Behavior and psychological state monitoring method and system

Abstract

The invention belongs to the technical field of artificial intelligence and image processing, and particularly discloses a behavior and psychological state monitoring method and system.The behavior and psychological state monitoring method comprises the steps that video data are captured through a commercial camera, and encryption and timestamp processing are conducted on the video data through a data acquisition module; the method comprises the following steps: preprocessing video data to obtain preprocessed image data; based on the preprocessed image data, using a convolutional neural network to identify the behavior pattern of an individual, and generating a corresponding behavior data set; based on the behavior data in the corresponding behavior data set, evaluating the psychological state of the individual, and generating an evaluation result; and generating an alarm based on an evaluation result, and notifying related personnel to carry out intervention. According to the invention, social anxiety is reduced, and the efficiency of public safety and mental health management is improved.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and image processing technology, and in particular to a behavior and psychological state monitoring method and system. Background Art

[0002] With the reduction of family size and the increase of work pressure on young people, the health and mental health of the elderly are often neglected. With the aging of the global population, the mental health problems of the elderly are becoming more and more prominent. At the same time, children and adolescents are also facing school violence, academic pressure and social isolation. The increase of depression and anxiety has led to serious social consequences. At present, the identification and intervention of mental health problems mainly rely on regular assessments by doctors or psychologists, or observations by non-professionals such as family members and teachers.

[0003] However, the above methods have latency issues, rely on subjective judgment, have limited coverage, and are difficult to monitor high-risk individuals in real time and continuously. With the development of image processing technology and artificial intelligence, the use of cameras for behavioral analysis has shown great potential, especially in the early detection of behavioral anomalies. Despite this, existing camera surveillance systems are mainly used for security and environmental monitoring, lacking in-depth analysis of human behavior patterns and intelligent assessment of psychological states. Summary of the invention

[0004] The present invention aims to provide a behavior and psychological state monitoring method and system to solve or improve at least one of the above-mentioned problems such as delay in judging human behavior, reliance on subjectivity, limited coverage, difficulty in real-time and continuous monitoring of high-risk individuals, and lack of in-depth analysis of human behavior patterns and intelligent assessment of psychological states.

[0005] In a first aspect, a behavior and psychological state monitoring system is provided, comprising:

[0006] A video data acquisition module, configured to continuously capture video data in a specific environment using at least one commercial camera, wherein the specific environment includes a school, a nursing home, a home or a workplace of an elderly person or child living alone;

[0007] An image processing module, used for preprocessing the video data;

[0008] The behavior analysis module is used to extract key visual features based on the preprocessed image data using convolutional neural networks and long short-term memory networks to identify individual behavior patterns. The key visual features include: individual movement trajectory, posture changes, gestures, facial expressions, and interactions with the surrounding environment;

[0009] A psychological state assessment module, used to assess the psychological state of an individual based on the analysis results of the behavioral pattern by using a support vector machine algorithm;

[0010] The early warning mechanism module is used to generate an alarm based on the psychological state assessment result and notify relevant personnel.

[0011] In any of the above technical solutions, the video data acquisition module includes:

[0012] The camera configuration submodule is used to adjust the camera's resolution, frame rate, sensitivity, and viewing angle;

[0013] A data preprocessing submodule, for encrypting and time-stamping the video data captured after adjusting the camera;

[0014] The data storage and transmission submodule is used to compress and securely transmit the processed video data.

[0015] In any of the above technical solutions, the image processing module includes:

[0016] Noise filtering submodule, used to reduce random noise in the preprocessed video;

[0017] A contrast enhancement submodule, for enhancing the contrast of the pre-processed video image by histogram equalization;

[0018] The object tracking submodule is used to identify key visual features in the pre-processed video through edge detection and object tracking techniques.

[0019] In any of the above technical solutions, the behavior analysis module includes:

[0020] A feature extraction submodule, for extracting the key visual features from the preprocessed video frames using a convolutional neural network;

[0021] A sequence analysis submodule, for analyzing a behavioral feature sequence of the key visual features through a long short-term memory network;

[0022] The behavior recognition and classification submodule is used to use a support vector machine to recognize the behavior pattern from the behavior feature sequence and obtain a behavior feature vector.

[0023] In any of the above technical solutions, the mental state assessment module includes:

[0024] A classification submodule, used for classifying the behavior feature vector;

[0025] The probability model submodule is used to analyze the probability of psychological state changes for various behavioral feature vectors using the hidden Markov model;

[0026] The decision support submodule is used to provide a basis for psychological state inference based on the probability of the psychological state change through historical data and real-time data to obtain a psychological state assessment result.

[0027] In any of the above technical solutions, the early warning mechanism module includes:

[0028] A threshold setting submodule, used to set a threshold for triggering an alarm according to the psychological state assessment result;

[0029] The alarm logic optimization submodule is used to optimize the alarm triggering logic through logistic regression or decision tree models;

[0030] The alarm notification configuration submodule is used to allow users to customize the recipients and priorities of alarm notifications.

[0031] In a second aspect, the present invention provides a behavior and psychological state monitoring method, comprising:

[0032] Capturing video data using a commercial camera, and encrypting and time-stamping the video data;

[0033] By preprocessing the video data;

[0034] Based on the preprocessed image data, a convolutional neural network is applied to identify the behavior pattern of the individual and generate a corresponding behavior data set;

[0035] Based on the behavior data in the corresponding behavior data set, the psychological state of the individual is evaluated to generate an evaluation result, wherein the psychological state evaluation result is represented by a psychological state score;

[0036] An alert is generated based on the evaluation results, and relevant personnel are notified to intervene.

[0037] In any of the above technical solutions, the step of applying a convolutional neural network to identify individual behavior patterns based on the preprocessed image data to generate a corresponding behavior data set includes:

[0038] Based on the pre-processed image data, a convolutional neural network is used to analyze individual behaviors in real time to identify individual behavior patterns;

[0039] Based on the individual's behavior pattern, a corresponding behavior data set is generated, and specific behavior events are marked for each behavior data set.

[0040] In any of the above technical solutions, the individual's psychological state is evaluated based on the behavioral data in the corresponding behavioral data set, and an evaluation result is generated, including: based on the behavioral data in the corresponding behavioral data set, combined with multiple psychological models, a multi-dimensional evaluation of the individual's psychological state is performed, and a psychological state report is generated, wherein the psychological model includes an emotional model and a behavioral model.

[0041] In any of the above technical solutions, generating an alarm based on the evaluation result and notifying relevant personnel to intervene includes:

[0042] Set up multi-level alarms to generate corresponding levels of intervention notifications based on different psychological status assessment results;

[0043] Based on the corresponding level of intervention notification, relevant personnel are notified to intervene, where the expression for generating the alarm is as follows:

[0044] Alert(t)=1{E(t)>τ}

[0045] Among them, Alert(t) indicates whether the alarm is triggered, E(t) represents the psychological state assessment result, that is, the psychological state score, τ represents the preset risk threshold, and 1 represents the indicator function, which is used to determine whether the threshold is exceeded.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. Automatically analyze captured images through artificial intelligence algorithms, focus on real-time assessment of behavioral and mental health conditions for the elderly and children, and provide real-time warnings when abnormal behavior or psychological problems occur;

[0048] 2. The present invention uses the intelligent system of commercial cameras to conduct timely intervention when abnormal behavior or psychological problems are found, thereby reducing social anxiety and improving the efficiency of public safety and mental health management;

[0049] 3. The present invention uses commercial cameras combined with artificial intelligence to automatically analyze psychological states, which not only achieves non-invasive, real-time monitoring, but also can continuously track behavioral changes, timely discover and warn of potential mental health problems, and has rare foresight for preventing psychological problems in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0051] Figure 1 A behavior and psychological state monitoring method of the present invention;

[0052] Figure 2is a block diagram of a behavior and psychological state monitoring system of the present invention;

[0053] Figure 3 The figure is a schematic diagram of the structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0054] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0056] Figure 1 A behavior and psychological state monitoring method of the present invention;

[0057] The following describes a behavior and psychological state monitoring method of the present invention. The method comprises:

[0058] S1. Capture video data using a commercial camera and encrypt and timestamp the video data through a data acquisition module;

[0059] In one embodiment, at least one commercial camera is used to continuously capture video data of an individual in a specific environment. At the same time, the captured video data is encrypted through a data acquisition module to ensure data security, and a timestamp is added to the video data to ensure time synchronization, so that the data acquisition time can be tracked during subsequent analysis. Commercial cameras usually have higher pixels and clearer picture quality, and can capture more details, which is crucial for security monitoring. Suitable camera types include webcams or surveillance cameras; examples of specific environments include schools, nursing homes, homes or workplaces of elderly people or children living alone, etc.

[0060] Specifically, commercial cameras are configured to meet the needs of different monitoring environments. The configuration includes adjusting the camera's resolution, frame rate, light sensitivity, and viewing angle to ensure that a clear and continuous video stream is captured. The camera output is defined as V(t,x,y,λ), where V(t,x,y,λ) represents the camera output, t represents time, x and y represent image space coordinates, and λ represents the color channel. By setting R x ×R y resolution and f frame rate to optimize image quality and data processing efficiency.

[0061] Process the original video stream and perform necessary data security and quality assurance measures, such as video data encryption and time stamping. The above measures ensure the security and traceability of the data in subsequent processing. Apply the encryption function E(V) to the video stream to generate an encrypted video stream V'(t,x,y,λ), and implement time stamping for each frame to ensure data synchronization and integrity.

[0062] To ensure that the encrypted and time-stamped video data is safely and efficiently compressed and stored, while ensuring the security and integrity of the data transmission process. The compression function C(V') is used to process the encrypted video to reduce the data storage space and transmission bandwidth requirements. The compressed video V"(t,x,y,λ) is transmitted through a security protocol T(V") to ensure the security of the data in the network.

[0063] S2. By preprocessing the video data;

[0064] In one embodiment, the captured video is preprocessed by denoising, contrast enhancement, etc. to generate preprocessed clear and stable image data, providing high-quality input data for subsequent behavior recognition.

[0065] Specifically, it reduces random noise in the image and improves image quality. This is especially important for subsequent image analysis, because noise can significantly affect the accuracy of the analysis algorithm. Linear smoothing filtering technology is generally used.

[0066] By enhancing the contrast of the image, the details of the image are made more obvious, which helps to better identify and analyze the content in the image. The main techniques used are histogram equalization, which is used to enhance the contrast by adjusting the brightness distribution of the image. Specifically, it is expressed by adjusting the brightness value of each pixel in the image to make the brightness distribution more uniform.

[0067] Detecting edge information in images and tracking targets is particularly critical for analyzing behavioral patterns in dynamic scenes. Edge detection is mainly performed in combination with the Canny edge detector and the Sobel operator. The target tracking part is implemented using deep learning networks such as GOTURN and the twin fully convolutional network. The network tracks the target by comparing features between consecutive frames. Canny is a commonly used image processing algorithm, Sobel is the name of the edge detection operator, and GOTURN is a deep learning network.

[0068] S3. Based on the preprocessed image data, applying a convolutional neural network to identify individual behavior patterns and generate a corresponding behavior data set;

[0069] In one embodiment, a convolutional neural network is used to analyze the preprocessed image data, identify individual behavior patterns, and generate a corresponding behavior data set. The behavior data set will contain behavioral event information of the identified behavioral events, such as action type, duration, and visual features. Specific behavior tags include "extreme", "angry", "self-harm", etc. to identify potential abnormal psychological behaviors.

[0070] Based on the pre-processed image data, a convolutional neural network is used to analyze individual behaviors in real time to identify individual behavior patterns;

[0071] Based on the individual's behavior pattern, a corresponding behavior data set is generated, and specific behavior events are marked for each behavior data set.

[0072] In one embodiment, when using a convolutional neural network to analyze preprocessed image data, the system first extracts key visual features, including the individual's motion trajectory, posture changes, gestures, facial expressions, and interactions with the surrounding environment. Through real-time analysis, the system can quickly identify the individual's behavior patterns, such as walking, sitting, standing, waving, etc., and mark each behavior. The present invention uses a deep learning framework to process and analyze video frame sequences, realizing the recognition of complex behaviors and the analysis of long-term behavioral trends.

[0073] Each behavioral event will be generated into a corresponding behavioral data set, which not only contains the identification of the behavior type, but also records the timestamp, duration, and possible triggering factors of the behavior. In addition, the system will continuously update this data set to provide accurate information for psychological state assessment at any time. Possible triggering factors include, for example, environmental changes or specific events.

[0074] Specifically, CNN is used to automatically extract key visual features from each frame of video. Feature extraction can be expressed as F(t) = CNN(I'(t); θ), where I'(t) is the t-th frame image processed by the image processing module, and θ is the parameter of the CNN model. Among them, F(t) represents the feature vector extracted from the t-th frame, and CNN(I'(t); θ) represents the automatic extraction of key visual features from each frame of video. Among them, CNN (Convolutional Neural Network) is translated into Chinese as convolutional neural network.

[0075] Apply RNN or LSTM to analyze the extracted feature sequence, identify behavior patterns and predict future behavior. Sequence analysis can be expressed as B(t)=LSTM(F(t);φ), where B(t) represents the behavior feature sequence, φ is the parameter of the LSTM model, and LSTM(F(t);φ) represents the feature sequence extracted by applying LSTM analysis. RNN (Recurrent Neural Network) is translated into Chinese as recurrent neural network, and LSTM (Long Short-Term Memory) is translated into Chinese as long short-term memory network.

[0076] Based on the results of sequence analysis, a classification algorithm is used to identify and mark specific behavior patterns. Behavior classification can be performed using SVM or other suitable classifiers, expressed as C(t)=classify(B(t);ψ), where C(t) represents the behavior classification and ψ is the parameter of the classifier. Among them, SVM (Support Vector Machine) is translated into Chinese as support vector machine.

[0077] S4. Based on the behavior data in the corresponding behavior data set, the psychological state of the individual is evaluated and an evaluation result is generated, wherein the psychological state evaluation result is represented by a psychological state score;

[0078] In one embodiment, the mental state of an individual is evaluated based on the generated behavioral data set. First, the system analyzes the individual's behavior frequency, change trend, and time pattern through the behavioral data set. For example, if the individual's social behavior is significantly reduced, or if they show repeated negative behavior patterns, it may point to potential mental health problems.

[0079] Next, we will combine multiple psychological models such as emotional models and behavioral models to conduct multi-dimensional assessments and analyze the changes in the individual's mental state in different time periods. Based on the in-depth analysis of behavioral data, a mental state report will be generated, which will describe the individual's mental health in detail and propose possible mental state indicators, such as anxiety, depression, etc.

[0080] Among them, psychological models include emotional models, behavioral models, etc.

[0081] Specifically, based on the behavioral data set, the system evaluates the individual's mental state through a behavioral analysis model and generates a mental state evaluation result, which may include labels such as mild depression, depression, and schizophrenia. The state evaluation formula is as follows:

[0082]

[0083] Among them, E(t) represents the psychological state score, ω i is the weight of each behavior, f i(B(t)) is the behavior characteristic value function, and B(t) is the behavior pattern.

[0084] If the assessment results show that the individual's mental state is abnormal, the system will generate an alarm and notify relevant personnel so that necessary intervention can be carried out.

[0085] Specifically, SVM is used to classify the psychological state according to the behavioral feature vector. The psychological state classification is represented by a mathematical model P(t) = SVM(B(t); θ), where P(t) represents the probability distribution of the current state, θ is the parameter of the SVM model, and SVM(P(t); θ) represents the classification prediction result of the psychological state by a support vector machine model given the behavioral feature vector P(t) and the model parameter θ, which is used to estimate the current psychological state.

[0086] Combined with probability graphical models such as HMM, it is used to analyze the transition probability between psychological states and enhance the dynamic understanding of the evolution of psychological states. The state transition probability can be expressed by HMM as State(t+1)=HMM(P(t);φ), where HMM (Hidden Markov Model), translated into Chinese as Hidden Markov Model, is a mathematical tool for exploring the invisible world, phenomena or facts, which is the probability of a series of events occurring in succession; State(t+1) represents the psychological state at time t+1, which represents the state of the system at the next time step, calculated based on the current state and the transition probability model; HMM(P(t);φ) represents the probability distribution of the next (future) psychological state based on the currently observed psychological state P(t) and the model parameter φ.

[0087] Using algorithms such as decision trees, we provide decision makers with early warnings of their mental states and action recommendations based on historical and real-time data.

[0088] S5. Generate an alarm based on the evaluation result and notify relevant personnel to intervene.

[0089] Set up multi-level alarms to generate corresponding levels of intervention notifications based on different psychological status assessment results;

[0090] Based on the corresponding level of intervention notification, relevant personnel are notified to intervene, where the expression for generating the alarm is as follows:

[0091] Alert(t)=1{E(t)>τ}

[0092] Among them, Alert(t) indicates whether the alarm is triggered, E(t) represents the psychological state assessment result, that is, the psychological state score, τ represents the preset risk threshold, and 1 represents the indicator function, which is used to determine whether the threshold is exceeded.

[0093] Among them, relevant personnel include: parents, teachers, children and other people who have a close relationship with the monitored person.

[0094] In one embodiment, the system implements alarm generation and intervention measures based on the generated psychological state assessment results. First, the system will set different alarm trigger thresholds, divide the individual's psychological state into low risk, medium risk and high risk intervals according to the assessment results, and generate corresponding alarms for different risk levels.

[0095] For example, when the individual's psychological status assessment results show a high risk, the system will immediately trigger a high-priority alarm to notify professional psychological counselors or relevant family members. At the same time, the system sends the alarm notification to the designated recipients via mobile phone applications or emails so that they can intervene in time. Then, users can customize the recipients and priorities of the alarm notifications as needed to ensure that intervention measures can be flexibly adjusted according to actual conditions to meet the specific needs of individuals.

[0096] Specifically, the alarm trigger threshold is set according to the results of the psychological status assessment, the corresponding alarm is generated, and relevant personnel are notified to intervene. The alarm levels are divided into three levels: low, medium, and high. Low-risk alarms can notify designated personnel through mobile phone applications, medium-risk alarms are sent to family members or caregivers via email, and high-risk alarms immediately notify professional psychological counselors or emergency contacts for rapid response.

[0097] It allows users to configure the recipients and priorities of alarm notifications through a customized interface. It provides a user interface that allows setting the details of the alarm, such as the recipient's contact information, alarm level, and response requirements.

[0098] The present invention automatically analyzes captured images through artificial intelligence algorithms, focusing on real-time assessment of behavior and mental health status of the elderly and children, and providing real-time warnings when abnormal behavior or psychological problems occur; through the intelligent system of commercial cameras, timely intervention is carried out when abnormal behavior or psychological problems are found, thereby reducing social anxiety and improving the efficiency of public safety and mental health management; through the automatic analysis of psychological states by commercial cameras combined with artificial intelligence, not only non-invasive and real-time monitoring is achieved, but also behavioral changes can be continuously tracked, potential mental health problems can be discovered and warned in a timely manner, and it has rare foresight for preventing psychological problems in advance.

[0099] Figure 2 is a block diagram of a behavior and psychological state monitoring system of the present invention;

[0100] A behavior and psychological state monitoring system, characterized by comprising:

[0101] A video data acquisition module, configured to continuously capture video data in an environment using at least one commercial camera, wherein the specific environment includes a school, a nursing home, a home or a workplace of an elderly person or child living alone;

[0102] The camera configuration submodule is used to adjust the camera's resolution, frame rate, sensitivity, and viewing angle;

[0103] A data preprocessing submodule, for encrypting and time-stamping the video data captured after adjusting the camera;

[0104] The data storage and transmission submodule is used to compress and securely transmit the processed video data. The image processing module is used to pre-process the video data;

[0105] Noise filtering submodule, used to reduce random noise in the preprocessed video;

[0106] A contrast enhancement submodule, for enhancing the contrast of the pre-processed video image by histogram equalization;

[0107] The object tracking submodule is used to identify key visual features in the pre-processed video through edge detection and object tracking techniques.

[0108] The behavior analysis module is used to extract key visual features based on the preprocessed image data using convolutional neural networks and long short-term memory networks to identify individual behavior patterns. The key visual features include: individual movement trajectory, posture changes, gestures, facial expressions, and interactions with the surrounding environment;

[0109] A feature extraction submodule, for extracting the key visual features from the preprocessed video frames using a convolutional neural network;

[0110] A sequence analysis submodule, for analyzing a behavioral feature sequence of the key visual features through a long short-term memory network;

[0111] The behavior recognition and classification submodule is used to use a support vector machine to recognize the behavior pattern from the behavior feature sequence and obtain a behavior feature vector.

[0112] A psychological state assessment module, which assesses the psychological state of an individual based on the analysis results of the behavioral pattern through a support vector machine algorithm;

[0113] A classification submodule, used for classifying the behavior feature vector;

[0114] The probability model submodule is used to analyze the probability of psychological state changes for various behavioral feature vectors using the hidden Markov model;

[0115] The decision support submodule is used to provide a basis for psychological state inference based on the probability of the psychological state change through historical data and real-time data to obtain a psychological state assessment result.

[0116] The early warning mechanism module is used to generate an alarm based on the psychological state assessment result and notify relevant personnel.

[0117] A threshold setting submodule, used to set a threshold for triggering an alarm according to the psychological state assessment result;

[0118] The alarm logic optimization submodule is used to optimize the alarm triggering logic through logistic regression or decision tree models;

[0119] The alarm notification configuration submodule is used to allow users to customize the recipients and priorities of alarm notifications.

[0120] Figure 3 The figure is a schematic diagram of the structure of an electronic device of the present invention.

[0121] In some embodiments of the present invention, Figure 3 As shown, an electronic device is provided, which includes: electronic devices such as desktop computers, notebooks, handheld computers, and cloud servers. The electronic device 3 may include but is not limited to a processor 301 and a memory 302. Those skilled in the art will understand that Figure 3 The electronic device 3 is merely an example and does not limit the electronic device 3 , and may include more or less components than those shown in the figure, or different components.

[0122] The processor 301 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0123] The memory 302 may be an internal storage unit of the electronic device 3, for example, a hard disk or memory of the electronic device 3. The memory 302 may also be an external storage device of the electronic device 3, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 3. The memory 302 may also include both an internal storage unit of the electronic device 3 and an external storage device. The memory 302 is used to store computer programs and other programs and data required by the electronic device.

[0124] In some embodiments of the present invention, a computer-readable storage medium is provided, which implements the steps of the above method when executed by the processor 301. Therefore, the computer-readable storage medium provided by the present invention has all the technical effects of the above steps and will not be repeated here.

[0125] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0126] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0127] In the embodiments provided in the present disclosure, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, which may be electrical, mechanical or other forms.

[0128] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present disclosure implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0129] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A behavior and psychological state monitoring system, characterized in that: Include: A video data acquisition module, configured to continuously capture video data in a specific environment using at least one commercial camera, wherein the specific environment includes a school, a nursing home, a home or a workplace of an elderly person or child living alone; An image processing module, used for preprocessing the video data; The behavior analysis module is used to extract key visual features based on the preprocessed image data using convolutional neural networks and long short-term memory networks to identify individual behavior patterns. The key visual features include: individual movement trajectory, posture changes, gestures, facial expressions, and interactions with the surrounding environment; A psychological state assessment module, used to assess the psychological state of an individual based on the analysis results of the behavioral pattern by using a support vector machine algorithm; The early warning mechanism module is used to generate an alarm based on the psychological state assessment result and notify relevant personnel.

2. The behavior and psychological state monitoring system according to claim 1, characterized in that: The video data acquisition module comprises: The camera configuration submodule is used to adjust the camera's resolution, frame rate, sensitivity, and viewing angle; A data preprocessing submodule, for encrypting and time-stamping the video data captured after adjusting the camera; The data storage and transmission submodule is used to compress and securely transmit the processed video data.

3. The behavior and psychological state monitoring system according to claim 1, characterized in that: The image processing module comprises: Noise filtering submodule, used to reduce random noise in the preprocessed video; A contrast enhancement submodule, for enhancing the contrast of the pre-processed video image by histogram equalization; The object tracking submodule is used to identify key visual features in the pre-processed video through edge detection and object tracking techniques.

4. The behavior and psychological state monitoring system according to claim 1, characterized in that: The behavior analysis module includes: A feature extraction submodule, for extracting the key visual features from the preprocessed video frames using a convolutional neural network; A sequence analysis submodule, for analyzing a behavioral feature sequence of the key visual features through a long short-term memory network; The behavior recognition and classification submodule is used to use a support vector machine to recognize the behavior pattern from the behavior feature sequence and obtain a behavior feature vector.

5. The behavior and psychological state monitoring system according to claim 1, characterized in that: The mental state assessment module includes: A classification submodule, used for classifying the behavior feature vector; The probability model submodule is used to analyze the probability of psychological state changes for various behavioral feature vectors using the hidden Markov model; The decision support submodule is used to provide a basis for psychological state inference based on the probability of the psychological state change through historical data and real-time data to obtain a psychological state assessment result.

6. The behavior and psychological state monitoring system according to claim 1, characterized in that: The early warning mechanism module includes: A threshold setting submodule, used to set a threshold for triggering an alarm according to the psychological state assessment result; The alarm logic optimization submodule is used to optimize the alarm triggering logic through logistic regression or decision tree models; The alarm notification configuration submodule is used to allow users to customize the recipients and priorities of alarm notifications.

7. A method for monitoring behavior and psychological state, characterized in that: include: Capturing video data using a commercial camera, and encrypting and time-stamping the video data; By preprocessing the video data; Based on the preprocessed image data, a convolutional neural network is applied to identify the behavior pattern of the individual and generate a corresponding behavior data set; Based on the behavior data in the corresponding behavior data set, the psychological state of the individual is evaluated to generate an evaluation result, wherein the psychological state evaluation result is represented by a psychological state score; An alert is generated based on the evaluation results, and relevant personnel are notified to intervene.

8. The behavior and psychological state monitoring method according to claim 7, characterized in that: The step of applying a convolutional neural network to identify individual behavior patterns based on the preprocessed image data to generate a corresponding behavior data set includes: Based on the pre-processed image data, a convolutional neural network is used to analyze individual behaviors in real time to identify individual behavior patterns; Based on the individual's behavior pattern, a corresponding behavior data set is generated, and specific behavior events are marked for each behavior data set.

9. The behavior and psychological state monitoring method according to claim 7, characterized in that: The method of evaluating the individual's psychological state based on the behavioral data in the corresponding behavioral data set and generating an evaluation result includes: based on the behavioral data in the corresponding behavioral data set, combining multiple psychological models, conducting a multi-dimensional evaluation of the individual's psychological state and generating a psychological state report, wherein the psychological model includes an emotion model and a behavior model.

10. The behavior and psychological state monitoring method according to claim 7, characterized in that: The generating of an alarm based on the evaluation result and notifying relevant personnel to intervene includes: Set up multi-level alarms to generate corresponding levels of intervention notifications based on different psychological status assessment results; Based on the corresponding level of intervention notification, relevant personnel are notified to intervene, where the expression for generating the alarm is as follows: Alert(t)=1{E(t)>τ} Among them, Alert(t) indicates whether the alarm is triggered, E(t) represents the psychological state assessment result, that is, the psychological state score, τ represents the preset risk threshold, and 1 represents the indicator function, which is used to determine whether the threshold is exceeded.

Images