify children's behavioral patterns and emotions, instantly detecting abnormal behavior and dangerous situations.

[0584] Sending an alert

[0585] If abnormal behavior is detected, the server generates an alert message containing a timestamp and location information and sends it to the parent or caregiver's device via push notification, SMS, email, etc.

[0586] Optimizing childcare plans

[0587] The server accumulates data over the long term and provides information to help optimize childcare plans based on the analysis results, allowing childcare workers to make data-based decisions and improve the quality of childcare.

[0588] Specific examples

[0589] At 9:00 AM, a network camera installed in the playroom of a childcare facility captures children playing and sends the data to a server in real time. The server analyzes the data using a TensorFlow-based deep learning model to ensure that the children are playing normally. However, if one child falls off the playground equipment, the server instantly detects this abnormal behavior. It also uses the Emotion API to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the results of this analysis, an alert is sent to the childcare worker's smartphone via push notification. The childcare worker receives the alert and can respond promptly.

[0590] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0591] Step 1:

[0592] Video data capture

[0593] Terminals: Terminals are multiple camera devices installed in each room and play area of ​​the childcare facility. The terminals capture children's behavior 24 hours a day, 365 days a year. The inputs include real-time video and audio of the children. The terminals capture this video and audio data and store it in digital format. The output is the captured video and audio data.

[0594] Specific operation: The camera records video and audio from within the room in real time.

[0595] Step 2:

[0596] Video data transmission

[0597] Terminal: The terminal transmits the captured video and audio data to the processing unit in real time. The input is the digital video and audio data generated in step 1. Using the Real Time Streaming Protocol (RTSP), a live data stream is generated that is sent to the processing unit as output.

[0598] Specific operation: The terminal transmits real-time streaming data to the processing device.

[0599] Step 3:

[0600] Receiving data

[0601] Server: The server receives the video and audio data sent from the device in real time. The input is the live data stream sent in step 2. The server receives this data and temporarily stores it in memory. The output is the video and audio data stored in memory.

[0602] Specific operation: The server receives the data stream in real time and stores it in memory.

[0603] Step 4:

[0604] Behavioral pattern analysis

[0605] Server: The server uses a deep learning model to analyze the video and audio data stored in memory. The input is the data stored in step 3. The deep learning model uses TensorFlow and PyTorch to identify children's behavioral patterns from the data. The output is the identified behavioral patterns, such as "running," "playing," and "resting."

[0606] Specific operation: The server applies a deep learning model to analyze behavioral patterns.

[0607] Step 5:

[0608] Emotional Data Analysis

[0609] Server: The server uses an emotion engine to identify the user's emotions from the video and audio data. The input is the data saved in step 3. The emotion engine can be, for example, Microsoft's Azure Emotion API. The output is the identified emotion data.

[0610] Specific operation: The server uses the emotion engine to identify the user's emotion.

[0611] Step 6:

[0612] Abnormal behavior detection

[0613] Server: The server detects abnormal behaviors and dangerous situations based on behavioral patterns and emotional data. The inputs are the behavioral patterns identified in step 4 and the emotional data identified in step 5. Using a specific algorithm, it identifies sudden movements, excessive excitement, etc. as abnormal behaviors. The output is information about the detected abnormal behaviors.

[0614] Specific operation: The server integrates the analysis results and detects abnormal behavior.

[0615] Step 7:

[0616] Generate and send alerts

[0617] Server: The server generates an alert message containing a timestamp and location information based on the detected abnormal behavior information. The input is the abnormal behavior information obtained in step 6. The alert message contains the details of the abnormal behavior and its urgency. This is sent to the parent or caregiver's device via push notification, SMS, email, etc. The output is the sent alert message.

[0618] Specific operation: The server generates an alert message and sends it to the target device.

[0619] (Application example 2)

[0620] 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."

[0621] Childcare facilities are required to ensure the safety of children while reducing the burden on caregivers. However, conventional systems have low accuracy in detecting abnormal behavior and dangerous situations, making it difficult to respond quickly. In addition, because changes in emotions are not reflected in behavior analysis, there is a delay in determining the level of urgency. A system that can solve these problems, detect abnormal behavior with higher accuracy, and respond quickly is needed.

[0622] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0623] In this invention, the server includes means for acquiring video data from multiple video capture devices installed in the childcare facility, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for issuing an alert to parents or caregivers when abnormal behavior or a dangerous situation is detected, means for continuously monitoring the video data 24 hours a day, 365 days a year, means for recognizing user emotions using an emotion engine when detecting abnormal behavior and improving the accuracy of abnormal behavior detection, and means for providing information for optimizing childcare plans based on the analysis results, thereby enabling abnormal behavior to be detected with high accuracy and responding promptly and appropriately.

[0624] A "video capture device" is a camera device installed to capture visual information.

[0625] "Video data" is a collection of visual information recorded by a video capture device.

[0626] A "server device" is a computer system that processes and stores data over a network and communicates with multiple devices.

[0627] "Means for transmitting in real time" refers to means that has the function of transferring data sequentially with minimal delay.

[0628] A "behavioral pattern" refers to a consistent pattern or tendency of behavior of a particular person or group.

[0629] "Abnormal behavior" refers to movements or actions that are different from normal, such as sudden movements or falls.

[0630] A "hazardous situation" is a condition or situation that presents a high risk of accident or injury.

[0631] An "alert" is an emergency notification issued to warn of an abnormality or danger.

[0632] The "emotion engine" is an algorithm that analyzes video and audio data to identify a person's emotions.

[0633] A "care plan" is a set of activities and schedules designed to ensure the safe and proper development of children in a childcare facility.

[0634] MODE FOR CARRYING OUT THE INVENTION

[0635] This invention provides a system that ensures the safety of children and customers in childcare facilities and brick-and-mortar stores and reduces the burden on managers and caregivers. This system starts by acquiring video data from multiple video capture devices and transmitting it to a server device in real time.

[0636] The server device is implemented using a programming language such as Python and uses deep learning libraries such as TensorFlow and Keras to analyze the video data. This analysis uses a deep learning model to detect abnormal behavior and dangerous situations. This model is capable of identifying behavioral patterns obtained from the video and detecting abnormal behavior.

[0637] The server device is equipped with an emotion engine that can recognize the user's emotions from the video and audio data. This emotion engine distinguishes between emotions such as surprise, sadness, and excitement, improving the accuracy of detecting abnormal behavior.

[0638] Furthermore, if abnormal behavior or a dangerous situation is detected, the server device will issue an alert. This alert will be sent to parents or caregivers via push notification, SMS, or email. The server will include the details of the abnormal behavior, the urgency level, a timestamp, and location information in the alert message, enabling a prompt response.

[0639] This system monitors video data 24 hours a day, 365 days a year, providing data that can be used to optimize childcare plans and store management. For example, the accumulation and analysis of long-term behavioral and emotional data is expected to improve the operation of childcare facilities and increase customer satisfaction.

[0640] Specific examples

[0641] Here is a specific example. First, a video camera installed in a childcare facility's playroom captures children playing and sends the video data in real time to a server device. The server device analyzes this video data to confirm that the children are playing normally. However, if one child falls off the playground equipment, the server device immediately detects the abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. An alarm is then sent to the childcare worker's smartphone, allowing the childcare worker to respond immediately.

[0642] As an example of application in a physical store, consider the case where a customer suddenly falls in the store. At that time, a surveillance camera captures the customer's behavior, and the video data is sent to a server in real time. If the deep learning model analyzes the behavior and determines that there is an abnormality, an alert is immediately sent to the store manager based on that information. Furthermore, the emotion engine analyzes the facial expressions and voices of other customers in the store to assess the urgency of the situation.

[0643] Prompt Sentence Examples

[0644] Below is an example of a prompt sentence to input to the generative AI model.

[0645] Your goal is to create a system that detects abnormal behavior in real time from in-store surveillance footage and sends immediate alerts. The technology stack used is TensorFlow, Keras, OpenCV, and Python. Please also implement a specific alert mechanism.

[0646]

[0647] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0648] Step 1:

[0649] A video capture device acquires multiple visual information: the input is the real-time video data captured by the camera, and the output is the raw video data sent from the camera.

[0650] Step 2:

[0651] Video data captured by a video capture device is sent to a server device in real time. The input is video data from the video capture device, and the output is a video stream sent to the server via a network. The Real Time Streaming Protocol (RTSP) is used to minimize delays.

[0652] Step 3:

[0653] The server device receives the video data and begins analysis. The input is the video data sent to the server, and the output is the analyzed behavioral pattern data. The behavioral patterns are identified using deep learning models (TensorFlow, Keras).

[0654] Step 4:

[0655] The server device detects abnormal behavior and dangerous situations based on the identified behavior patterns. The input is behavior pattern data, and the output is a judgment result of abnormal or normal behavior. The analysis is performed using a deep learning algorithm.

[0656] Step 5:

[0657] The emotion engine analyzes video and audio data to recognize the user's emotions. The input is video and audio data, and the output is a judgment result of the user's emotional state (surprise, sadness, excitement, etc.). This improves the accuracy of detecting abnormal behavior.

[0658] Step 6:

[0659] When the server device detects abnormal behavior or a dangerous situation, it generates an alert. The input is the abnormal behavior judgment result and the emotional state, and the output is an alert message. The alert message includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[0660] Step 7:

[0661] Sends alerts to parents' or caregivers' devices. The input is the alert message, and the output is a notification sent in the form of push notification, SMS, email, etc. This allows parents or caregivers to respond quickly.

[0662] Step 8:

[0663] The server device continuously monitors video data 24 hours a day, 365 days a year, and stores the necessary data. The input is real-time video data, and the output is a database of long-term behavioral patterns and emotional states.

[0664] Step 9:

[0665] The accumulated data is analyzed to provide information for optimizing childcare plans and store management. The input is the accumulated behavioral patterns and emotional state data, and the output is an optimized childcare plan and store management proposal.

[0666] 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.

[0667] 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.

[0668] 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.

[0669] [Third embodiment]

[0670] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0671] 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.

[0672] 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).

[0673] 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.

[0674] 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.

[0675] 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).

[0676] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0677] 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.

[0678] 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.

[0679] 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.

[0680] 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.

[0681] 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."

[0682] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected.

[0683] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[0684] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[0685] The server device detects abnormal behavior based on the results of behavioral pattern analysis. Abnormal behavior includes falls, sudden collisions, and excessive excitement. If abnormal behavior is detected, the server device immediately generates an alert and sends it to the device of the parent or caregiver. The alert is sent in the form of a push notification, SMS, email, or other format, allowing caregivers to respond quickly.

[0686] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral data, it can provide information to help optimize childcare plans. This allows childcare workers to make data-based decisions and improve the quality of childcare.

[0687] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server device instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker will immediately receive the alert and be able to respond promptly.

[0688] In this way, this system provides an effective means of ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[0689] The processing flow will be explained below.

[0690] Step 1:

[0691] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[0692] Step 2:

[0693] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[0694] Step 3:

[0695] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[0696] Step 4:

[0697] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[0698] Step 5:

[0699] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[0700] Step 6:

[0701] If an abnormal behavior is detected, the server generates an alert message including a timestamp and location information, as well as the specific details of the abnormal behavior and its urgency.

[0702] Step 7:

[0703] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[0704] Step 8:

[0705] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[0706] Step 9:

[0707] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[0708] Step 10:

[0709] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns and abnormal behavior. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[0710] The above are the specific processing steps and details of each operation of the AI ​​support and care system. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers.

[0711] Example 1

[0712] 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."

[0713] Conventional safety management systems in childcare facilities lacked the means to monitor children's daily behavior 24 hours a day, 365 days a year. Furthermore, alarm systems to quickly detect and respond to abnormal behavior or dangerous situations were also ineffective. This increased the burden on childcare workers, and led to problems such as insufficient efforts to ensure the safety of children.

[0714] 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.

[0715] In this invention, the server includes a means for temporarily storing video data in a buffer and checking for missing data and compression status, a means for identifying children's behavioral patterns from the video data after preprocessing using a deep learning model, and a means for comparing the identified behavioral patterns with set abnormal behavior conditions to detect abnormal behavior or dangerous situations. This enables real-time monitoring of children's behavior 24 hours a day, 365 days a year, quickly detecting abnormal behavior or dangerous situations, and issuing an alarm. This allows parents and caregivers to respond quickly, effectively ensuring children's safety and reducing the burden on caregivers. It can also provide information to support the optimization of childcare plans based on the accumulated behavioral data.

[0716] A "video recording device" is a device installed in a childcare facility to capture the daily activities of children.

[0717] "Video data" refers to data that includes video information of children's daily activities captured by a video camera.

[0718] The "server device" is a central processing unit for receiving and analyzing video data transmitted from multiple video capture devices installed in the childcare facility.

[0719] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time, and is a technology that minimizes data delays.

[0720] The "deep learning model" is an advanced machine learning model used to identify children's behavioral patterns from received video data and detect abnormal behavior.

[0721] A "behavioral pattern" is a sequence of specific movements or activities that children exhibit within a childcare facility.

[0722] "Abnormal behavior" refers to behavior that deviates from children's normal patterns of behavior and may lead directly to danger or crisis.

[0723] An "alert" is a warning message sent to parents or caregivers when abnormal behavior or a dangerous situation is detected.

[0724] A "push notification" is a real-time alert notification message sent directly from a server device to the device of a parent or caregiver.

[0725] "Optimizing childcare plans" refers to analyzing accumulated behavioral data to improve the quality of childcare and provide information to plan and execute more efficient childcare activities.

[0726] The present invention is a system for ensuring the safety of children in childcare facilities and reducing the burden on caregivers, and is composed of multiple video capture devices, a server device, and a notification system. Specific embodiments for carrying out the invention are described below.

[0727] Installation of video recording equipment and data acquisition

[0728] Device:

[0729] Video recording devices are installed in each room and play area within the childcare facility. 360-degree cameras or high-resolution cameras can be used for these. The video recording devices capture the children's daily activities 24 hours a day, 365 days a year, and transmit the video data to a server device in real time. A real-time streaming protocol (such as RTSP) is used to transmit the video data.

[0730] Server device data reception and analysis

[0731] server:

[0732] The server temporarily stores the received video data in a buffer and then analyzes it using a deep learning model, performing preprocessing such as color space conversion and frame resizing.

[0733] As a specific example, the server uses deep learning models such as ResNet and YOLO to identify children's behavioral patterns, including "running," "playing," and "resting."

[0734] Detecting abnormal behavior and generating alerts

[0735] server:

[0736] The server device compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior or dangerous situations, such as when a child falls off playground equipment or has a sudden collision.

[0737] When abnormal behavior is detected, an alert is immediately generated, containing information about the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[0738] Notification to parents and caregivers

[0739] server:

[0740] Alerts are sent to parents' and caregivers' devices in the form of push notifications, SMS, emails, etc. For example, an alert such as "A child has fallen off play equipment in the playroom" is immediately sent to a caregiver's smartphone.

[0741] Data accumulation and support for optimizing childcare plans

[0742] server:

[0743] The server device stores daily captured video data and analysis results over the long term, making it possible to track long-term changes in behavioral patterns. Based on the stored data, it can provide information to help optimize childcare plans. For example, it can provide statistical information on the activities of children at specific times of the day.

[0744] Specific examples

[0745] At 9:00 a.m., a video camera installed in the playroom of a childcare facility captures children playing and sends the video data to a server in real time. The server analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[0746] Prompt Sentence Examples

[0747] "Please explain the surveillance system in your childcare facility. This system uses video cameras to monitor children's behavior in real time and issues an alert if it detects abnormal behavior, such as falls or sudden bumps. The system also has 24 / 7 monitoring capabilities and analyzes the accumulated data to help optimize childcare planning."

[0748] In this way, the present invention provides a specific system for ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[0749] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0750] Program processing steps

[0751] Step 1:

[0752] Video data capture and transmission

[0753] Devices: Video cameras installed in each room and play area of ​​the childcare facility capture the children's daily activities 24 hours a day, 365 days a year. For example, they capture footage of children playing in the playroom at 9:00 AM.

[0754] Input: Children's daily activities

[0755] Output: Real-time video data

[0756] How it works: The video capture device is equipped with a 360-degree camera that captures children's movements from multiple angles. The captured video data is encrypted and sent to a server using a real-time streaming protocol (such as RTSP).

[0757] Step 2:

[0758] Data reception and primary analysis by the server

[0759] Server: The server temporarily stores the received video data in a buffer and checks for any data loss or compression issues.

[0760] Input: Real-time video data

[0761] Output: Pre-processed video data

[0762] Specific operation: The server checks the frame rate and resolution, checks for missing data, and then performs preprocessing such as color space conversion and frame resizing.

[0763] Step 3:

[0764] Behavioral pattern analysis

[0765] Server: The preprocessed video data is input into a deep learning model to identify children's behavioral patterns. Deep learning models such as ResNet and YOLO are used.

[0766] Input: Preprocessed video data

[0767] Output: Identified behavioral pattern data

[0768] Specific actions: The deep learning model in the server sequentially analyzes actions such as "running," "playing," and "resting" from video data.

[0769] Step 4:

[0770] Abnormal behavior detection

[0771] Server: Compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior and dangerous situations.

[0772] Input: Identified behavioral pattern data

[0773] Output: Abnormal behavior detection alert

[0774] Specific behavior: For example, the server detects a child falling off a playground equipment and flags it as abnormal behavior.

[0775] Step 5:

[0776] Alert generation and notification

[0777] Server: When abnormal behavior is detected, an alert is immediately generated and sent to the device of the parent or caregiver. The alert includes information such as the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[0778] Input: Abnormal behavior detection alert

[0779] Output: Warning message

[0780] Specific operation: The server generates an alert message stating, "A child has fallen off the playground equipment in the playroom," and sends it to the caregiver's smartphone in the form of a push notification, SMS, email, etc.

[0781] Step 6:

[0782] Data accumulation and support for optimizing childcare plans

[0783] Server: Stores daily captured video data and analysis results over the long term. Based on the stored data, it provides information to help optimize childcare plans.

[0784] Input: Identified behavioral pattern data and alarm data

[0785] Output: Statistics for optimizing childcare plans

[0786] Specific operation: The server generates statistics on the activities of children during specific time periods, which is used to review and improve childcare plans.

[0787] The above is the specific flow of program processing for this system.

[0788] (Application example 1)

[0789] 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."

[0790] While there are systems in childcare facilities that safely monitor children's behavior and quickly detect abnormal behavior or dangerous situations and notify parents and caregivers, similar functions are required in home and small childcare environments. There is also a need to provide a simple means of ensuring children's safety using the smartphones of parents and caregivers.

[0791] 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.

[0792] In this invention, the server includes means for acquiring video data from multiple video capture devices, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for the server device to detect abnormal behavior or dangerous situations, means for issuing an alert to parents or caregivers when the abnormal behavior or dangerous situation is detected, means for the server device to continuously monitor the video data 24 hours a day, 365 days a year, means for providing information to optimize childcare plans based on the analysis results, and means for detecting abnormal behavior and sending push notifications to smartphone applications used by families and certified childcare providers. This makes it possible to ensure the safety of children even in home or small childcare environments.

[0793] A "server device" is a central processing unit that analyzes video data over a network and issues an alarm based on the results.

[0794] A "video recording device" is a camera device that is installed in a facility or home to capture the behavior of children.

[0795] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time without delay.

[0796] A "deep learning model" is a machine learning model that analyzes children's behavioral patterns and detects abnormal behavior.

[0797] "Abnormal behavior" refers to behavior that is dangerous for children, such as falling or sudden collisions.

[0798] "Push notifications" are real-time notifications sent to users' smartphones when abnormal behavior is detected.

[0799] A "childcare plan" is a plan for childcare activities that is optimized based on children's behavioral data.

[0800] The "smartphone application used by families and certified childcare mothers" is a mobile application that monitors children's behavior at home and receives notifications when abnormal behavior is detected.

[0801] An "alert" is an alert issued when abnormal behavior or a dangerous situation is detected.

[0802] "Means for continuously monitoring the video data 24 hours a day, 365 days a year" refers to a function that continuously monitors children's behavior, regardless of day or night.

[0803] The present invention is a system for monitoring the behavior of children in childcare facilities and at home to ensure their safety. The system is configured as follows.

[0804] Hardware and software configuration

[0805] Hardware

[0806] Video recording equipment: Cameras installed in childcare facilities and homes monitor children's behavior 24 hours a day, 365 days a year.

[0807] Server device: A central processing unit that receives and analyzes video data. The server runs at all times and continuously monitors data.

[0808] Smartphone: A mobile device used by parents and certified childcare providers. Push notifications are sent when abnormal behavior is detected.

[0809] software

[0810] OpenCV: A library for preprocessing video data and image processing.

[0811] TensorFlow: A library that uses deep learning to analyze children's behavioral patterns and detect abnormal behavior.

[0812] smtplib: A Python module for sending email notifications when anomalous behavior is detected.

[0813] Data processing and calculation

[0814] 1. Camera input: The video camera continuously captures the children's behavior and transmits the video data to the server in real time using the real-time streaming protocol.

[0815] 2. Image preprocessing: The server preprocesses the received video frames to make them easier for the deep learning model to process. OpenCV is used for this preprocessing.

[0816] 3. Behavioral Analysis: Using a trained deep learning model, the server analyzes children's behavioral patterns and uses TensorFlow to identify specific movements and activities.

[0817] 4. Abnormal behavior detection: Abnormal behavior is detected based on the behavioral pattern analysis. If abnormal behavior is detected, an alert is sent to the parents using smtplib.

[0818] 5. Push notification: When abnormal behavior is detected, a push notification will be sent to the smartphone, allowing parents or certified childcare providers to take immediate action.

[0819] 6. Optimization of childcare plans: Through long-term data accumulation, the server provides information that is useful for optimizing childcare plans.

[0820] Specific examples

[0821] For example, if a camera installed in a home monitors a child's behavior and the child suddenly runs off and falls, the server will instantly detect this abnormal behavior and send a push notification to the parent's smartphone, allowing the parent to quickly rush to the scene.

[0822] Prompt Sentence Examples

[0823] "I would like to develop a real-time behavior analysis system using a camera. I need a function that detects abnormal behavior in children and notifies them by email. Please provide me with Python code that uses OpenCV and TensorFlow."

[0824] As described above, the present invention is a system that ensures the safety of children in childcare facilities and homes and reduces the burden on parents and caregivers.

[0825] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0826] Step 1:

[0827] Video data acquisition and transmission:

[0828] The video camera is installed in a childcare facility or a home and captures children's behavior 24 hours a day, 365 days a year. The captured video data is sent to a server using a real-time streaming protocol. The input is a video frame, and the output is real-time video data sent to the server.

[0829] Step 2:

[0830] Image preprocessing:

[0831] The server preprocesses the received video frames to make them easier for the deep learning model to process. Specifically, it resizes and normalizes the video frames using OpenCV. The input is real-time video data, and the output is preprocessed image data.

[0832] Step 3:

[0833] Behavioral pattern analysis:

[0834] Using the preprocessed image data, the server uses a deep learning model to analyze the children's behavioral patterns. It uses TensorFlow to identify specific movements and activities. The input is the preprocessed image data, and the output is the behavioral pattern data resulting from the analysis.

[0835] Step 4:

[0836] Anomalous behavior detection:

[0837] Based on the results of the behavioral pattern analysis, the server detects abnormal behavior. If the behavioral pattern predicted by the deep learning model is determined to be abnormal, it flags the behavior as abnormal. The input is behavioral pattern data, and the output is an abnormal behavior flag.

[0838] Step 5:

[0839] Sending an alert:

[0840] When abnormal behavior is detected, the server uses smtplib to send an alert to parents or caregivers. Specifically, it sends an alert message to a specified email address. It also uses a push notification service to send notifications in real time to smartphones. The input is an abnormal behavior flag, and the output is an alert message and a push notification.

[0841] Step 6:

[0842] Optimize childcare planning:

[0843] The server analyzes behavioral data accumulated over the long term and provides information useful for optimizing childcare plans. Specifically, it statistically processes behavioral patterns and the frequency of abnormal behavior, and provides feedback to caregivers. The input is the accumulated behavioral data, and the output is information about an optimized childcare plan.

[0844] Through these steps, this system can ensure the safety of children in childcare facilities and at home, and reduce the burden on parents and caregivers.

[0845] 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.

[0846] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected. In addition, by incorporating an emotion engine, the system recognizes user emotions and improves the accuracy of abnormal behavior detection.

[0847] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[0848] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[0849] The server device applies an algorithm to detect abnormal behavior based on the results of behavioral pattern analysis. For example, abnormal behavior such as falls, sudden collisions, and excessive excitement can be detected in real time. Furthermore, the server device is equipped with an emotion engine that can recognize the user's emotions from video and audio data. This emotion data is used as feedback to improve the accuracy of abnormal behavior detection.

[0850] When abnormal behavior is detected, the server device generates an alert message containing a timestamp and location information. The alert message includes specific details of the abnormal behavior and the level of urgency. The generated alert message is sent to the device of the parent or caregiver. The message is sent via push notification, SMS, email, etc., allowing caregivers to respond quickly.

[0851] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral and emotional data, it can provide information to help optimize childcare plans. This allows caregivers to make data-based decisions and improve the quality of childcare.

[0852] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one child falls off the playground equipment, the server device instantly detects this abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the comprehensive analysis results, an alert is sent to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[0853] In this way, this system provides an effective means of reducing the burden on caregivers and improving the quality of childcare while ensuring the safety of children. The introduction of an emotion engine further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[0854] The processing flow will be explained below.

[0855] Step 1:

[0856] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[0857] Step 2:

[0858] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[0859] Step 3:

[0860] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[0861] Step 4:

[0862] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[0863] Step 5:

[0864] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[0865] Step 6:

[0866] When abnormal behavior is detected, the server activates the emotion engine, which analyzes the facial expressions and voices of the children and caregivers around the child to assess the level of urgency.

[0867] Step 7:

[0868] Based on the evaluation results from the emotion engine, the server generates an alert message, which includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[0869] Step 8:

[0870] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[0871] Step 9:

[0872] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[0873] Step 10:

[0874] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[0875] Step 11:

[0876] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns, abnormal behavior, and emotional data. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[0877] The above are the specific processing steps and details of each operation of the AI ​​support and care system incorporating an emotion engine. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers. The introduction of the emotion engine also further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[0878] Example 2

[0879] 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."

[0880] While there are systems in childcare facilities that ensure the safety of children while reducing the burden on caregivers, conventional systems lack sufficient accuracy in detecting abnormal behavior or dangerous situations. In particular, they are unable to detect abnormal behavior while taking into account the children's emotions, making it difficult to respond quickly and appropriately. Furthermore, a system that can monitor 24 hours a day, 365 days a year is required, but achieving this requires a high-performance monitoring system, which poses challenges in terms of cost and operation.

[0881] 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.

[0882] In this invention, the server includes means for acquiring data from multiple image capture devices installed in the childcare facility, means for transmitting the data to a processing device in real time, means for the processing device to analyze the data and identify the behavioral patterns of the monitored person, means for the processing device to detect abnormal behavior or dangerous situations, means for issuing an alert to the monitored person when the abnormal behavior or dangerous situation is detected, means for the processing device to continuously monitor the data 24 hours a day, 365 days a year, means for providing information to optimize the monitoring target's plan based on the analysis results, and means for identifying the monitored person's emotions from the data and improving the accuracy of detecting abnormal behavior. This makes it possible to ensure the safety of children, reduce the burden on caregivers, and improve the accuracy of detecting abnormal behavior.

[0883] A "camera" is a device that is installed in each room and play area within a childcare facility to capture the behavior of the person being monitored.

[0884] "Data" refers to the video and audio information captured by the imaging device.

[0885] A "processing device" is a device for receiving and analyzing data and detecting anomalous behavior.

[0886] The "Real Time Streaming Protocol" is a protocol for transmitting data with minimal delay.

[0887] "Behavioral patterns" are mental categories that refer to specific movements or activities of the monitored person (e.g., running, playing, resting).

[0888] "Abnormal behavior" refers to movements or activities that deviate from normal patterns of behavior, such as sudden movements or excessive excitement.

[0889] An "alert" is an emergency message sent to a target when abnormal behavior or a dangerous situation is detected.

[0890] "Information for optimizing childcare plans" is information that includes suggestions and methods for improving the quality of childcare, based on the results of continuous data analysis.

[0891] "Emotion" refers to the psychological state of the monitored person as identified from the video and audio data.

[0892] "Continuous monitoring 24 hours a day, 365 days a year" means constantly monitoring and analyzing data 365 days a year.

[0893] MODE FOR CARRYING OUT THE INVENTION

[0894] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. This system uses multiple image capture devices to monitor children's behavior and analyzes the data with a processing device to detect abnormal behavior and dangerous situations in real time. In addition, the system improves the accuracy of abnormal behavior detection by identifying emotional data.

[0895] Hardware and Software Configuration

[0896] Imaging device

[0897] The devices used are multiple camera devices installed in each room and play area of ​​the childcare facility. Specifically, network cameras are an example. These camera devices capture children's behavior 24 hours a day, 365 days a year, and transmit the data to a processing device in real time.

[0898] Processing equipment

[0899] The server is equipped with a high-performance processor to receive and analyze data sent in real time, such as a cloud-based server (e.g., AWS EC2 instance) or an on-premise high-performance server (e.g., Dell PowerEdge series).

[0900] Data analysis

[0901] Deep learning models based on TensorFlow and PyTorch are used for data analysis. These models run on a server and identify children's behavioral patterns. Microsoft's Emotion API is used as the emotion engine, implementing algorithms to recognize user emotions from video and audio data.

[0902] Acquiring and Sending Data

[0903] The camera records the children's behavior and transmits the data to the processing unit using a real-time streaming protocol, which minimizes data latency.

[0904] Data analysis

[0905] The server analyzes the received data in real time and uses deep learning models and an emotion engine to identify children's behavioral patterns and emotions, instantly detecting abnormal behavior and dangerous situations.

[0906] Sending an alert

[0907] If abnormal behavior is detected, the server generates an alert message containing a timestamp and location information and sends it to the parent or caregiver's device via push notification, SMS, email, etc.

[0908] Optimizing childcare plans

[0909] The server accumulates data over the long term and provides information to help optimize childcare plans based on the analysis results, allowing childcare workers to make data-based decisions and improve the quality of childcare.

[0910] Specific examples

[0911] At 9:00 AM, a network camera installed in the playroom of a childcare facility captures children playing and sends the data to a server in real time. The server analyzes the data using a TensorFlow-based deep learning model to ensure that the children are playing normally. However, if one child falls off the playground equipment, the server instantly detects this abnormal behavior. It also uses the Emotion API to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the results of this analysis, an alert is sent to the childcare worker's smartphone via push notification. The childcare worker receives the alert and can respond promptly.

[0912] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0913] Step 1:

[0914] Video data capture

[0915] Terminals: Terminals are multiple camera devices installed in each room and play area of ​​the childcare facility. The terminals capture children's behavior 24 hours a day, 365 days a year. The inputs include real-time video and audio of the children. The terminals capture this video and audio data and store it in digital format. The output is the captured video and audio data.

[0916] Specific operation: The camera records video and audio from within the room in real time.

[0917] Step 2:

[0918] Video data transmission

[0919] Terminal: The terminal transmits the captured video and audio data to the processing unit in real time. The input is the digital video and audio data generated in step 1. Using the Real Time Streaming Protocol (RTSP), a live data stream is generated that is sent to the processing unit as output.

[0920] Specific operation: The terminal transmits real-time streaming data to the processing device.

[0921] Step 3:

[0922] Receiving data

[0923] Server: The server receives the video and audio data sent from the device in real time. The input is the live data stream sent in step 2. The server receives this data and temporarily stores it in memory. The output is the video and audio data stored in memory.

[0924] Specific operation: The server receives the data stream in real time and stores it in memory.

[0925] Step 4:

[0926] Behavioral pattern analysis

[0927] Server: The server uses a deep learning model to analyze the video and audio data stored in memory. The input is the data stored in step 3. The deep learning model uses TensorFlow and PyTorch to identify children's behavioral patterns from the data. The output is the identified behavioral patterns, such as "running," "playing," and "resting."

[0928] Specific operation: The server applies a deep learning model to analyze behavioral patterns.

[0929] Step 5:

[0930] Emotional Data Analysis

[0931] Server: The server uses an emotion engine to identify the user's emotions from the video and audio data. The input is the data saved in step 3. The emotion engine can be, for example, Microsoft's Azure Emotion API. The output is the identified emotion data.

[0932] Specific operation: The server uses the emotion engine to identify the user's emotion.

[0933] Step 6:

[0934] Abnormal behavior detection

[0935] Server: The server detects abnormal behaviors and dangerous situations based on behavioral patterns and emotional data. The inputs are the behavioral patterns identified in step 4 and the emotional data identified in step 5. Using a specific algorithm, it identifies sudden movements, excessive excitement, etc. as abnormal behaviors. The output is information about the detected abnormal behaviors.

[0936] Specific operation: The server integrates the analysis results and detects abnormal behavior.

[0937] Step 7:

[0938] Generate and send alerts

[0939] Server: The server generates an alert message containing a timestamp and location information based on the detected abnormal behavior information. The input is the abnormal behavior information obtained in step 6. The alert message contains the details of the abnormal behavior and its urgency. This is sent to the parent or caregiver's device via push notification, SMS, email, etc. The output is the sent alert message.

[0940] Specific operation: The server generates an alert message and sends it to the target device.

[0941] (Application example 2)

[0942] 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."

[0943] Childcare facilities are required to ensure the safety of children while reducing the burden on caregivers. However, conventional systems have low accuracy in detecting abnormal behavior and dangerous situations, making it difficult to respond quickly. In addition, because changes in emotions are not reflected in behavior analysis, there is a delay in determining the level of urgency. A system that can solve these problems, detect abnormal behavior with higher accuracy, and respond quickly is needed.

[0944] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0945] In this invention, the server includes means for acquiring video data from multiple video capture devices installed in the childcare facility, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for issuing an alert to parents or caregivers when abnormal behavior or a dangerous situation is detected, means for continuously monitoring the video data 24 hours a day, 365 days a year, means for recognizing user emotions using an emotion engine when detecting abnormal behavior and improving the accuracy of abnormal behavior detection, and means for providing information for optimizing childcare plans based on the analysis results, thereby enabling abnormal behavior to be detected with high accuracy and responding promptly and appropriately.

[0946] A "video capture device" is a camera device installed to capture visual information.

[0947] "Video data" is a collection of visual information recorded by a video capture device.

[0948] A "server device" is a computer system that processes and stores data over a network and communicates with multiple devices.

[0949] "Means for transmitting in real time" refers to means that has the function of transferring data sequentially with minimal delay.

[0950] A "behavioral pattern" refers to a consistent pattern or tendency of behavior of a particular person or group.

[0951] "Abnormal behavior" refers to movements or actions that are different from normal, such as sudden movements or falls.

[0952] A "hazardous situation" is a condition or situation that presents a high risk of accident or injury.

[0953] An "alert" is an emergency notification issued to warn of an abnormality or danger.

[0954] The "emotion engine" is an algorithm that analyzes video and audio data to identify a person's emotions.

[0955] A "care plan" is a set of activities and schedules designed to ensure the safe and proper development of children in a childcare facility.

[0956] MODE FOR CARRYING OUT THE INVENTION

[0957] This invention provides a system that ensures the safety of children and customers in childcare facilities and brick-and-mortar stores and reduces the burden on managers and caregivers. This system starts by acquiring video data from multiple video capture devices and transmitting it to a server device in real time.

[0958] The server device is implemented using a programming language such as Python and uses deep learning libraries such as TensorFlow and Keras to analyze the video data. This analysis uses a deep learning model to detect abnormal behavior and dangerous situations. This model is capable of identifying behavioral patterns obtained from the video and detecting abnormal behavior.

[0959] The server device is equipped with an emotion engine that can recognize the user's emotions from the video and audio data. This emotion engine distinguishes between emotions such as surprise, sadness, and excitement, improving the accuracy of detecting abnormal behavior.

[0960] Furthermore, if abnormal behavior or a dangerous situation is detected, the server device will issue an alert. This alert will be sent to parents or caregivers via push notification, SMS, or email. The server will include the details of the abnormal behavior, the urgency level, a timestamp, and location information in the alert message, enabling a prompt response.

[0961] This system monitors video data 24 hours a day, 365 days a year, providing data that can be used to optimize childcare plans and store management. For example, the accumulation and analysis of long-term behavioral and emotional data is expected to improve the operation of childcare facilities and increase customer satisfaction.

[0962] Specific examples

[0963] Here is a specific example. First, a video camera installed in a childcare facility's playroom captures children playing and sends the video data in real time to a server device. The server device analyzes this video data to confirm that the children are playing normally. However, if one child falls off the playground equipment, the server device immediately detects the abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. An alarm is then sent to the childcare worker's smartphone, allowing the childcare worker to respond immediately.

[0964] As an example of application in a physical store, consider the case where a customer suddenly falls in the store. At that time, a surveillance camera captures the customer's behavior, and the video data is sent to a server in real time. If the deep learning model analyzes the behavior and determines that there is an abnormality, an alert is immediately sent to the store manager based on that information. Furthermore, the emotion engine analyzes the facial expressions and voices of other customers in the store to assess the urgency of the situation.

[0965] Prompt Sentence Examples

[0966] Below is an example of a prompt sentence to input to the generative AI model.

[0967] Your goal is to create a system that detects abnormal behavior in real time from in-store surveillance footage and sends immediate alerts. The technology stack used is TensorFlow, Keras, OpenCV, and Python. Please also implement a specific alert mechanism.

[0968]

[0969] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0970] Step 1:

[0971] A video capture device acquires multiple visual information: the input is the real-time video data captured by the camera, and the output is the raw video data sent from the camera.

[0972] Step 2:

[0973] Video data captured by a video capture device is sent to a server device in real time. The input is video data from the video capture device, and the output is a video stream sent to the server via a network. The Real Time Streaming Protocol (RTSP) is used to minimize delays.

[0974] Step 3:

[0975] The server device receives the video data and begins analysis. The input is the video data sent to the server, and the output is the analyzed behavioral pattern data. The behavioral patterns are identified using deep learning models (TensorFlow, Keras).

[0976] Step 4:

[0977] The server device detects abnormal behavior and dangerous situations based on the identified behavior patterns. The input is behavior pattern data, and the output is a judgment result of abnormal or normal behavior. The analysis is performed using a deep learning algorithm.

[0978] Step 5:

[0979] The emotion engine analyzes video and audio data to recognize the user's emotions. The input is video and audio data, and the output is a judgment result of the user's emotional state (surprise, sadness, excitement, etc.). This improves the accuracy of detecting abnormal behavior.

[0980] Step 6:

[0981] When the server device detects abnormal behavior or a dangerous situation, it generates an alert. The input is the abnormal behavior judgment result and the emotional state, and the output is an alert message. The alert message includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[0982] Step 7:

[0983] Sends alerts to parents' or caregivers' devices. The input is the alert message, and the output is a notification sent in the form of push notification, SMS, email, etc. This allows parents or caregivers to respond quickly.

[0984] Step 8:

[0985] The server device continuously monitors video data 24 hours a day, 365 days a year, and stores the necessary data. The input is real-time video data, and the output is a database of long-term behavioral patterns and emotional states.

[0986] Step 9:

[0987] The accumulated data is analyzed to provide information for optimizing childcare plans and store management. The input is the accumulated behavioral patterns and emotional state data, and the output is an optimized childcare plan and store management proposal.

[0988] 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.

[0989] 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.

[0990] 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.

[0991] [Fourth embodiment]

[0992] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0993] 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.

[0994] 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).

[0995] 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.

[0996] 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.

[0997] 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).

[0998] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0999] 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.

[1000] 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.

[1001] 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.

[1002] 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.

[1003] 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.

[1004] 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."

[1005] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected.

[1006] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[1007] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[1008] The server device detects abnormal behavior based on the results of behavioral pattern analysis. Abnormal behavior includes falls, sudden collisions, and excessive excitement. If abnormal behavior is detected, the server device immediately generates an alert and sends it to the device of the parent or caregiver. The alert is sent in the form of a push notification, SMS, email, or other format, allowing caregivers to respond quickly.

[1009] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral data, it can provide information to help optimize childcare plans. This allows childcare workers to make data-based decisions and improve the quality of childcare.

[1010] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server device instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker will immediately receive the alert and be able to respond promptly.

[1011] In this way, this system provides an effective means of ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[1012] The processing flow will be explained below.

[1013] Step 1:

[1014] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[1015] Step 2:

[1016] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[1017] Step 3:

[1018] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[1019] Step 4:

[1020] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[1021] Step 5:

[1022] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[1023] Step 6:

[1024] If an abnormal behavior is detected, the server generates an alert message including a timestamp and location information, as well as the specific details of the abnormal behavior and its urgency.

[1025] Step 7:

[1026] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[1027] Step 8:

[1028] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[1029] Step 9:

[1030] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[1031] Step 10:

[1032] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns and abnormal behavior. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[1033] The above are the specific processing steps and details of each operation of the AI ​​support and care system. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers.

[1034] Example 1

[1035] 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."

[1036] Conventional safety management systems in childcare facilities lacked the means to monitor children's daily behavior 24 hours a day, 365 days a year. Furthermore, alarm systems to quickly detect and respond to abnormal behavior or dangerous situations were also ineffective. This increased the burden on childcare workers, and led to problems such as insufficient efforts to ensure the safety of children.

[1037] 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.

[1038] In this invention, the server includes a means for temporarily storing video data in a buffer and checking for missing data and compression status, a means for identifying children's behavioral patterns from the video data after preprocessing using a deep learning model, and a means for comparing the identified behavioral patterns with set abnormal behavior conditions to detect abnormal behavior or dangerous situations. This enables real-time monitoring of children's behavior 24 hours a day, 365 days a year, quickly detecting abnormal behavior or dangerous situations, and issuing an alarm. This allows parents and caregivers to respond quickly, effectively ensuring children's safety and reducing the burden on caregivers. It can also provide information to support the optimization of childcare plans based on the accumulated behavioral data.

[1039] A "video recording device" is a device installed in a childcare facility to capture the daily activities of children.

[1040] "Video data" refers to data that includes video information of children's daily activities captured by a video camera.

[1041] The "server device" is a central processing unit for receiving and analyzing video data transmitted from multiple video capture devices installed in the childcare facility.

[1042] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time, and is a technology that minimizes data delays.

[1043] The "deep learning model" is an advanced machine learning model used to identify children's behavioral patterns from received video data and detect abnormal behavior.

[1044] A "behavioral pattern" is a sequence of specific movements or activities that children exhibit within a childcare facility.

[1045] "Abnormal behavior" refers to behavior that deviates from children's normal patterns of behavior and may lead directly to danger or crisis.

[1046] An "alert" is a warning message sent to parents or caregivers when abnormal behavior or a dangerous situation is detected.

[1047] A "push notification" is a real-time alert notification message sent directly from a server device to the device of a parent or caregiver.

[1048] "Optimizing childcare plans" refers to analyzing accumulated behavioral data to improve the quality of childcare and provide information to plan and execute more efficient childcare activities.

[1049] The present invention is a system for ensuring the safety of children in childcare facilities and reducing the burden on caregivers, and is composed of multiple video capture devices, a server device, and a notification system. Specific embodiments for carrying out the invention are described below.

[1050] Installation of video recording equipment and data acquisition

[1051] Device:

[1052] Video recording devices are installed in each room and play area within the childcare facility. 360-degree cameras or high-resolution cameras can be used for these. The video recording devices capture the children's daily activities 24 hours a day, 365 days a year, and transmit the video data to a server device in real time. A real-time streaming protocol (such as RTSP) is used to transmit the video data.

[1053] Server device data reception and analysis

[1054] server:

[1055] The server temporarily stores the received video data in a buffer and then analyzes it using a deep learning model, performing preprocessing such as color space conversion and frame resizing.

[1056] As a specific example, the server uses deep learning models such as ResNet and YOLO to identify children's behavioral patterns, including "running," "playing," and "resting."

[1057] Detecting abnormal behavior and generating alerts

[1058] server:

[1059] The server device compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior or dangerous situations, such as when a child falls off playground equipment or has a sudden collision.

[1060] When abnormal behavior is detected, an alert is immediately generated, containing information about the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[1061] Notification to parents and caregivers

[1062] server:

[1063] Alerts are sent to parents' and caregivers' devices in the form of push notifications, SMS, emails, etc. For example, an alert such as "A child has fallen off play equipment in the playroom" is immediately sent to a caregiver's smartphone.

[1064] Data accumulation and support for optimizing childcare plans

[1065] server:

[1066] The server device stores daily captured video data and analysis results over the long term, making it possible to track long-term changes in behavioral patterns. Based on the stored data, it can provide information to help optimize childcare plans. For example, it can provide statistical information on the activities of children at specific times of the day.

[1067] Specific examples

[1068] At 9:00 a.m., a video camera installed in the playroom of a childcare facility captures children playing and sends the video data to a server in real time. The server analyzes the video data and confirms that the children are playing normally. However, if one of the children falls off the playground equipment, the server instantly detects this abnormal behavior and sends an alert to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[1069] Prompt Sentence Examples

[1070] "Please explain the surveillance system in your childcare facility. This system uses video cameras to monitor children's behavior in real time and issues an alert if it detects abnormal behavior, such as falls or sudden bumps. The system also has 24 / 7 monitoring capabilities and analyzes the accumulated data to help optimize childcare planning."

[1071] In this way, the present invention provides a specific system for ensuring the safety of children, reducing the burden on caregivers, and improving the quality of childcare.

[1072] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1073] Program processing steps

[1074] Step 1:

[1075] Video data capture and transmission

[1076] Devices: Video cameras installed in each room and play area of ​​the childcare facility capture the children's daily activities 24 hours a day, 365 days a year. For example, they capture footage of children playing in the playroom at 9:00 AM.

[1077] Input: Children's daily activities

[1078] Output: Real-time video data

[1079] How it works: The video capture device is equipped with a 360-degree camera that captures children's movements from multiple angles. The captured video data is encrypted and sent to a server using a real-time streaming protocol (such as RTSP).

[1080] Step 2:

[1081] Data reception and primary analysis by the server

[1082] Server: The server temporarily stores the received video data in a buffer and checks for any data loss or compression issues.

[1083] Input: Real-time video data

[1084] Output: Pre-processed video data

[1085] Specific operation: The server checks the frame rate and resolution, checks for missing data, and then performs preprocessing such as color space conversion and frame resizing.

[1086] Step 3:

[1087] Behavioral pattern analysis

[1088] Server: The preprocessed video data is input into a deep learning model to identify children's behavioral patterns. Deep learning models such as ResNet and YOLO are used.

[1089] Input: Preprocessed video data

[1090] Output: Identified behavioral pattern data

[1091] Specific actions: The deep learning model in the server sequentially analyzes actions such as "running," "playing," and "resting" from video data.

[1092] Step 4:

[1093] Abnormal behavior detection

[1094] Server: Compares the identified behavioral patterns with the set abnormal behavior conditions to detect abnormal behavior and dangerous situations.

[1095] Input: Identified behavioral pattern data

[1096] Output: Abnormal behavior detection alert

[1097] Specific behavior: For example, the server detects a child falling off a playground equipment and flags it as abnormal behavior.

[1098] Step 5:

[1099] Alert generation and notification

[1100] Server: When abnormal behavior is detected, an alert is immediately generated and sent to the device of the parent or caregiver. The alert includes information such as the type of abnormal behavior, the time of occurrence, and the location of the occurrence.

[1101] Input: Abnormal behavior detection alert

[1102] Output: Warning message

[1103] Specific operation: The server generates an alert message stating, "A child has fallen off the playground equipment in the playroom," and sends it to the caregiver's smartphone in the form of a push notification, SMS, email, etc.

[1104] Step 6:

[1105] Data accumulation and support for optimizing childcare plans

[1106] Server: Stores daily captured video data and analysis results over the long term. Based on the stored data, it provides information to help optimize childcare plans.

[1107] Input: Identified behavioral pattern data and alarm data

[1108] Output: Statistics for optimizing childcare plans

[1109] Specific operation: The server generates statistics on the activities of children during specific time periods, which is used to review and improve childcare plans.

[1110] The above is the specific flow of program processing for this system.

[1111] (Application example 1)

[1112] 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."

[1113] While there are systems in childcare facilities that safely monitor children's behavior and quickly detect abnormal behavior or dangerous situations and notify parents and caregivers, similar functions are required in home and small childcare environments. There is also a need to provide a simple means of ensuring children's safety using the smartphones of parents and caregivers.

[1114] 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.

[1115] In this invention, the server includes means for acquiring video data from multiple video capture devices, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for the server device to detect abnormal behavior or dangerous situations, means for issuing an alert to parents or caregivers when the abnormal behavior or dangerous situation is detected, means for the server device to continuously monitor the video data 24 hours a day, 365 days a year, means for providing information to optimize childcare plans based on the analysis results, and means for detecting abnormal behavior and sending push notifications to smartphone applications used by families and certified childcare providers. This makes it possible to ensure the safety of children even in home or small childcare environments.

[1116] A "server device" is a central processing unit that analyzes video data over a network and issues an alarm based on the results.

[1117] A "video recording device" is a camera device that is installed in a facility or home to capture the behavior of children.

[1118] "Real-time streaming protocol" is a communication protocol for transmitting video data in real time without delay.

[1119] A "deep learning model" is a machine learning model that analyzes children's behavioral patterns and detects abnormal behavior.

[1120] "Abnormal behavior" refers to behavior that is dangerous for children, such as falling or sudden collisions.

[1121] "Push notifications" are real-time notifications sent to users' smartphones when abnormal behavior is detected.

[1122] A "childcare plan" is a plan for childcare activities that is optimized based on children's behavioral data.

[1123] The "smartphone application used by families and certified childcare mothers" is a mobile application that monitors children's behavior at home and receives notifications when abnormal behavior is detected.

[1124] An "alert" is an alert issued when abnormal behavior or a dangerous situation is detected.

[1125] "Means for continuously monitoring the video data 24 hours a day, 365 days a year" refers to a function that continuously monitors children's behavior, regardless of day or night.

[1126] The present invention is a system for monitoring the behavior of children in childcare facilities and at home to ensure their safety. The system is configured as follows.

[1127] Hardware and software configuration

[1128] Hardware

[1129] Video recording equipment: Cameras installed in childcare facilities and homes monitor children's behavior 24 hours a day, 365 days a year.

[1130] Server device: A central processing unit that receives and analyzes video data. The server runs at all times and continuously monitors data.

[1131] Smartphone: A mobile device used by parents and certified childcare providers. Push notifications are sent when abnormal behavior is detected.

[1132] software

[1133] OpenCV: A library for preprocessing video data and image processing.

[1134] TensorFlow: A library that uses deep learning to analyze children's behavioral patterns and detect abnormal behavior.

[1135] smtplib: A Python module for sending email notifications when anomalous behavior is detected.

[1136] Data processing and calculation

[1137] 1. Camera input: The video camera continuously captures the children's behavior and transmits the video data to the server in real time using the real-time streaming protocol.

[1138] 2. Image preprocessing: The server preprocesses the received video frames to make them easier for the deep learning model to process. OpenCV is used for this preprocessing.

[1139] 3. Behavioral Analysis: Using a trained deep learning model, the server analyzes children's behavioral patterns and uses TensorFlow to identify specific movements and activities.

[1140] 4. Abnormal behavior detection: Abnormal behavior is detected based on the behavioral pattern analysis. If abnormal behavior is detected, an alert is sent to the parents using smtplib.

[1141] 5. Push notification: When abnormal behavior is detected, a push notification will be sent to the smartphone, allowing parents or certified childcare providers to take immediate action.

[1142] 6. Optimization of childcare plans: Through long-term data accumulation, the server provides information that is useful for optimizing childcare plans.

[1143] Specific examples

[1144] For example, if a camera installed in a home monitors a child's behavior and the child suddenly runs off and falls, the server will instantly detect this abnormal behavior and send a push notification to the parent's smartphone, allowing the parent to quickly rush to the scene.

[1145] Prompt Sentence Examples

[1146] "I would like to develop a real-time behavior analysis system using a camera. I need a function that detects abnormal behavior in children and notifies them by email. Please provide me with Python code that uses OpenCV and TensorFlow."

[1147] As described above, the present invention is a system that ensures the safety of children in childcare facilities and homes and reduces the burden on parents and caregivers.

[1148] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1149] Step 1:

[1150] Video data acquisition and transmission:

[1151] The video camera is installed in a childcare facility or a home and captures children's behavior 24 hours a day, 365 days a year. The captured video data is sent to a server using a real-time streaming protocol. The input is a video frame, and the output is real-time video data sent to the server.

[1152] Step 2:

[1153] Image preprocessing:

[1154] The server preprocesses the received video frames to make them easier for the deep learning model to process. Specifically, it resizes and normalizes the video frames using OpenCV. The input is real-time video data, and the output is preprocessed image data.

[1155] Step 3:

[1156] Behavioral pattern analysis:

[1157] Using the preprocessed image data, the server uses a deep learning model to analyze the children's behavioral patterns. It uses TensorFlow to identify specific movements and activities. The input is the preprocessed image data, and the output is the behavioral pattern data resulting from the analysis.

[1158] Step 4:

[1159] Anomalous behavior detection:

[1160] Based on the results of the behavioral pattern analysis, the server detects abnormal behavior. If the behavioral pattern predicted by the deep learning model is determined to be abnormal, it flags the behavior as abnormal. The input is behavioral pattern data, and the output is an abnormal behavior flag.

[1161] Step 5:

[1162] Sending an alert:

[1163] When abnormal behavior is detected, the server uses smtplib to send an alert to parents or caregivers. Specifically, it sends an alert message to a specified email address. It also uses a push notification service to send notifications in real time to smartphones. The input is an abnormal behavior flag, and the output is an alert message and a push notification.

[1164] Step 6:

[1165] Optimize childcare planning:

[1166] The server analyzes behavioral data accumulated over the long term and provides information useful for optimizing childcare plans. Specifically, it statistically processes behavioral patterns and the frequency of abnormal behavior, and provides feedback to caregivers. The input is the accumulated behavioral data, and the output is information about an optimized childcare plan.

[1167] Through these steps, this system can ensure the safety of children in childcare facilities and at home, and reduce the burden on parents and caregivers.

[1168] 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.

[1169] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. Specifically, video data from multiple video capture devices is transmitted in real time to a server device, which analyzes the video data to detect abnormal behavior and issues an alarm if abnormal behavior is detected. In addition, by incorporating an emotion engine, the system recognizes user emotions and improves the accuracy of abnormal behavior detection.

[1170] First, video cameras are installed in each room and play area within the childcare facility. This allows for 24 / 7 monitoring of children's behavior. The video cameras capture the children's daily activities and transmit the video data to a server device in real time. A real-time streaming protocol is used to transmit the video data, minimizing delays in the video data.

[1171] The server then analyzes the received video data. The server uses a deep learning model to identify children's behavioral patterns from the video data. Deep learning algorithms are used to analyze the behavioral patterns and identify specific movements and activities (e.g., running, playing, resting, etc.).

[1172] The server device applies an algorithm to detect abnormal behavior based on the results of behavioral pattern analysis. For example, abnormal behavior such as falls, sudden collisions, and excessive excitement can be detected in real time. Furthermore, the server device is equipped with an emotion engine that can recognize the user's emotions from video and audio data. This emotion data is used as feedback to improve the accuracy of abnormal behavior detection.

[1173] When abnormal behavior is detected, the server device generates an alert message containing a timestamp and location information. The alert message includes specific details of the abnormal behavior and the level of urgency. The generated alert message is sent to the device of the parent or caregiver. The message is sent via push notification, SMS, email, etc., allowing caregivers to respond quickly.

[1174] The system also has a 24 / 7 continuous monitoring function, allowing for monitoring of children's safety day and night. Furthermore, the server device also has a long-term data storage function, and by analyzing the accumulated behavioral and emotional data, it can provide information to help optimize childcare plans. This allows caregivers to make data-based decisions and improve the quality of childcare.

[1175] As a concrete example, let us consider an actual usage scenario in a childcare facility. At 9:00 AM, a video camera installed in the playroom captures children playing and sends the video data to a server device in real time. The server device analyzes the video data and confirms that the children are playing normally. However, if one child falls off the playground equipment, the server device instantly detects this abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the comprehensive analysis results, an alert is sent to the childcare worker's smartphone. The childcare worker receives the alert immediately and can respond promptly.

[1176] In this way, this system provides an effective means of reducing the burden on caregivers and improving the quality of childcare while ensuring the safety of children. The introduction of an emotion engine further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[1177] The processing flow will be explained below.

[1178] Step 1:

[1179] The devices (video recording devices) are installed in each room and play area of ​​the childcare facility. The devices capture video data at specified intervals and store it in a buffer.

[1180] Step 2:

[1181] The terminal prepares to transmit the captured video data to the server in real time using the Real Time Streaming Protocol (RTSP).

[1182] Step 3:

[1183] The server decodes the video data received from the device and stores it in an analysis buffer. The server checks the timestamp of the video data to ensure data continuity and consistency.

[1184] Step 4:

[1185] The server's deep learning model takes as input the video data stored in the analysis buffer and identifies children's behavioral patterns. The model uses a pre-trained behavior classification algorithm to categorize specific behaviors.

[1186] Step 5:

[1187] Based on the identified behavioral patterns, the server applies algorithms to detect abnormal behavior, such as falls, sudden collisions, or excessive excitement, in real time.

[1188] Step 6:

[1189] When abnormal behavior is detected, the server activates the emotion engine, which analyzes the facial expressions and voices of the children and caregivers around the child to assess the level of urgency.

[1190] Step 7:

[1191] Based on the evaluation results from the emotion engine, the server generates an alert message, which includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[1192] Step 8:

[1193] The server sends the generated alert message to the user's (caregiver's) device via push notification, SMS, email, etc., allowing the caregiver to respond quickly.

[1194] Step 9:

[1195] The user (caregiver) checks the received alarm message on the device and immediately rushes to the scene. The caregiver takes appropriate action based on the content of the alarm.

[1196] Step 10:

[1197] The server monitors video data 24 hours a day, 365 days a year, and also accumulates all analysis results and alarm data, saving them in a database.

[1198] Step 11:

[1199] The server periodically analyzes the accumulated data and generates statistical information on behavioral patterns, abnormal behavior, and emotional data. Based on this statistical information, users (caregivers) can obtain information to optimize their childcare plans.

[1200] The above are the specific processing steps and details of each operation of the AI ​​support and care system incorporating an emotion engine. This system can ensure the safety of children in childcare facilities and reduce the burden on caregivers. The introduction of the emotion engine also further improves the accuracy of detecting abnormal behavior, enabling faster and more appropriate responses.

[1201] Example 2

[1202] 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."

[1203] While there are systems in childcare facilities that ensure the safety of children while reducing the burden on caregivers, conventional systems lack sufficient accuracy in detecting abnormal behavior or dangerous situations. In particular, they are unable to detect abnormal behavior while taking into account the children's emotions, making it difficult to respond quickly and appropriately. Furthermore, a system that can monitor 24 hours a day, 365 days a year is required, but achieving this requires a high-performance monitoring system, which poses challenges in terms of cost and operation.

[1204] 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.

[1205] In this invention, the server includes means for acquiring data from multiple image capture devices installed in the childcare facility, means for transmitting the data to a processing device in real time, means for the processing device to analyze the data and identify the behavioral patterns of the monitored person, means for the processing device to detect abnormal behavior or dangerous situations, means for issuing an alert to the monitored person when the abnormal behavior or dangerous situation is detected, means for the processing device to continuously monitor the data 24 hours a day, 365 days a year, means for providing information to optimize the monitoring target's plan based on the analysis results, and means for identifying the monitored person's emotions from the data and improving the accuracy of detecting abnormal behavior. This makes it possible to ensure the safety of children, reduce the burden on caregivers, and improve the accuracy of detecting abnormal behavior.

[1206] A "camera" is a device that is installed in each room and play area within a childcare facility to capture the behavior of the person being monitored.

[1207] "Data" refers to the video and audio information captured by the imaging device.

[1208] A "processing device" is a device for receiving and analyzing data and detecting anomalous behavior.

[1209] The "Real Time Streaming Protocol" is a protocol for transmitting data with minimal delay.

[1210] "Behavioral patterns" are mental categories that refer to specific movements or activities of the monitored person (e.g., running, playing, resting).

[1211] "Abnormal behavior" refers to movements or activities that deviate from normal patterns of behavior, such as sudden movements or excessive excitement.

[1212] An "alert" is an emergency message sent to a target when abnormal behavior or a dangerous situation is detected.

[1213] "Information for optimizing childcare plans" is information that includes suggestions and methods for improving the quality of childcare, based on the results of continuous data analysis.

[1214] "Emotion" refers to the psychological state of the monitored person as identified from the video and audio data.

[1215] "Continuous monitoring 24 hours a day, 365 days a year" means constantly monitoring and analyzing data 365 days a year.

[1216] MODE FOR CARRYING OUT THE INVENTION

[1217] The system of the present invention is designed to ensure the safety of children in childcare facilities and reduce the burden on caregivers. This system uses multiple image capture devices to monitor children's behavior and analyzes the data with a processing device to detect abnormal behavior and dangerous situations in real time. In addition, the system improves the accuracy of abnormal behavior detection by identifying emotional data.

[1218] Hardware and Software Configuration

[1219] Imaging device

[1220] The devices used are multiple camera devices installed in each room and play area of ​​the childcare facility. Specifically, network cameras are an example. These camera devices capture children's behavior 24 hours a day, 365 days a year, and transmit the data to a processing device in real time.

[1221] Processing equipment

[1222] The server is equipped with a high-performance processor to receive and analyze data sent in real time, such as a cloud-based server (e.g., AWS EC2 instance) or an on-premise high-performance server (e.g., Dell PowerEdge series).

[1223] Data analysis

[1224] Deep learning models based on TensorFlow and PyTorch are used for data analysis. These models run on a server and identify children's behavioral patterns. Microsoft's Emotion API is used as the emotion engine, implementing algorithms to recognize user emotions from video and audio data.

[1225] Acquiring and Sending Data

[1226] The camera records the children's behavior and transmits the data to the processing unit using a real-time streaming protocol, which minimizes data latency.

[1227] Data analysis

[1228] The server analyzes the received data in real time and uses deep learning models and an emotion engine to identify children's behavioral patterns and emotions, instantly detecting abnormal behavior and dangerous situations.

[1229] Sending an alert

[1230] If abnormal behavior is detected, the server generates an alert message containing a timestamp and location information and sends it to the parent or caregiver's device via push notification, SMS, email, etc.

[1231] Optimizing childcare plans

[1232] The server accumulates data over the long term and provides information to help optimize childcare plans based on the analysis results, allowing childcare workers to make data-based decisions and improve the quality of childcare.

[1233] Specific examples

[1234] At 9:00 AM, a network camera installed in the playroom of a childcare facility captures children playing and sends the data to a server in real time. The server analyzes the data using a TensorFlow-based deep learning model to ensure that the children are playing normally. However, if one child falls off the playground equipment, the server instantly detects this abnormal behavior. It also uses the Emotion API to analyze the facial expressions and voices of other children nearby to determine the level of urgency. Based on the results of this analysis, an alert is sent to the childcare worker's smartphone via push notification. The childcare worker receives the alert and can respond promptly.

[1235] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1236] Step 1:

[1237] Video data capture

[1238] Terminals: Terminals are multiple camera devices installed in each room and play area of ​​the childcare facility. The terminals capture children's behavior 24 hours a day, 365 days a year. The inputs include real-time video and audio of the children. The terminals capture this video and audio data and store it in digital format. The output is the captured video and audio data.

[1239] Specific operation: The camera records video and audio from within the room in real time.

[1240] Step 2:

[1241] Video data transmission

[1242] Terminal: The terminal transmits the captured video and audio data to the processing unit in real time. The input is the digital video and audio data generated in step 1. Using the Real Time Streaming Protocol (RTSP), a live data stream is generated that is sent to the processing unit as output.

[1243] Specific operation: The terminal transmits real-time streaming data to the processing device.

[1244] Step 3:

[1245] Receiving data

[1246] Server: The server receives the video and audio data sent from the device in real time. The input is the live data stream sent in step 2. The server receives this data and temporarily stores it in memory. The output is the video and audio data stored in memory.

[1247] Specific operation: The server receives the data stream in real time and stores it in memory.

[1248] Step 4:

[1249] Behavioral pattern analysis

[1250] Server: The server uses a deep learning model to analyze the video and audio data stored in memory. The input is the data stored in step 3. The deep learning model uses TensorFlow and PyTorch to identify children's behavioral patterns from the data. The output is the identified behavioral patterns, such as "running," "playing," and "resting."

[1251] Specific operation: The server applies a deep learning model to analyze behavioral patterns.

[1252] Step 5:

[1253] Emotional Data Analysis

[1254] Server: The server uses an emotion engine to identify the user's emotions from the video and audio data. The input is the data saved in step 3. The emotion engine can be, for example, Microsoft's Azure Emotion API. The output is the identified emotion data.

[1255] Specific operation: The server uses the emotion engine to identify the user's emotion.

[1256] Step 6:

[1257] Abnormal behavior detection

[1258] Server: The server detects abnormal behaviors and dangerous situations based on behavioral patterns and emotional data. The inputs are the behavioral patterns identified in step 4 and the emotional data identified in step 5. Using a specific algorithm, it identifies sudden movements, excessive excitement, etc. as abnormal behaviors. The output is information about the detected abnormal behaviors.

[1259] Specific operation: The server integrates the analysis results and detects abnormal behavior.

[1260] Step 7:

[1261] Generate and send alerts

[1262] Server: The server generates an alert message containing a timestamp and location information based on the detected abnormal behavior information. The input is the abnormal behavior information obtained in step 6. The alert message contains the details of the abnormal behavior and its urgency. This is sent to the parent or caregiver's device via push notification, SMS, email, etc. The output is the sent alert message.

[1263] Specific operation: The server generates an alert message and sends it to the target device.

[1264] (Application example 2)

[1265] 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."

[1266] Childcare facilities are required to ensure the safety of children while reducing the burden on caregivers. However, conventional systems have low accuracy in detecting abnormal behavior and dangerous situations, making it difficult to respond quickly. In addition, because changes in emotions are not reflected in behavior analysis, there is a delay in determining the level of urgency. A system that can solve these problems, detect abnormal behavior with higher accuracy, and respond quickly is needed.

[1267] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1268] In this invention, the server includes means for acquiring video data from multiple video capture devices installed in the childcare facility, means for transmitting the video data to a server device in real time, means for the server device to analyze the video data and identify children's behavior patterns, means for issuing an alert to parents or caregivers when abnormal behavior or a dangerous situation is detected, means for continuously monitoring the video data 24 hours a day, 365 days a year, means for recognizing user emotions using an emotion engine when detecting abnormal behavior and improving the accuracy of abnormal behavior detection, and means for providing information for optimizing childcare plans based on the analysis results, thereby enabling abnormal behavior to be detected with high accuracy and responding promptly and appropriately.

[1269] A "video capture device" is a camera device installed to capture visual information.

[1270] "Video data" is a collection of visual information recorded by a video capture device.

[1271] A "server device" is a computer system that processes and stores data over a network and communicates with multiple devices.

[1272] "Means for transmitting in real time" refers to means that has the function of transferring data sequentially with minimal delay.

[1273] A "behavioral pattern" refers to a consistent pattern or tendency of behavior of a particular person or group.

[1274] "Abnormal behavior" refers to movements or actions that are different from normal, such as sudden movements or falls.

[1275] A "hazardous situation" is a condition or situation that presents a high risk of accident or injury.

[1276] An "alert" is an emergency notification issued to warn of an abnormality or danger.

[1277] The "emotion engine" is an algorithm that analyzes video and audio data to identify a person's emotions.

[1278] A "care plan" is a set of activities and schedules designed to ensure the safe and proper development of children in a childcare facility.

[1279] MODE FOR CARRYING OUT THE INVENTION

[1280] This invention provides a system that ensures the safety of children and customers in childcare facilities and brick-and-mortar stores and reduces the burden on managers and caregivers. This system starts by acquiring video data from multiple video capture devices and transmitting it to a server device in real time.

[1281] The server device is implemented using a programming language such as Python and uses deep learning libraries such as TensorFlow and Keras to analyze the video data. This analysis uses a deep learning model to detect abnormal behavior and dangerous situations. This model is capable of identifying behavioral patterns obtained from the video and detecting abnormal behavior.

[1282] The server device is equipped with an emotion engine that can recognize the user's emotions from the video and audio data. This emotion engine distinguishes between emotions such as surprise, sadness, and excitement, improving the accuracy of detecting abnormal behavior.

[1283] Furthermore, if abnormal behavior or a dangerous situation is detected, the server device will issue an alert. This alert will be sent to parents or caregivers via push notification, SMS, or email. The server will include the details of the abnormal behavior, the urgency level, a timestamp, and location information in the alert message, enabling a prompt response.

[1284] This system monitors video data 24 hours a day, 365 days a year, providing data that can be used to optimize childcare plans and store management. For example, the accumulation and analysis of long-term behavioral and emotional data is expected to improve the operation of childcare facilities and increase customer satisfaction.

[1285] Specific examples

[1286] Here is a specific example. First, a video camera installed in a childcare facility's playroom captures children playing and sends the video data in real time to a server device. The server device analyzes this video data to confirm that the children are playing normally. However, if one child falls off the playground equipment, the server device immediately detects the abnormal behavior and uses its emotion engine to analyze the facial expressions and voices of other children nearby to determine the level of urgency. An alarm is then sent to the childcare worker's smartphone, allowing the childcare worker to respond immediately.

[1287] As an example of application in a physical store, consider the case where a customer suddenly falls in the store. At that time, a surveillance camera captures the customer's behavior, and the video data is sent to a server in real time. If the deep learning model analyzes the behavior and determines that there is an abnormality, an alert is immediately sent to the store manager based on that information. Furthermore, the emotion engine analyzes the facial expressions and voices of other customers in the store to assess the urgency of the situation.

[1288] Prompt Sentence Examples

[1289] Below is an example of a prompt sentence to input to the generative AI model.

[1290] Your goal is to create a system that detects abnormal behavior in real time from in-store surveillance footage and sends immediate alerts. The technology stack used is TensorFlow, Keras, OpenCV, and Python. Please also implement a specific alert mechanism.

[1291]

[1292] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1293] Step 1:

[1294] A video capture device acquires multiple visual information: the input is the real-time video data captured by the camera, and the output is the raw video data sent from the camera.

[1295] Step 2:

[1296] Video data captured by a video capture device is sent to a server device in real time. The input is video data from the video capture device, and the output is a video stream sent to the server via a network. The Real Time Streaming Protocol (RTSP) is used to minimize delays.

[1297] Step 3:

[1298] The server device receives the video data and begins analysis. The input is the video data sent to the server, and the output is the analyzed behavioral pattern data. The behavioral patterns are identified using deep learning models (TensorFlow, Keras).

[1299] Step 4:

[1300] The server device detects abnormal behavior and dangerous situations based on the identified behavior patterns. The input is behavior pattern data, and the output is a judgment result of abnormal or normal behavior. The analysis is performed using a deep learning algorithm.

[1301] Step 5:

[1302] The emotion engine analyzes video and audio data to recognize the user's emotions. The input is video and audio data, and the output is a judgment result of the user's emotional state (surprise, sadness, excitement, etc.). This improves the accuracy of detecting abnormal behavior.

[1303] Step 6:

[1304] When the server device detects abnormal behavior or a dangerous situation, it generates an alert. The input is the abnormal behavior judgment result and the emotional state, and the output is an alert message. The alert message includes the details of the abnormal behavior, the urgency level, a timestamp, and location information.

[1305] Step 7:

[1306] Sends alerts to parents' or caregivers' devices. The input is the alert message, and the output is a notification sent in the form of push notification, SMS, email, etc. This allows parents or caregivers to respond quickly.

[1307] Step 8:

[1308] The server device continuously monitors video data 24 hours a day, 365 days a year, and stores the necessary data. The input is real-time video data, and the output is a database of long-term behavioral patterns and emotional states.

[1309] Step 9:

[1310] The accumulated data is analyzed to provide information for optimizing childcare plans and store management. The input is the accumulated behavioral patterns and emotional state data, and the output is an optimized childcare plan and store management proposal.

[1311] 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.

[1312] 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.

[1313] 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.

[1314] 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.

[1315] 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.

[1316] 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.

[1317] 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).

[1318] 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.

[1319] 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."

[1320] 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.

[1321] 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).

[1322] 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.

[1323] 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.

[1324] 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.

[1325] 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.

[1326] 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.

[1327] 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.

[1328] 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.

[1329] 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.

[1330] 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.

[1331] 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.

[1332] The following is further disclosed regarding the above embodiment.

[1333] (Claim 1)

[1334] means for acquiring video data from a plurality of video capture devices installed within the childcare facility;

[1335] means for transmitting the video data to a server device in real time;

[1336] a means for analyzing the video data and identifying a child's behavior pattern, in the server device;

[1337] The server device has means for detecting abnormal behavior and dangerous situations;

[1338] means for issuing an alarm to a parent or caregiver when the abnormal behavior or dangerous situation is detected;

[1339] means for the server device to continuously monitor the video data 24 hours a day, 365 days a year;

[1340] A means for providing information for optimizing a childcare plan based on the analysis results;

[1341] A system including:

[1342] (Claim 2)

[1343] 10. The system of claim 1, wherein the means for transmitting the video data in real time uses a real-time streaming protocol.

[1344] (Claim 3)

[1345] 2. The system of claim 1, wherein the means for detecting abnormal behavior or dangerous situations performs analysis based on a deep learning model.

[1346] "Example 1"

[1347] (Claim 1)

[1348] means for acquiring video data from a plurality of video capture devices;

[1349] means for transmitting the video data to a server device in real time;

[1350] a means for temporarily storing the video data in a buffer in the server device and checking for missing data and a compression state;

[1351] means for identifying children's behavior patterns from the video data using a deep learning model after preprocessing in the server device;

[1352] a means for detecting abnormal behavior or a dangerous situation by comparing the identified behavior pattern with a set abnormal behavior condition in the server device;

[1353] means for generating an alarm when the abnormal behavior or dangerous situation is detected and sending the alarm to a device of a parent or caregiver in the form of a push notification, SMS, email, etc.;

[1354] means for the server device to continuously monitor the video data 24 hours a day, 365 days a year;

[1355] The server device has a means for accumulating daily behavior data over a long period of time based on the analysis results and providing information for optimizing childcare plans;

[1356] A system including:

[1357] (Claim 2)

[1358] 10. The system of claim 1, wherein the means for transmitting the video data in real time uses a real-time streaming protocol.

[1359] (Claim 3)

[1360] 2. The system of claim 1, wherein the means for detecting abnormal behavior or dangerous situations performs analysis based on a deep learning model.

[1361] "Application Example 1"

[1362] (Claim 1)

[1363] means for acquiring video data from a plurality of video capture devices installed within the childcare facility;

[1364] means for transmitting the video data to a server device in real time;

[1365] a means for analyzing the video data and identifying a child's behavior pattern, in the server device;

[1366] The server device has means for detecting abnormal behavior and dangerous situations;

[1367] means for issuing an alarm to a parent or caregiver when the abnormal behavior or dangerous situation is detected;

[1368] means for the server device to continuously monitor the video data 24 hours a day, 365 days a year;

[1369] A means for providing information for optimizing a childcare plan based on the analysis results;

[1370] A method for detecting abnormal behavior and sending push notifications to smartphone applications used by families and certified childcare mothers.

[1371] A system including:

[1372] (Claim 2)

[1373] 10. The system of claim 1, wherein the means for transmitting the video data in real time uses a real-time streaming protocol.

[1374] (Claim 3)

[1375] 2. The system of claim 1, wherein the means for detecting abnormal behavior or dangerous situations performs analysis based on a deep learning model.

[1376] "Example 2: Combining Emotion Engines"

[1377] (Claim 1)

[1378] a means for acquiring data from a plurality of imaging devices installed in the childcare facility;

[1379] means for transmitting said data to a processing device in real time;

[1380] means for the processing device to analyze the data and identify patterns of behavior of the monitored person;

[1381] means for detecting abnormal behavior or dangerous situations;

[1382] means for issuing an alarm to a target person when the abnormal behavior or dangerous situation is detected;

[1383] means for the processing device to continuously monitor the data 24 hours a day, 365 days a year;

[1384] means for providing information for optimizing a monitoring plan based on the analysis results;

[1385] means for identifying emotions of the monitored person from the data and improving the accuracy of detecting the abnormal behavior;

[1386] A system including:

[1387] (Claim 2)

[1388] 10. The system of claim 1, wherein the means for transmitting data in real time uses a real-time streaming protocol.

[1389] (Claim 3)

[1390] 2. The system according to claim 1, wherein the means for detecting abnormal behavior or dangerous situations performs analysis based on a machine learning model.

[1391] "Application example 2 when combining emotion engines"

[1392] (Claim 1)

[1393] means for acquiring video data from a plurality of video capture devices installed within the childcare facility;

[1394] means for transmitting the video data to a server device in real time;

[1395] a means for analyzing the video data and identifying a child's behavior pattern, in the server device;

[1396] The server device has means for detecting abnormal behavior and dangerous situations;

[1397] means for issuing an alarm to a parent or caregiver when the abnormal behavior or dangerous situation is detected;

[1398] means for the server device to continuously monitor the video data 24 hours a day, 365 days a year;

[1399] a means for recognizing a user's emotion using an emotion engine when detecting the abnormal behavior, thereby improving the accuracy of detecting the abnormal behavior;

[1400] A means for providing information for optimizing a childcare plan based on the analysis results;

[1401] A system including:

[1402] (Claim 2)

[1403] 10. The system of claim 1, wherein the means for transmitting the video data in real time uses a real-time streaming protocol.

[1404] (Claim 3)

[1405] 2. The system of claim 1, wherein the means for detecting abnormal behavior or dangerous situations performs analysis based on a deep learning model. [Explanation of symbols]

[1406] 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. means for acquiring video data from a plurality of video capture devices installed within the childcare facility; means for transmitting the video data to a server device in real time; a means for analyzing the video data and identifying a child's behavior pattern, in the server device; The server device has means for detecting abnormal behavior and dangerous situations; means for issuing an alarm to a parent or caregiver when the abnormal behavior or dangerous situation is detected; means for the server device to continuously monitor the video data 24 hours a day, 365 days a year; A means for providing information for optimizing a childcare plan based on the analysis results; A system including:

2. 2. The system of claim 1, wherein the means for transmitting video data in real time uses a real-time streaming protocol.

3. The system of claim 1 , wherein the means for detecting abnormal behavior or dangerous situations performs analysis based on a deep learning model.

Images