Psychological consultation pavilion and intelligent analysis system thereof

Abstract

The invention relates to the field of intelligent analysis, and discloses a psychological counseling pavilion and an intelligent analysis system.The intelligent analysis system comprises a user data collection module, and the user data collection module comprises a physiological data collection unit, a brain wave detection unit and a data processing unit, the behavior data collection unit is used for collecting facial expression and body language behavior data of the user by using a monitoring camera; the user data analysis module analyzes the physiological data and the behavior data of the user to obtain a user emotional health score, and compares the user emotional health score with a plurality of preset score intervals to judge the psychological anxiety degree of the user, and the psychological anxiety degree of the user is divided into a normal level, a mild level, a severe level and a severe level; the personalized coping module provides personalized psychological suggestions and intervention measures according to the psychological anxiety degree of the user; and the user early warning prompt module automatically gives out an early warning prompt when judging that the user is at the psychological anxiety degree.

Description

Technical Field

[0001] The present invention relates to the field of intelligent analysis, and particularly to a psychological counseling kiosk and its intelligent analysis system. Background Art

[0002] Traditional psychological counseling services mainly rely on psychological counseling institutions or the psychiatry departments of hospitals, and there are many limitations. On the one hand, these places are often relatively concentrated geographically, making it extremely inconvenient for people living in remote areas or with limited mobility to obtain services. On the other hand, the time of counseling services is usually relatively fixed, making it difficult to meet the sudden psychological needs of people during non-working hours. In addition, during face-to-face counseling, some clients may be reluctant to open up due to concerns about privacy leakage or social prejudice.

[0003] With the development of technology, some online psychological counseling platforms have emerged, alleviating the limitations of time and space to a certain extent. However, online platforms lack the authenticity and atmosphere of face-to-face communication. It is difficult for counselors to capture non-verbal information of clients, such as facial expressions and body language, which are crucial for accurately judging the psychological state of clients. Moreover, it is difficult to ensure absolute privacy and quietness in the online environment, and it is easily interfered by the outside world, affecting the counseling effect.

[0004] In order to overcome the deficiencies of traditional psychological counseling methods, the research and development of psychological counseling kiosks and their intelligent analysis systems that integrate advanced technologies are particularly urgent. Summary of the Invention

[0005] The purpose of the present invention is to provide a psychological counseling kiosk and its intelligent analysis system to solve the above technical problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A psychological counseling kiosk, comprising:

[0008] A kiosk body, made of soundproof materials, having a closed space for blocking external noise and providing a private environment for users;

[0009] A seat for users to sit on during the counseling process;

[0010] A heart rate monitor and an electroencephalogram detector for collecting physiological data of users during the counseling process;

[0011] A surveillance camera for collecting image data of users during the counseling process;

[0012] Consultation communication module, the consultation communication module includes a video call device and a voice message device, the video call device is used to connect with a psychological counselor to achieve remote real-time consultation, and the voice message device is used for users to leave questions during non-consultation periods and wait for the psychological counselor to reply;

[0013] Psychological assessment module, the psychological assessment module has a variety of psychological assessment scales built-in, and is equipped with a touch screen for users to conduct assessments and automatically generate assessment reports.

[0014] An intelligent analysis system applied to the psychological consultation kiosk, including:

[0015] User data collection module, the user data collection module includes:

[0016] Physiological data collection unit, used to connect a heart rate monitor and an electroencephalogram detector to collect users' physiological data;

[0017] Behavior data collection unit, using a surveillance camera to collect users' facial expressions and body language behavior data;

[0018] User data analysis module, used to analyze users' physiological data and behavior data, obtain a user emotional health score, and then compare the user emotional health score with multiple preset score intervals to judge the degree of users' psychological anxiety. The degree of users' psychological anxiety is arranged in ascending order as four levels: normal, mild, severe, and extremely severe;

[0019] Personalized coping module, used to provide personalized psychological suggestions and intervention measures according to the degree of users' psychological anxiety;

[0020] User warning prompt module, automatically issues a warning prompt when it is judged that the user is in an extremely severe psychological anxiety state.

[0021] As a further technical solution, the method for obtaining the user emotional health score is:

[0022] Process the users' physiological data and behavior data respectively to obtain the user physiological index D k and the user behavior index D f ;

[0023] Through the formula:

[0024]

[0025] Calculate to obtain the user emotional health score G kf ;

[0026] where, μ 1 、μ 2 are weight coefficients, ρ 1 、ρ 2is a preset proportionality coefficient, D kc , D fc are respectively the reference values of the user's physiological indicators and behavior coefficients. i is the i-th sampling, and n represents the total number of samplings within a unit time.

[0027] As a further technical solution, the process of obtaining the user's physiological indicators is as follows:

[0028] Substitute the collected user heart rate, blood pressure, respiratory rate, and electroencephalogram parameters into the formula:

[0029]

[0030] Calculate to obtain the user's physiological indicator D k ;

[0031] Among them, A(t) is the curve of heart rate changing with time, A c is the reference heart rate, t j , t j+1 are respectively the starting point and ending point of the monitoring period, ΔP h is the diastolic blood pressure coefficient, ΔP l is the systolic blood pressure coefficient, B Δ is the electroencephalogram coefficient, S(t) is the curve of respiratory rate changing with time, S c is the reference respiratory rate, τ 1 , τ 2 are reference coefficients.

[0032] As a further technical solution, the process of obtaining the electroencephalogram coefficient is as follows:

[0033] Input the powers of the obtained α, β, γ, θ, and δ waves into a pre-trained recurrent neural network model, and output to obtain the electroencephalogram coefficient B Δ ;

[0034] Among them, the recurrent neural network model is a long short-term memory network.

[0035] As a further technical solution, the process of obtaining the user's behavior indicators is as follows:

[0036] Compare the calculated user physiological indicator D k with the preset user physiological indicator threshold D kth ;

[0037] If D k ≥D kth , then it is initially judged that the current user has a tendency to have psychological problems;

[0038] Otherwise, it is initially judged that the current user does not have a tendency to have psychological problems;

[0039] When condition D k ≥D kth is satisfied, continuously collect facial expression pictures of the current user through a surveillance camera at a set frame rate. If the current user has body movements, obtain body movement pictures;

[0040] The process of obtaining body movement pictures is as follows:

[0041] Calculate the image difference between consecutive frames through computer vision algorithms. If the difference exceeds the set threshold, it is considered that a body movement has occurred; once a body movement is detected, intercept the corresponding image from the frame sequence of the surveillance camera as the body movement picture;

[0042] After preprocessing the facial expression pictures and body movement pictures, obtain standard facial expression pictures and body movement pictures;

[0043] Then input the facial expression pictures and body movement pictures into the trained convolutional neural network model to output the user's expression and body movement categories;

[0044] Through the formula:

[0045]

[0046] Calculate the user behavior metric D f ;

[0047] x f is the facial expression category mapping score, y b is the body movement category mapping score, I f is the facial expression intensity value, I b is the body movement intensity value, and r is the conversion coefficient.

[0048] As a further technical solution, the process of obtaining the facial expression intensity value I f is as follows:

[0049] Obtain the set of facial feature points detected at time t1, which is

[0050] Obtain the set of facial feature points detected at time t2, which is

[0051] Through the formula:

[0052]

[0053] Calculate the displacement of each facial feature point Then substitute it into the following formula:

[0054]

[0055] Calculate the facial expression intensity value I f ;

[0056] Among them, is the weight coefficient of the th facial feature point.

[0057] As a further technical solution, the process of obtaining the limb movement intensity value I b is as follows:

[0058] Obtain the limb key point coordinates of the tε-th frame as where δ is the number of limb key points;

[0059] Calculate the sum K of the Euclidean distances between limb key points in two adjacent frames λ , and the expression is:

[0060]

[0061] where λ = 1, 2, 3,..., ζ;

[0062] Substitute into the formula: Calculate the limb movement intensity value I b .

[0063] Advantages of the present invention:

[0064] By separately refining the physiological data and behavioral data of the user to obtain the user physiological index and the user behavior index, the present invention not only relies on a single type of data, but comprehensively considers the information in two dimensions of physiology and behavior, and establishes a more reasonable scoring system; through the fusion of multi-source data, it can more comprehensively and accurately depict the user's psychological anxiety state, and greatly improves the accuracy and comprehensiveness of the evaluation compared with the evaluation method that only relies on a single data source. Description of the drawings

[0065] The present invention will be further described below with reference to the accompanying drawings.

[0066] Figure 1 is a schematic structural diagram of the present invention;

[0067] Figure 2 is a system block diagram of the present invention. Specific embodiments

[0068] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0069] Please refer to Figure 1 as shown, the present invention is a psychological counseling kiosk, including:

[0070] A kiosk body, which is made of soundproof materials and has a closed space for blocking external noises and providing a private environment for users;

[0071] A seat for users to sit on during the consultation process;

[0072] A heart rate monitor and an electroencephalogram detector for collecting physiological data of users during the consultation process;

[0073] A monitoring camera for collecting image data of users during the consultation process;

[0074] A consultation communication module, which includes a video call device and a voice message device. The video call device is used to connect with a psychological counselor to achieve remote real-time consultation, and the voice message device is used for users to leave questions during non-consultation periods and wait for the psychological counselor to reply;

[0075] A psychological assessment module, which is built-in with a variety of psychological assessment scales, equipped with a touch screen for users to conduct assessments, and automatically generates assessment reports.

[0076] In this embodiment, the kiosk body is made of soundproof materials to form a closed space, which can effectively block external noise interference, create a private and quiet consultation environment for users, help users relax physically and mentally, express their inner thoughts and feelings more freely, and improve the consultation effect; the setting of the seat provides a comfortable sitting experience for users during the consultation process. During a long consultation process, a comfortable seat can reduce the physical fatigue of users, make them more focused on the consultation content, and ensure the smooth progress of the consultation; the heart rate monitor and the electroencephalogram detector can collect the physiological data of users during the consultation process in real time. These data can intuitively reflect the physical state and psychological stress level of users. For example, the change in heart rate can detect the fluctuation of users' emotions, and the electroencephalogram data can further reveal the activity state of users' brains, providing an objective physiological basis for psychological counselors, assisting them to analyze the psychological conditions of users more accurately, and formulating more targeted consultation plans; the monitoring camera collects the image data of users during the consultation process, which can capture information such as users' facial expressions and body languages; facial expressions and body languages are the external manifestations of psychological states, which helps counselors understand the emotions and attitudes that users do not express through words, supplement the deficiencies of verbal information, and improve the comprehensiveness and accuracy of the assessment of users' psychological states;

[0077] Meanwhile, the psychological assessment module incorporates a variety of psychological assessment scales that cover different aspects of psychological traits and problems, enabling quantitative assessment of the user's psychological state from multiple dimensions. Through the touch screen, users can independently conduct assessments easily. They can choose appropriate scales according to their own situations for testing to comprehensively understand their psychological conditions. After the assessment is completed, an assessment report is automatically generated, providing intuitive and systematic assessment results for users and counselors. The content of the report can serve as an important reference for counseling, helping counselors quickly understand the user's psychological characteristics and problems, formulating more targeted counseling plans, and at the same time enabling users to have a clearer understanding of their own psychological state and actively cooperate with the counseling work. This invention is more suitable for people whose psychological problems have not reached an extreme state and those who are trapped in psychological problems but are unaware of it.

[0078] Please refer to Figure 2 as shown, an intelligent analysis system applied to the psychological counseling kiosk, including:

[0079] User data collection module, the user data collection module includes:

[0080] Physiological data collection unit, used to connect a heart rate monitor and an electroencephalogram detector to collect the user's physiological data;

[0081] Behavior data collection unit, using a surveillance camera to collect the user's facial expressions and body language behavior data;

[0082] User data analysis module, used to analyze the user's physiological data and behavior data to obtain the user's emotional health score, and then compare the user's emotional health score with multiple preset score intervals to judge the user's psychological anxiety level. The user's psychological anxiety level is arranged in ascending order as four levels: normal, mild, severe, and critical.

[0083] Personalized coping module, used to provide personalized psychological suggestions and intervention measures according to the user's psychological anxiety level; the personalized coping module retrieves corresponding personalized psychological suggestions and intervention measures from a preset coping strategy library according to the judgment result of severe anxiety. For example, provide some deep relaxation training methods, such as progressive muscle relaxation method, and suggest that the user practice meditation regularly. At the same time, recommend some relevant psychological self-help books and online courses.

[0084] User warning and reminder module, automatically issues a warning reminder when it is judged that the user is in a critical psychological anxiety level. The reminder method is the flashing of lights and graphic prompts in the kiosk, informing the user of their current psychological anxiety level and reminding them to take corresponding measures according to the suggestions. At the same time, the warning information will also be sent to the counselor's terminal device so that the counselor can contact the user in time to provide further assistance.

[0085] The method for obtaining the user's emotional health score is as follows:

[0086] Process the physiological data and behavioral data of the user respectively to obtain the user's physiological index D k and the user's behavioral index D f ;

[0087] Through the formula:

[0088]

[0089] Calculate to obtain the user's emotional health score G kf ;

[0090] Among them, μ 1 、μ 2 are weight coefficients, formulated comprehensively based on historical data and empirical data, ρ 1 、ρ 2 are preset proportionality coefficients, determined based on the analysis of historical data and experimental data, D kc 、D fc are the reference values of the user's physiological index and behavioral coefficient respectively, i is the i-th sampling, and n represents the total number of samplings within a unit time.

[0091] In this embodiment, by refining the physiological data and behavioral data of the user respectively to obtain the user's physiological index D k and the user's behavioral index D f , it does not rely solely on a single type of data, but comprehensively considers the information in two dimensions of physiology and behavior to establish a more reasonable scoring system. Among them, physiological data reflects the internal stress response of the body, such as changes in indicators such as heart rate and brain waves, which can intuitively reflect the physiological response of the body to the mental state; behavioral data provides supplementary information for the assessment of the mental state from external manifestations, such as facial expressions and body language; through the fusion of multi-source data, it can more comprehensively and accurately depict the user's psychological anxiety state, and compared with the assessment method that only relies on a single data source, it greatly improves the accuracy and comprehensiveness of the assessment.

[0092] The process of obtaining the user's physiological index is as follows:

[0093] Substitute the collected user's heart rate, blood pressure, respiratory rate, and brain wave parameters into the formula:

[0094]

[0095] Calculate to obtain the user's physiological index D k ;

[0096] Among them, A(t) is the curve of heart rate changing with time, A c is the reference heart rate, t k, t j+1 are the start and end points of the monitoring period respectively, and ΔP h is the diastolic blood pressure coefficient, and ΔP l is the systolic blood pressure coefficient, B Δ is the electroencephalogram coefficient, S(t) is the curve of the respiratory frequency changing with time, and S c is the reference respiratory frequency, and τ 1 , τ 2 are reference coefficients; P hmax , P hmin are the upper limit value and the lower limit value of the diastolic blood pressure respectively; P lmax , P lmin are the upper limit value and the lower limit value of the systolic blood pressure respectively, and P h , P l are the measured values of the user's diastolic blood pressure and systolic blood pressure respectively.

[0097] The process of obtaining the electroencephalogram coefficient is as follows:

[0098] Input the powers of the obtained α, β, γ, θ, and δ waves into a pre-trained recurrent neural network model, and output to obtain the electroencephalogram coefficient B Δ ;

[0099] Among them, the recurrent neural network model is a long short-term memory network.

[0100] In this embodiment, a specific method for obtaining the user's physiological indicators is provided. First, collect the user's heart rate, blood pressure, respiratory frequency, and electroencephalogram parameters; then substitute the above parameters into the formula to calculate the user's physiological indicator D k . It can be clearly seen from the above formula that the closer the diastolic blood pressure coefficient and the systolic blood pressure coefficient are to 1, the more normal the user's blood pressure is, otherwise it means it is more abnormal; similarly, the larger the electroencephalogram coefficient B Δ , the more normal the user's electroencephalogram power is, otherwise it means the user's electroencephalogram power is more abnormal; and the user's heart rate and respiratory frequency are reflected by the formula . Obviously, if the deviation of the user's heart rate from the reference value within the monitoring period is smaller, the closer it is to 1, indicating that the user's heart rate and respiratory frequency are more normal, otherwise it is more abnormal;

[0101] Through the above technical solution, it is possible to score comprehensively based on heart rate, blood pressure, respiratory rate, and different brain wave powers, and establish a user physiological model through comprehensive evaluation according to the corresponding weights. The score range for each parameter evaluation is between 0 and 1, and the closer the score is to 1, the closer the indicator is to the normal range. For example, if the user's heart rate is 85 beats per minute, blood pressure is 120 / 80 mmHg, respiratory rate is 15 breaths per minute, and the brain wave powers are within the ideal range, then the scores for each item will be relatively high, and the comprehensive score will also be high, indicating that the user's physiological state is relatively good; conversely, if some indicators deviate significantly from the normal range, the corresponding scores will decrease, and the comprehensive score will also decrease, suggesting that there may be physiological or psychological abnormalities.

[0102] The process of obtaining the user behavior indicators is as follows:

[0103] Compare the calculated user physiological indicator D k with the pre-set user physiological indicator threshold D kth ;

[0104] If D k ≥D kth , then initially judge that the current user has a tendency of psychological problems;

[0105] Otherwise, initially judge that the current user does not have a tendency of psychological problems;

[0106] When the condition of D k ≥D kth is satisfied, continuously collect the facial expression pictures of the current user at the set frame rate through the monitoring camera. If the current user has body movements, obtain the body movement pictures;

[0107] The process of obtaining the body movement pictures is as follows:

[0108] Through computer vision algorithms, such as the optical flow method, calculate the image differences between consecutive frames. If the difference exceeds the set threshold, it is considered that a body movement has occurred; once a body movement is detected, intercept the corresponding image from the frame sequence of the monitoring camera as the body movement picture; it should be noted that the computer vision algorithm used here is a mature existing technology, so it will not be elaborated further;

[0109] After preprocessing the facial expression pictures and body movement pictures, obtain the standard facial expression pictures and body movement pictures;

[0110] Then input the facial expression pictures and body movement pictures into the trained convolutional neural network model to output the user's expression and body movement categories;

[0111] Through the formula:

[0112]

[0113] Calculate the user behavior metric D f ;

[0114] x f is the mapping score of facial expression categories, y b is the mapping score of body movement categories, I f is the facial expression intensity value, i b is the body movement intensity value, and r is the conversion coefficient.

[0115] In this embodiment, by comparing the calculated user physiological metrics with pre-set thresholds, it is possible to quickly and preliminarily determine whether a user has a tendency towards psychological problems, providing an effective screening for subsequent behavioral data collection, avoiding unnecessary behavioral data collection for users without a tendency towards psychological problems, and improving the operating efficiency of the entire system; at the same time, for users who may have psychological problems, it is possible to promptly initiate further behavioral monitoring processes, enabling early problem detection and providing the possibility for timely intervention; and by using mature computer vision algorithms such as the optical flow method to calculate the image differences between consecutive frames to determine the occurrence of body movements, it has high accuracy and real-time performance, can accurately capture the subtle body movements of users, and once a movement is detected, immediately intercept the corresponding image from the frame sequence of the surveillance camera to ensure that the captured body movement pictures completely and accurately reflect the user's behavior, providing a reliable data basis for subsequent analysis;

[0116] Through the formula Calculate the user behavior metric D f , which integrates the information conveyed by facial expressions and body movements, comprehensively quantifies the user's behavior performance from multiple dimensions. Compared with analyzing facial expressions or body movements alone, the comprehensive behavior metric can more comprehensively reflect the user's psychological state and behavior characteristics, providing richer and more accurate information for psychological counseling and intervention;

[0117] It should be noted that: the mapping scores of facial expression categories and body movement categories are obtained by looking up the mapping table established based on empirical data and historical data;

[0118] Different facial expression categories have different psychological anxiety tendencies. A mapping table can be established, for example: Happy: 0 points (indicating a decrease in anxiety level); Sad: +20 points; Angry: +30 points; Surprised: +10 points; Fear: +40 points; Disgusted: +25 points. Different body movement categories and intensities can also be mapped to different scores, for example: Raising the hand: +15 points; Turning around: +25 points; Nodding: -15 points; -20 points.

[0119] The process of obtaining the facial expression intensity value I f is as follows:

[0120] Obtain the set of facial feature points detected at time t1, which is

[0121] Obtain the set of facial feature points detected at time t2, which is

[0122] Through the formula:

[0123]

[0124] Calculate the displacement of each facial feature point Then substitute it into the following formula:

[0125]

[0126] Calculate the facial expression intensity value I f ;

[0127] Among them, is the weight coefficient of the th facial feature point.

[0128] In this embodiment, the change of facial expression is usually accompanied by the movement of facial muscles, and these movements can be reflected by the displacement of facial feature points. First, use a facial feature point detection algorithm (such as the 68-point facial feature point detection in the dlib library) to detect the key feature points of the face, and these feature points usually include the key points of parts such as eyebrows, eyes, mouth, nose, etc.;

[0129] The generation of facial expressions stems from the contraction and relaxation of facial muscles, and these muscle movements directly cause the displacement of facial feature points; calculating the expression intensity by tracking the feature point displacement is closely related to the physiological mechanism of expression generation; for example, when smiling, the corners of the mouth turn up and the eye corner muscles contract, which will change the positions of the corresponding feature points, and the feature point displacement amount can intuitively reflect the degree of these muscle movements, thus directly reflecting the essence of expression changes and laying a foundation for accurately measuring the expression intensity; at the same time, it can real-time track the positions of facial feature points at different times and accurately capture the dynamic change process of expressions from start to end; compared with some other static analysis methods, based on feature point displacement, the fluctuation of expression intensity can be real-time monitored. For example, in a conversation, the instantaneous fluctuations of the user's expression intensity with the change of topics can be timely captured, providing strong support for analyzing the real-time evolution of emotions; finally, converting the facial expression intensity into specific numerical values, by calculating the average value or weighted average value of feature point displacements, the quantification of expression intensity is realized. For example, when comparing the expression intensities of different users or the same user at different times, accurate comparisons can be made based on the quantified values, providing a standardized measurement basis for fields such as psychological research and human-computer interaction.

[0130] Obtain the limb movement intensity value I b The process is as follows:

[0131] Obtain the coordinates of limb key points at the tε-th frame as where ζ is the number of limb key points;

[0132] Calculate the sum K of the Euclidean distances between limb key points in two adjacent frames λ , and the expression is:

[0133]

[0134] where λ = 1, 2, 3, …, ζ;

[0135] Substitute into the formula: Calculate to obtain the limb movement intensity value I b .

[0136] In this embodiment, calculating the limb movement intensity value through the Euclidean distance is relatively objective, reducing the interference of subjective factors in human judgment; it provides a unified measurement standard for the limb movement intensity. No matter what type of limb movement, the corresponding intensity value can be calculated through the Euclidean distance, thus facilitating direct comparison between different movements, different individuals, or different movement performances of the same individual.

[0137] It should be noted that: the calculation formula and each parameter participating in the operation in the present invention have been pre-dimensionless processed, and the process of dimensionless processing is well-known in the industry and will not be described here.

[0138] The above has described a detailed description of an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as used to limit the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the patent of the present invention.

Claims

1. A psychological consultation booth, characterized in that: include: The pavilion body is made of sound-proof material and has a closed space for blocking external noise and providing a private environment for users; A seat for users to sit during the consultation process; Heart rate monitors and brain wave detectors are used to collect users’ physiological data during the consultation process; Monitoring cameras, used to collect image data of users during the consultation process; A consultation communication module, which includes a video call device and a voice message device. The video call device is used to connect to a psychological counselor to achieve remote real-time consultation, and the voice message device is used for users to leave questions during non-consultation periods and wait for the psychological counselor to reply; The psychological assessment module has a variety of built-in psychological assessment scales and is equipped with a touch screen for users to conduct assessments and automatically generates assessment reports.

2. An intelligent analysis system applied to the psychological counseling booth described in claim 1, characterized in that: include: A user data collection module, wherein the user data collection module comprises: A physiological data collection unit, used to connect to a heart rate monitor and an electroencephalogram detector to collect the user's physiological data; A behavior data collection unit, which uses surveillance cameras to collect the user's facial expression and body language behavior data; A user data analysis module is used to analyze the user's physiological data and behavioral data to obtain the user's emotional health score, and then compare the user's emotional health score with a plurality of preset score intervals to determine the user's psychological anxiety level, wherein the user's psychological anxiety level is arranged in ascending order into four levels: normal, mild, moderate, and severe; Personalized coping module, used to provide personalized psychological advice and intervention measures based on the user's psychological anxiety level; The user early warning prompt module automatically issues an early warning prompt when it determines that the user is in a serious level of psychological anxiety.

3. The intelligent analysis system according to claim 2, characterized in that: The method for obtaining the user's emotional health score is: Process the user's physiological data and behavioral data separately to obtain the user's physiological index D k and user behavior indicators D f ; By formula: Calculate the user's emotional health score G kf ; Among them, μ1 and μ2 are weight coefficients, ρ1 and ρ2 are preset proportional coefficients, and D kc , D fc are the reference values ​​of the user's physiological indicators and behavior coefficients, respectively; i is the i-th sampling; and n represents the total number of sampling times per unit time.

4. The intelligent analysis system according to claim 3, characterized in that: The process of obtaining the user's physiological indicators is as follows: Substitute the collected user's heart rate, blood pressure, breathing rate and brain wave parameters into the formula: Calculate the user's physiological index D k ; A(t) is the curve of heart rate changing with time, A c is the reference heart rate, t j ,t j+1 are the starting point and end point of the monitoring period, ΔP h is the diastolic pressure coefficient, ΔP l is the systolic blood pressure coefficient, B Δ is the brain wave coefficient, S(t) is the curve of respiratory frequency changing with time, S c is the reference respiratory rate, τ1 and τ2 are reference coefficients; P hmax , P hmin are the upper and lower limits of diastolic blood pressure, respectively; P lmax , P lmin are the upper and lower limits of systolic blood pressure, respectively. h , P l They are the actual measured values ​​of the user's diastolic and systolic blood pressure respectively.

5. The intelligent analysis system according to claim 4, characterized in that: The process of obtaining the brain wave coefficient is as follows: The power of the acquired α, β, γ, θ, and δ waves is input into the pre-trained recurrent neural network model, and the output is the brain wave coefficient B Δ ; Wherein, the recurrent neural network model is a long short-term memory network.

6. The intelligent analysis system according to claim 2 or 4, characterized in that: The process of obtaining the user behavior indicator is as follows: The calculated user physiological index D k Compared with the preset user physiological index threshold D kth Make a comparison; If D k ≥D kth , it is preliminarily judged that the current user has a tendency to have psychological problems; Otherwise, it is preliminarily determined that the current user does not have a tendency to have psychological problems; When D is satisfied k ≥D kth When the condition is met, the facial expression pictures of the current user are continuously collected by the surveillance camera at the set frame rate. If the current user has body movements, the body movement pictures are obtained; The process of obtaining body movement pictures is: The image difference between consecutive frames is calculated through computer vision algorithms. If the difference exceeds a set threshold, it is considered that a body movement has occurred. Once a body movement is detected, the corresponding image is captured from the frame sequence of the surveillance camera as a body movement picture. After preprocessing the facial expression pictures and body movement pictures, standard facial expression pictures and body movement pictures are obtained; Then input the facial expression pictures and body movement pictures into the trained convolutional neural network model, and output the user's expression and body movement categories; By formula: Calculate the user behavior index D f ; x f Map scores for facial expression categories, y b Mapping scores for body action categories, I f is the facial expression intensity value, I b is the limb movement intensity value, and r is the conversion coefficient.

7. The intelligent analysis system according to claim 6, characterized in that: Get the facial expression intensity value I f The process is: Get the facial feature point set detected at time t1, which is Get the facial feature point set detected at time t2, By formula: Calculate the displacement of each facial feature point Substitute the following formula: Calculate the facial expression intensity value I f ; in, For the The weight coefficient of each facial feature point.

8. The intelligent analysis system according to claim 6, characterized in that: Get the limb movement intensity value i b The process is: Get the coordinates of the limb key points in the tεth frame: Where ζ is the number of limb key points; Calculate the sum of the Euclidean distances K of the limb key points between two adjacent frames λ , the expression is: Among them, λ = 1, 2, 3,..., ζ; Substituting into the formula: Calculate the limb movement intensity value I b .

Images