Teenager psychological behavior transaction monitoring and early warning method based on machine learning

Abstract

The invention relates to the technical field of psychological health monitoring and early warning, in particular to a teenager psychological behavior transaction monitoring and early warning method based on machine learning, which performs early recognition on the psychological health state of a teenager user through a machine learning model, timely discovers potential psychological health problems, and improves the accuracy of the teenager psychological behavior transaction monitoring and early warning. Intervention measures can be taken before problem deterioration, long-period and non-perceptual mental health fluctuation tracking is carried out through monitoring videos, under the condition that normal life of teenagers is not affected, the mental health conditions of the teenagers are continuously monitored, early warning is provided in time once symptom development or deterioration is found, rapid response and intervention are facilitated, and the mental health monitoring method is suitable for being popularized and applied. Dangerous area video management and control are carried out on teenager users in abnormal psychological states, dangerous behaviors possibly occurring due to psychological health problems of the teenager users are effectively prevented, early warning is immediately provided and security personnel are reminded to deal with the dangerous behaviors once the teenager users enter the dangerous area, and the occurrence probability of accidents can be greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of mental health monitoring and early warning, and in particular to a method for monitoring and early warning of abnormal psychological and behavioral changes in adolescents based on machine learning. Background Art

[0003] Non-suicidal self-injury and suicide among adolescents is a very common and serious public health problem. It is necessary to pay attention to, detect, study and warn adolescent mental health. Conventional mental health prediction is carried out through questionnaire surveys, which is inefficient and has a limited sample size. It can only cover a small number of people and is difficult to fully and accurately reflect the situation of the entire population.

[0004] The existing publication number CN116759081A discloses a method for monitoring and warning of abnormal psychological behavior of adolescents based on machine learning, including the following steps: obtaining the physical sign data, emotional data and psychological test self-assessment form data of adolescent users; extracting the psychological behavior characteristic data of adolescent users based on the physical sign data, emotional data and psychological test self-assessment form data of adolescent users; using the pre-built psychological behavior monitoring and early warning model to output the mental health status corresponding to the current psychological behavior characteristic data; grading and classifying the output mental health status of adolescent users, and issuing crisis warnings to adolescent users with abnormal mental states based on the grading and classification results. The comprehensive monitoring and early warning of adolescents' mental health based on machine learning has the automation and efficiency of data collection and processing compared to traditional questionnaire surveys, and can provide more accurate and timely early warning and intervention measures.

[0005] However, the above-mentioned machine learning-based method for monitoring and warning of abnormal psychological and behavioral changes in adolescents relies on monitoring personnel for early warning and intervention, but the monitoring personnel have limited energy and cannot provide all-weather and comprehensive monitoring. As a result, when adolescents engage in abnormal behavior, they cannot be discovered in time, leading to accidents. Summary of the invention

[0007] The purpose of the present invention is to provide a method for monitoring and early warning of abnormal psychological and behavioral changes in adolescents based on machine learning, which solves the problem that adolescents' abnormal behaviors cannot be discovered in time, leading to accidents.

[0008] To achieve the above object, the present invention provides a method for monitoring and early warning of adolescent psychological and behavioral abnormalities based on machine learning, comprising the following steps: Collect data on adolescent users, output the corresponding mental health status through machine learning models, and screen out adolescent users with abnormal mental states; Through monitoring videos, the app can track the mental health fluctuations of adolescent users with abnormal mental states over a long period of time without being perceived, and provide timely monitoring and early warning of symptom development; Conduct video control of dangerous areas for young users with abnormal psychological states. Once the person enters, an early warning will be provided to remind security personnel to deal with the situation at the front, and the person's trajectory retrieval capability will also be provided.

[0009] Among them, by monitoring the video, long-term and imperceptible psychological health fluctuations of adolescent users with abnormal psychological states are tracked, and symptom development monitoring and early warning are provided in a timely manner. The steps also include: Install dedicated cameras in key areas; Collect video data through the camera and perform emotion recognition based on the video data; Set a warning threshold, and when the subject's emotional state exceeds the threshold, a warning signal will be automatically issued.

[0010] Wherein, the video data is collected by a camera, and emotion recognition is performed based on the video data, and the steps further include: Clean and preprocess the collected video data to remove noise and outliers; Analyze the preprocessed data to identify the emotional state of the subjects and determine whether there are abnormal emotions; Long-term follow-up is conducted on subjects with abnormal emotions to record their emotional change trends and symptom development.

[0011] Wherein, a warning threshold is set, and when the emotional state of the subject exceeds the threshold, a warning signal is automatically issued, and the steps further include: Early warning signals include sound and light alarms, and information pushed to relevant personnel’s mobile phones or computer terminals.

[0012] Among them, video control of dangerous areas is performed for young users with abnormal psychological states. Once the person enters, an early warning is provided to remind security personnel to deal with it in advance, and the person trajectory retrieval capability is provided. The steps also include: Establish a personnel database containing information about adolescent users with abnormal psychological conditions, and use face recognition technology to compare video data with the personnel database in real time; When a young user with a suspected abnormal psychological state is detected entering a dangerous area, the warning mechanism is immediately triggered; After receiving the early warning, the security personnel immediately rushed to the scene to deal with the situation according to the preset emergency plan.

[0013] When a young user with a suspected abnormal psychological state is detected to have entered a dangerous area, an early warning mechanism is immediately triggered, and the steps further include: The early warning information is conveyed to security personnel and relevant departments through sound and light alarms, SMS notifications, and APP push.

[0014] After receiving the warning, the security personnel will immediately go to the scene to deal with it according to the preset emergency plan. The steps also include: Trace the action trajectory of the young user before entering the dangerous area, and analyze the trajectory data to understand the user's activity patterns and possible risk factors, providing a reference for subsequent intervention measures.

[0015] A kind of monitoring and early warning method of abnormal psychological behavior of adolescents based on machine learning of the present invention, through the machine learning model, the psychological health status of adolescent users is identified early, and potential psychological health problems can be discovered in time, which helps to take intervention measures before the problem worsens, reduce the negative impact of psychological health problems on adolescent growth and development, and identify adolescent users with abnormal psychological states, and provide them with personalized psychological health care and support. This includes psychological counseling, treatment advice, educational resources, etc., which help them better understand and deal with their mental health problems. Using monitoring video to track long-term, imperceptible psychological health fluctuations, it is possible to continuously monitor their mental health without affecting the normal life of adolescents. Once symptoms develop or deteriorate, timely provide early warnings, which help to respond quickly and intervene. Video control of dangerous areas for adolescent users with abnormal psychological states can effectively prevent dangerous behaviors that may occur due to mental health problems. Once they enter a dangerous area, they immediately provide early warnings and remind security personnel to deal with it in advance, which can greatly reduce the probability of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0018] Figure 1 It is a flowchart of the steps of the method for monitoring and early warning of abnormal psychological and behavioral changes in adolescents based on machine learning according to the first embodiment of the present invention.

[0019] Figure 2 This is a flowchart of the steps of the first embodiment of the present invention for tracking the long-term, imperceptible psychological health fluctuations of adolescent users with abnormal psychological states through monitoring videos, and providing timely symptom development monitoring and early warning.

[0020] Figure 3 This is a flowchart of the steps of the first embodiment of the present invention for conducting video control of dangerous areas for adolescent users with abnormal psychological states. Once the person enters, an early warning is provided to remind security personnel to deal with the situation in advance, and the person trajectory retrieval capability is provided. DETAILED DESCRIPTION

[0022] Embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0023] The first embodiment of the present application is: See also Figures 1 to 3 ,in, Figure 1 It is a flowchart of the steps of the method for monitoring and early warning of abnormal psychological and behavioral changes in adolescents based on machine learning according to the first embodiment of the present invention. Figure 2 This is a flowchart of the steps of the first embodiment of the present invention for tracking the long-term, imperceptible psychological health fluctuations of adolescent users with abnormal psychological states through monitoring videos, and providing timely symptom development monitoring and early warning. Figure 3 This is a flowchart of the steps of the first embodiment of the present invention for conducting video control of dangerous areas for adolescent users with abnormal psychological states. Once the person enters, an early warning is provided to remind security personnel to deal with it in advance, and the ability to retrieve the trajectory of the person is also provided. The present invention provides a method for monitoring and early warning of abnormal psychological behavior of adolescents based on machine learning, which includes the following steps: S100: Collect data of adolescent users, output corresponding mental health status through machine learning model, and screen out adolescent users with abnormal mental status; Specifically, the psychological behavior monitoring and early warning model disclosed in publication number CN116759081A is used to monitor the mental health status. First, the physical sign data, emotional data and psychological test self-assessment form data of the adolescent users are obtained; the psychological behavior characteristic data of the adolescent users are extracted based on the physical sign data, emotional data and psychological test self-assessment form data of the adolescent users; the pre-constructed psychological behavior monitoring and early warning model is used to output the mental health status corresponding to the current psychological behavior characteristic data; the output mental health status of the adolescent users is graded and classified, and crisis warnings are issued to adolescent users with abnormal psychological states based on the grading and classification results, and relevant departments or professional institutions are cooperated to seek professional support for intervention and treatment of psychological problems. After the intervention and treatment, it is necessary to pay attention to the situation of the students, follow up the effect of the crisis intervention and the psychological changes of the students, adjust the intervention measures in time, and screen out adolescent users with abnormal psychological states for further monitoring.

[0024] S200: Through monitoring videos, the app can track the mental health fluctuations of adolescent users with abnormal mental states over a long period of time without being perceived, and provide timely monitoring and early warning of symptom development; S201: Install dedicated cameras in key areas; S202: Collecting video data through a camera, cleaning and preprocessing the collected video data, and removing noise and abnormal values; S203: Analyze the preprocessed data, identify the emotional state of the subject, and determine whether there is abnormal emotion; S204: Conduct long-term follow-up on subjects with abnormal emotions and record their emotional change trends and symptom development; S205: Setting a warning threshold, when the emotional state of the subject exceeds the threshold, a warning signal is automatically issued.

[0025] Specifically, special cameras are installed in key areas such as campus gates, teaching buildings, dormitory buildings, libraries, restaurants, and homes to ensure that the micro-vibration data of the subjects' faces and neck muscles can be captured. The cameras are connected to the emotion recognition system, and the system is configured and debugged to ensure that the system can operate normally and accurately identify the emotional state of the subjects. The emotion recognition system adopts an AI-based dynamic abnormal emotion screening system, relying on advanced AI technology and emotion recognition analysis technology, combined with the Internet of Things smart terminal to achieve real-time, non-contact detection and early warning of the emotional state of the subjects, and can analyze and detect the emotional state of the subjects in real time. The system is started to collect data, including the emotional state and behavioral information of the subjects, and the collected data is cleaned and preprocessed to remove noise and outliers. The preprocessed data is analyzed using the emotion recognition system to identify the emotional state of the subjects and determine whether there are abnormal emotions. Subjects with abnormal emotions are tracked for a long time, their emotional change trends and symptom development are recorded, and an early warning threshold is set. When the emotional state of the subject exceeds the threshold, the system automatically sends a warning signal, which can include sound and light alarms, information push to relevant personnel's mobile phones or computer terminals, etc. Personalized intervention measures are formulated based on early warning signs and symptom development. These measures include psychological counseling, psychotherapy, drug intervention and other methods. A feedback mechanism is established to collect feedback from the subjects and their parents or guardians on the intervention measures so that the intervention plan can be adjusted and optimized in a timely manner. The subjects are tracked and followed up over the long term to evaluate the intervention effect and adjust the intervention strategy.

[0026] S300: Conduct video control of dangerous areas for young users with abnormal psychological states. Once the person enters, an early warning will be provided to remind security personnel to deal with the situation. At the same time, the person's trajectory retrieval capability will be provided.

[0027] S301: Establishing a personnel database containing information of adolescent users with abnormal psychological states, and performing real-time comparison between the video data and the personnel database through face recognition technology; S302: When a young user with a suspected abnormal psychological state is detected to enter a dangerous area, an early warning mechanism is immediately triggered; S303: After receiving the warning, the security personnel will immediately go to the scene to deal with it according to the preset emergency plan; S305: After the handling, the movement trajectory of the adolescent user before entering the dangerous area is traced back, and the activity patterns and possible risk factors of the user are understood by analyzing the trajectory data, so as to provide reference for subsequent intervention measures.

[0028] Specifically, clearly define which areas are "dangerous areas" and set up obvious signs and warning signs in these areas to physically remind all personnel to pay attention. Install high-definition video surveillance cameras in and around dangerous areas to ensure coverage without blind spots. The cameras should have functions such as night vision and zoom so that they can clearly capture images under different lighting conditions. Integrate the video surveillance system with intelligent analysis software to realize automatic recognition and analysis of video images. Set up specific algorithms or models to identify the characteristics of adolescents with abnormal psychological states, such as facial expressions, behavioral patterns, etc., and customize and develop based on knowledge in fields such as psychology and behavior. Establish a database containing information on adolescent users with abnormal psychological states, including but not limited to names, photos, and feature descriptions. Use technologies such as face recognition to compare the video surveillance system with the personnel database in real time. When the video surveillance system detects that a teenage user with suspected abnormal psychological state enters a dangerous area, the early warning mechanism is immediately triggered. The early warning information is quickly conveyed to security personnel and relevant departments through various means such as sound and light alarms, SMS notifications, and APP push. After receiving the early warning, the security personnel immediately go to the scene to deal with it according to the preset emergency plan. During the handling process, security personnel should remain calm and professional, keep a safe distance from adolescents with abnormal psychological states, try to communicate and comfort them, and contact professional psychological counselors or medical institutions for assistance when necessary. During the handling process or afterwards, use the personnel trajectory retrieval function provided by the video surveillance system to trace the movement trajectory of the adolescent user before entering the dangerous area. By analyzing the trajectory data, we can understand the user's activity patterns, possible risk factors, etc., and provide reference for subsequent intervention measures.

[0029] Early identification of the mental health status of adolescent users through machine learning models can timely detect potential mental health problems, which helps to take intervention measures before the problem worsens and reduce the negative impact of mental health problems on the growth and development of adolescents. After identifying adolescent users with abnormal mental states, they can be provided with personalized mental health care and support. This includes psychological counseling, treatment advice, educational resources, etc., which help them better understand and deal with their mental health problems. Using surveillance videos to track long-term, imperceptible mental health fluctuations can continuously monitor their mental health without affecting their normal lives. Once symptoms develop or worsen, timely warnings are provided, which helps to respond and intervene quickly. Video control of dangerous areas for adolescent users with abnormal mental states can effectively prevent dangerous behaviors that may occur due to mental health problems. Once they enter a dangerous area, immediate warnings are provided and security personnel are reminded to deal with it in advance, which can greatly reduce the probability of accidents.

[0030] What is disclosed above is only one or more preferred embodiments of the present application, and cannot be used to limit the scope of rights of the present application. Ordinary technicians in this field can understand that all or part of the processes of implementing the above embodiments and equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A method for monitoring and early warning of abnormal psychological and behavioral changes in adolescents based on machine learning, characterized in that: The following steps are involved: Collect data on adolescent users, output the corresponding mental health status through machine learning models, and screen out adolescent users with abnormal mental states; Through monitoring videos, the app can track the mental health fluctuations of adolescent users with abnormal mental states over a long period of time without being perceived, and provide timely monitoring and early warning of symptom development; Conduct video control of dangerous areas for young users with abnormal psychological states. Once the person enters, an early warning will be provided to remind security personnel to deal with the situation at the front, and the person's trajectory retrieval capability will also be provided.

2. The method for monitoring and early warning of psychological and behavioral abnormalities in adolescents based on machine learning as claimed in claim 1, characterized in that: By monitoring the video, the psychological health fluctuations of adolescent users with abnormal psychological states are tracked for a long period of time and without perception, and symptom development monitoring and early warning are provided in a timely manner. The steps also include: Install dedicated cameras in key areas; Collect video data through the camera and perform emotion recognition based on the video data; Set a warning threshold, and when the subject's emotional state exceeds the threshold, a warning signal will be automatically issued.

3. The method for monitoring and early warning of psychological and behavioral abnormalities in adolescents based on machine learning as claimed in claim 2, characterized in that: The video data is collected by a camera, and emotion recognition is performed based on the video data. The steps further include: Clean and preprocess the collected video data to remove noise and outliers; Analyze the preprocessed data to identify the emotional state of the subjects and determine whether there are abnormal emotions; Long-term follow-up is conducted on subjects with abnormal emotions to record their emotional change trends and symptom development.

4. The method for monitoring and early warning of psychological and behavioral abnormalities in adolescents based on machine learning as claimed in claim 2, characterized in that: Setting a warning threshold, when the subject's emotional state exceeds the threshold, automatically sending a warning signal, the steps also include: Early warning signals include sound and light alarms, and information pushed to relevant personnel’s mobile phones or computer terminals.

5. The method for monitoring and early warning of abnormal psychological and behavioral changes in adolescents based on machine learning as claimed in claim 3, characterized in that: Conduct video control of dangerous areas for adolescent users with abnormal psychological states. Once the person enters, an early warning is provided to remind security personnel to deal with it in advance, and the person's trajectory retrieval capability is provided. The steps also include: Establish a personnel database containing information about adolescent users with abnormal psychological conditions, and use face recognition technology to compare video data with the personnel database in real time; When a young user with a suspected abnormal psychological state is detected entering a dangerous area, the warning mechanism is immediately triggered; After receiving the early warning, the security personnel immediately rushed to the scene to deal with the situation according to the preset emergency plan.

6. The method for monitoring and early warning of psychological and behavioral abnormalities in adolescents based on machine learning as claimed in claim 5, characterized in that: When a young user with a suspected abnormal psychological state is detected to enter a dangerous area, an early warning mechanism is immediately triggered, and the steps further include: The early warning information is conveyed to security personnel and relevant departments through sound and light alarms, SMS notifications, and APP push.

7. The method for monitoring and early warning of adolescent psychological and behavioral abnormalities based on machine learning as claimed in claim 5, characterized in that: After receiving the warning, the security personnel will immediately go to the scene to deal with it according to the preset emergency plan. The steps mentioned above also include: Trace the action trajectory of the young user before entering the dangerous area, and analyze the trajectory data to understand the user's activity patterns and possible risk factors, providing a reference for subsequent intervention measures.

Images