Health monitoring method, device, equipment and storage medium based on smart helmet

Abstract

The present invention relates to the field of health monitoring technology, and in particular to a health monitoring method, device, equipment and storage medium based on a smart helmet. The method comprises the following steps: obtaining real-time physiological parameters of a user based on a biological monitoring chip; mining personalized physiological state characteristics of the real-time physiological parameters of the user, thereby generating a personalized physiological state portrait of the user; analyzing the time series state characteristic changes of the personalized physiological state portrait of the user, and predicting the state trend of multiple time periods, thereby obtaining physiological state trend prediction data for multiple time periods; obtaining the user's eyeball visual image and smart helmet posture data based on a multimodal sensor; tracking the optical flow of eye feature points and locating the center of the pupil on the user's eyeball visual image, generating a representation of the evolution of the user's eyeball state; calculating the overall head movement frequency on the smart helmet posture data, thereby obtaining a head muscle relaxation evaluation value. The present invention realizes real-time and accurate health monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring, and in particular to a health monitoring method, device, equipment and storage medium based on a smart helmet. Background Art

[0002] With the rapid development of intelligent transportation systems, driving safety and driver health monitoring have gradually become the focus of research. As a new type of wearable device, smart helmets can monitor the driver's physiological state and environmental changes in real time during driving through integrated sensors and communication technologies. Smart helmets not only provide safety protection for drivers, but also can effectively collect physiological parameters such as heart rate, body temperature, and brain waves, providing reliable data support for health monitoring during driving.

[0003] In a real driving environment, the driver's health status will be affected by many factors, such as fatigue, anxiety and environmental noise. These factors can lead to decreased attention, slow reaction and even traffic accidents during long-term driving. However, traditional health monitoring methods often rely on manual inspections and periodic physical examinations, which make it difficult to achieve real-time monitoring and rapid response, and cannot effectively prevent potential health risks. The application of smart helmets provides a new idea for solving this problem. Therefore, there is an urgent need for a user's driving health monitoring method based on a smart helmet, which can integrate multiple physiological indicators and analyze the driver's health status in real time, thereby improving driving safety and driving experience. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a health monitoring method, device, equipment and storage medium based on a smart helmet to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a health monitoring method based on a smart helmet, wherein a biological monitoring sensor chip and a multimodal sensor are embedded in the smart helmet, and the method comprises the following steps:

[0006] Step S1: obtaining the user's real-time physiological parameters based on the biological monitoring chip; mining the personalized physiological state characteristics of the user's real-time physiological parameters, thereby generating a personalized physiological state portrait of the user;

[0007] Step S2: Analyze the time series state characteristic changes of the user's personalized physiological state portrait, and perform multi-time period state trend prediction, so as to obtain physiological state trend prediction data for multiple time periods;

[0008] Step S3: obtaining the user's eyeball visual image and the smart helmet posture data based on the multimodal sensor; performing optical flow tracking of eye feature points and pupil center location on the user's eyeball visual image to generate a representation of the user's eyeball state evolution;

[0009] Step S4: calculating the overall head movement frequency of the smart helmet posture data, and evaluating the head muscle relaxation, thereby obtaining a head muscle relaxation evaluation value;

[0010] Step S5: Performing deep correlation analysis on the user's eyeball state evolution representation and head muscle relaxation evaluation value, and performing long-term motion fatigue evolution to generate long-term fatigue situation prediction data;

[0011] Step S6: Conduct a comprehensive health status assessment on the physiological status trend prediction data and long-term fatigue status prediction data of multiple time periods, make health risk decisions, and build a health monitoring risk decision engine.

[0012] The present invention also provides a health monitoring device based on a smart helmet, comprising:

[0013] The state portrait module is used to obtain the user's real-time physiological parameters based on the biological monitoring chip; the user's real-time physiological parameters are mined for personalized physiological state characteristics, thereby generating a personalized physiological state portrait of the user;

[0014] The state trend prediction module is used to analyze the time series state characteristic changes of the user's personalized physiological state portrait and perform multi-time state trend prediction, thereby obtaining physiological state trend prediction data for multiple time periods;

[0015] The eye state evolution module is used to obtain the user's eye visual image and smart helmet posture data based on multimodal sensors; perform optical flow tracking of eye feature points and pupil center location on the user's eye visual image to generate a representation of the user's eye state evolution;

[0016] The relaxation evaluation module is used to calculate the overall head movement frequency of the smart helmet posture data and evaluate the relaxation of the head muscles, thereby obtaining the evaluation value of the relaxation of the head muscles;

[0017] The fatigue state prediction module is used to conduct in-depth correlation analysis on the user's eye state evolution representation and head muscle relaxation evaluation value, and to perform long-term motion fatigue evolution to generate long-term fatigue state prediction data;

[0018] The health risk decision module is used to conduct a comprehensive health status assessment on the physiological status trend prediction data of multiple time periods and the long-term fatigue status prediction data, make health risk decisions, and build a health monitoring risk decision engine.

[0019] The present invention also provides a computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the health monitoring method for the smart helmet described in any one of the above items are implemented.

[0020] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the health monitoring method for the smart helmet described in any one of the above items are implemented.

[0021] The beneficial effects of the present invention are as follows: by acquiring real-time physiological parameters through the biological monitoring chip, the system can monitor the user's physiological condition in real time; personalized physiological state feature mining helps to gain a deep understanding of the user's health status and provide a personalized basis for subsequent health monitoring; through time series state feature change analysis and multi-period state trend prediction, the system can predict the user's physiological state change trend; multi-period physiological state trend prediction data helps the user understand his or her own health changes and provide timely health management suggestions; acquiring eyeball visual images and smart helmet posture data helps monitor the user's eye state and head posture during driving; generating a representation of the user's eyeball state evolution can help evaluate the user's alertness and attention level and improve driving safety; Through the calculation of the overall movement frequency of the head and the evaluation of muscle relaxation, the system can evaluate the fatigue of the user's head muscles. The evaluation value of head muscle relaxation helps to remind the user to adjust the posture in time, reduce driving fatigue and improve driving comfort. Through the correlation analysis of the eye state evolution representation and muscle relaxation evaluation value, the system can predict the user's long-term fatigue state. The long-term fatigue state prediction data helps to remind the user to pay attention to driving safety and reduce the impact of driving fatigue on health. By combining the physiological state trend prediction data and long-term fatigue state prediction data of multiple time periods, the system can conduct a comprehensive assessment of the user's health status. Building a health monitoring risk decision-making engine helps the system provide personalized health risk assessment and decision-making suggestions to ensure the user's driving safety and health. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic diagram of the steps of a health monitoring method based on a smart helmet according to the present invention;

[0023] Figure 2 Detailed implementation flow chart of step S1;

[0024] Figure 3 Detailed implementation flow chart of step S2;

[0025] Figure 4 It is a schematic diagram of the detailed implementation steps of step S4. DETAILED DESCRIPTION

[0026] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0027] The present application example provides a health monitoring method, device, equipment and storage medium based on a smart helmet. The execution subjects of the health monitoring method, device, equipment and storage medium based on a smart helmet include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload equipment, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0028] See also Figures 1 to 4 The present invention provides a health monitoring method based on a smart helmet, wherein a biological monitoring sensor chip and a multimodal sensor are embedded in the smart helmet, and the health monitoring method based on the smart helmet comprises the following steps:

[0029] Step S1: obtaining the user's real-time physiological parameters based on the biological monitoring chip; mining the personalized physiological state characteristics of the user's real-time physiological parameters, thereby generating a personalized physiological state portrait of the user;

[0030] Step S2: Analyze the time series state characteristic changes of the user's personalized physiological state portrait, and perform multi-time period state trend prediction, so as to obtain physiological state trend prediction data for multiple time periods;

[0031] Step S3: obtaining the user's eyeball visual image and the smart helmet posture data based on the multimodal sensor; performing optical flow tracking of eye feature points and pupil center location on the user's eyeball visual image to generate a representation of the user's eyeball state evolution;

[0032] Step S4: calculating the overall head movement frequency of the smart helmet posture data, and evaluating the head muscle relaxation, thereby obtaining a head muscle relaxation evaluation value;

[0033] Step S5: Performing deep correlation analysis on the user's eyeball state evolution representation and head muscle relaxation evaluation value, and performing long-term motion fatigue evolution to generate long-term fatigue situation prediction data;

[0034] Step S6: Conduct a comprehensive health status assessment on the physiological status trend prediction data and long-term fatigue status prediction data of multiple time periods, make health risk decisions, and build a health monitoring risk decision engine.

[0035] By acquiring real-time physiological parameters through the bio-monitoring chip, the system can monitor the user's physiological condition in real time. Personalized physiological state feature mining helps to gain a deeper understanding of the user's health status and provide a personalized basis for subsequent health monitoring. Through time series state feature change analysis and multi-period state trend prediction, the system can predict the user's physiological state change trend. Multi-period physiological state trend prediction data helps users understand their own health changes and provide timely health management suggestions. Obtaining eye visual images and smart helmet posture data helps monitor the user's eye state and head posture during driving. Generating a representation of the user's eye state evolution can help evaluate the user's alertness and attention level and improve driving safety. The overall head movement frequency Through rate calculation and muscle relaxation assessment, the system can assess the user's head muscle fatigue. The head muscle relaxation assessment value helps to remind the user to adjust the posture in time, reduce driving fatigue, and improve driving comfort. Through the correlation analysis of the eye state evolution representation and muscle relaxation assessment value, the system can predict the user's long-term fatigue state. The long-term fatigue state prediction data helps to remind the user to pay attention to driving safety and reduce the impact of driving fatigue on health. By combining the physiological state trend prediction data and long-term fatigue state prediction data of multiple time periods, the system can conduct a comprehensive assessment of the user's health status. Building a health monitoring risk decision-making engine helps the system provide personalized health risk assessment and decision-making suggestions to ensure the user's driving safety and health.

[0036] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a health monitoring method based on a smart helmet of the present invention. In this example, the steps of the health monitoring method based on a smart helmet include:

[0037] Step S1: obtaining the user's real-time physiological parameters based on the biological monitoring chip; mining the personalized physiological state characteristics of the user's real-time physiological parameters, thereby generating a personalized physiological state portrait of the user;

[0038] In this embodiment, a biological monitoring chip is used to collect the user's physiological signals in real time, and a signal processing technology (such as a low-pass filter and a moving average filter) is used to denoise the collected physiological data to ensure the accuracy of the data. Different types of physiological parameters are standardized to eliminate dimensional differences for subsequent analysis. Key indicators (such as heart rate variability, body temperature fluctuations, sweat component concentrations, etc.) are selected from the real-time physiological parameters as the basis for mining features. Statistical methods (such as mean, standard deviation, correlation coefficient, etc.) are used to analyze the selected features to identify the user's physiological state change pattern. A time series analysis method (such as a sliding window technology) is used to extract dynamic features to capture the instantaneous changes in the user's physiological state. Based on the mined physiological features, a personalized physiological state portrait model is constructed, which includes the user's physiological state description, health indicator range and potential risk assessment. The generated physiological state portrait is visualized, and charts, dashboards, etc. are used to make it easy for users to understand their own physiological states. The user's personalized physiological state portrait is stored in a database and updated regularly to reflect the user's physiological state changes and health trends.

[0039] Step S2: Analyze the time series state characteristic changes of the user's personalized physiological state portrait, and perform multi-time period state trend prediction, so as to obtain physiological state trend prediction data for multiple time periods;

[0040] In this embodiment, a time series analysis is performed on the physiological state portrait data of the user to extract key features, such as mean, standard deviation, maximum value, minimum value, etc., to identify the changing trend of the physiological state, and a sliding window technique is applied to analyze the physiological state characteristics within a continuous time period to identify significant state changes (such as sudden heart rate fluctuations). A suitable time series prediction model is selected according to the data characteristics, such as ARIMA (autoregressive integrated moving average model), LSTM (long short-term memory network) or Prophet model. The prediction model is trained using the user's historical physiological state data to learn the time series characteristics and changing laws of the physiological state. Based on the trained model, the physiological state of multiple time periods in the future is predicted to generate future physiological state trend data. The generated physiological state trend prediction data is visualized, and a line chart or a bar chart is used to display the prediction trend of each time period for easy understanding.

[0041] Step S3: obtaining the user's eyeball visual image and the smart helmet posture data based on the multimodal sensor; performing optical flow tracking of eye feature points and pupil center location on the user's eyeball visual image to generate a representation of the user's eyeball state evolution;

[0042] In this embodiment, a computer vision algorithm (such as Shi-Tomasi corner detection) is used to identify eye feature points, including key points such as pupils and eye corners. The Lucas-Kanade optical flow method is applied to track the extracted eye feature points to generate an eye optical flow movement trajectory. The data of the optical flow movement trajectory is recorded as a time series for subsequent analysis of eye state changes. The collected eye visual images are subjected to illumination correction and noise removal to improve the accuracy of pupil recognition. The pupil center point is located using image processing technology (such as Hough transform or circle detection), its coordinates are calculated, the pupil center point is marked on the image, and its coordinate information is saved as basic data for subsequent analysis. The eye optical flow movement trajectory and the pupil center point position information are integrated to form comprehensive feature data of the user's eye state. By analyzing the changes in the optical flow trajectory and the pupil center point, dynamic features (such as eye movement range, pupil size changes, etc.) are extracted. According to the analysis results, a representation of the user's eye state evolution is generated, including physiological indicators, dynamic changes and potential health status information.

[0043] Step S4: calculating the overall head movement frequency of the smart helmet posture data, and evaluating the head muscle relaxation, thereby obtaining a head muscle relaxation evaluation value;

[0044] In this embodiment, a threshold judgment method is applied to identify the time period of head movement, and a suitable threshold is set to distinguish between static and moving states (for example, an angular velocity exceeding a certain value is considered to be in a moving state). The number of movements within the effective movement period is counted and divided by the observation time period (such as per minute). The overall movement frequency of the head is calculated, and the calculated head movement frequency is recorded as time series data for subsequent analysis. The key features that affect the relaxation of head muscles are determined, including head movement frequency, movement amplitude, static duration, etc. A muscle relaxation evaluation model is designed based on physiological and kinematic principles. The model can be constructed using linear regression or machine learning algorithms (such as random forests). The model is trained using historical data, and the performance of the model is evaluated and parameters are optimized to improve the accuracy of the prediction. The real-time head movement frequency and other relevant features are input into the evaluation model to calculate the current head muscle relaxation evaluation value.

[0045] Step S5: Performing deep correlation analysis on the user's eyeball state evolution representation and head muscle relaxation evaluation value, and performing long-term motion fatigue evolution to generate long-term fatigue situation prediction data;

[0046] In this embodiment, relevant data are extracted from the user's eye state evolution representation (such as eye movement trajectory, pupil size change) and head muscle relaxation evaluation value to form a joint data set, and key features are determined, including the eye state change amplitude, muscle relaxation value, movement frequency, etc. As the basis for analysis, statistical methods (such as Pearson correlation coefficient, Spearman rank correlation) or machine learning methods (such as multivariate linear regression, random forest) are used to perform deep correlation analysis to identify the relationship between eye state and muscle relaxation, collect the user's long-term physiological state data, eye state data and muscle relaxation evaluation value to form a long-term monitoring data set, divide the data into time periods (such as daily, weekly) to provide a basis for subsequent analysis, select a suitable time series analysis model (such as ARIMA, LSTM) to analyze the evolution trend of sports fatigue, use historical data to train the evolution analysis model, evaluate the performance of the model and optimize the parameters to improve the prediction accuracy.

[0047] Step S6: Conduct a comprehensive health status assessment on the physiological status trend prediction data and long-term fatigue status prediction data of multiple time periods, make health risk decisions, and build a health monitoring risk decision engine.

[0048] In this embodiment, physiological status trend prediction data (such as heart rate, body temperature, etc.) and long-term fatigue situation prediction data for multiple time periods are collected to form a comprehensive health data set, and a comprehensive health status assessment model is designed. Considering multiple physiological indicators and fatigue status, the weight of each indicator is determined (based on medical expert opinions or historical data analysis), and a weighted average or multivariate statistical method is used to calculate a comprehensive health status assessment value to reflect the overall health status of the user. The calculated comprehensive health status assessment value is recorded in a database to facilitate subsequent analysis and user feedback. A health risk threshold is preset according to the comprehensive health status assessment value, and different health risk levels (such as low, medium, and High risk), use risk analysis models (such as logistic regression, decision tree) to analyze the comprehensive health status assessment value, generate user health risk data, use the set standards to judge the user's health risk level, and record the corresponding risk information. Based on the health risk data, design health intervention decision rules, clarify the intervention measures at different risk levels (such as diet adjustment, increased exercise, medical advice, etc.), develop a health monitoring risk decision engine, integrate user health data and health risk analysis results, automatically generate personalized health intervention suggestions, and build a real-time feedback mechanism to ensure that users can receive corresponding risk assessments and intervention suggestions in a timely manner when their health status changes.

[0049] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0050] Step S11: Real-time monitoring of the user's physiological signals based on the biological monitoring chip, and digital processing to generate real-time physiological parameters of the user, wherein the real-time physiological parameters of the user include heart rate parameters, body temperature parameters and sweat composition data;

[0051] Step S12: Calculating the real-time heart rate fluctuation to generate the user's heart rate fluctuation data;

[0052] Step S13: performing statistics on the variation range of the body temperature parameters to generate a body temperature variation range curve;

[0053] Step S14: performing electrolyte balance analysis on the sweat component data to generate electrolyte concentration data;

[0054] Step S15: Perform personalized physiological state feature mining on the user's heart rate fluctuation data, body temperature variation curve and electrolyte concentration data, so as to generate a personalized physiological state portrait of the user.

[0055] In this embodiment, a bio-monitoring chip is used to collect the user's physiological signals in real time, including heart rate, body temperature and sweat composition, and the collected analog signals are digitized and processed, and an analog-to-digital converter (ADC) is used to convert the analog signals into digital signals, and heart rate data is extracted from the real-time physiological parameters to ensure the time series integrity of the data, and a statistical method (such as standard deviation or autocorrelation analysis) is applied to calculate the fluctuation of the heart rate, identify the frequency and amplitude of the heart rate change, record the time series data of the heart rate fluctuation, and analyze its association with the user's activity status to provide a basis for subsequent analysis, extract relevant electrolyte information (such as sodium, potassium and chloride) from the sweat composition data to ensure the integrity of the data, and use biochemical analysis methods to evaluate the electrolyte concentration to determine whether it is within the normal range and identify potential electrolyte imbalances, integrate the user's heart rate fluctuation data, body temperature change amplitude curve and electrolyte concentration data to form a comprehensive physiological data set, and use machine learning algorithms (such as cluster analysis and principal component analysis) to perform personalized feature mining on the integrated data to identify the user's physiological state characteristics, and generate a personalized physiological state portrait of the user based on the analysis results, including the normal range, change trend and potential health risks of physiological indicators, to provide a comprehensive physiological state assessment for the user.

[0056] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0057] Step S21: extracting multiple timestamps based on the user's real-time physiological parameters;

[0058] Step S22: Calculate the exercise time of multiple timestamps to obtain the user's exercise duration;

[0059] Step S23: performing multi-time window division processing according to the user's exercise duration, thereby obtaining a multi-stage exercise time window;

[0060] Step S24: performing temporal state characteristic change analysis on the user's personalized physiological state portrait based on the multi-stage exercise time window to generate user temporal state characteristic change data;

[0061] Step S25: performing deep convolution learning processing on the user's temporal state feature change data to generate a temporal state change law;

[0062] Step S26: Utilizing the time series state change rule to perform multi-time period state trend prediction on the user's real-time physiological parameters, thereby obtaining physiological state trend prediction data for multiple time periods.

[0063] In this embodiment, when monitoring the real-time physiological parameters of the user, the timestamps of data such as heart rate, body temperature, and electrolytes are recorded to ensure the accuracy of the time information, and the extracted timestamps are organized into a unified format (year-month-day-hour-minute-second) for subsequent analysis and processing. The user's motion state is analyzed using sensor data (such as accelerometers and gyroscopes) to identify when the user is in motion. Based on the extracted timestamps, the duration of the user in motion is calculated, the start and end time of the motion is identified, the user's motion duration is recorded, and the duration is associated with the physiological parameter data to form a motion data set. According to the user's motion duration, it is divided into multiple time windows (such as every 5 minutes, every 10 minutes) to ensure that the data in each window is representative. The motion data is grouped by time window to form a multi-stage motion time window for subsequent analysis. Each time window is labeled to identify its corresponding motion stage or activity type, and physiological state features (such as average heart rate, body temperature change amplitude, electrolyte) are extracted from each multi-stage motion time window. The method comprises the following steps: a) using the time series state feature data set to form a time series state feature data set, applying time series analysis methods (such as sliding average method or time series decomposition) to study the changing trends of physiological state features in different time windows, and constructing a deep convolutional neural network (CNN) model to learn the spatial and temporal patterns of the user's time series state feature change data. The time series state feature change data is standardized and formatted to ensure that it is suitable for input into the convolutional network. The model is trained using historical data and the model parameters are optimized to improve the learning ability of the time series state change law. Through the output of the model, the user's time series state change law is extracted, and the main trends and patterns of feature changes are identified. Based on the generated time series state change law, a multi-period state trend prediction model is constructed, and a recursive neural network (RNN) or a long short-term memory network (LSTM) is used for prediction. The user's real-time physiological parameters are input into the prediction model to generate physiological state trend prediction data for multiple time periods in the future. The prediction results are sorted and stored in a database to provide a basis for subsequent user health management and feedback.

[0064] In this embodiment, the specific steps of step S3 are:

[0065] Step S31: Acquire the user's eye visual image and smart helmet posture data based on the multimodal sensor;

[0066] Step S32: performing optical flow tracking of eye feature points on the user's eyeball visual image to generate an eyeball optical flow movement trajectory;

[0067] Step S33: calculating the eye movement range of the eye optical flow movement trajectory to obtain the movement range of the optical flow trajectory;

[0068] Step S34: performing dynamic trajectory movement speed analysis on the eyeball optical flow movement trajectory to obtain the eyeball dynamic trajectory speed;

[0069] Step S35: locating the pupil center point of the user's eyeball visual image and marking the pupil center point;

[0070] Step S36: Analyze pupil size change characteristics of the pupil center point to generate pupil size change characteristics;

[0071] Step S37: Dynamically evolve the eye state of the optical flow trajectory moving range, the eye dynamic trajectory speed and the pupil size change characteristics to generate a representation of the user's eye state evolution.

[0072] In this embodiment, a multimodal sensor embedded in the smart helmet, including a camera and an IMU (inertial measurement unit), is used to ensure that it can capture the user's eye visual image and head posture data, start the sensor, collect the user's eye visual image and helmet posture data (including position and angle information) in real time, use a computer vision algorithm (such as Harris corner detection or SIFT) to extract eye feature points, select a suitable optical flow tracking algorithm, such as the Lucas-Kanade method, perform feature point optical flow tracking, record the position of the tracked feature points in continuous frames, generate an eye optical flow movement trajectory, calculate the minimum and maximum coordinates of the eye movement based on the points in the optical flow trajectory to determine the movement range, and generate a movement range indicator of the optical flow trajectory, such as a range surface. The method comprises the following steps: calculating the product or maximum moving distance, selecting a suitable speed calculation method, such as calculating the speed by time difference and position difference, calculating the moving speed of each segment in the optical flow trajectory, recording the speed change, selecting a suitable pupil detection algorithm, such as Hough transform or deep learning model, locating the center point of the pupil in the eye visual image, recording its coordinates, calculating the diameter or area of ​​the pupil according to the coordinates of the center point of the pupil and the surrounding edge points, recording the change in pupil size, generating pupil size change feature data, selecting a suitable dynamic state evolution model, such as Markov model or deep learning model, analyzing the user's eye state evolution according to the moving range of the optical flow trajectory, the dynamic trajectory speed and the pupil size change characteristics, generating a dynamic representation of the user's eye state, including feature descriptions in different states.

[0073] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S4. In this embodiment, the detailed implementation steps of step S4 include:

[0074] Step S41: Calculate the overall head movement frequency of the smart helmet posture data to generate the head movement frequency;

[0075] Step S42: performing time series frequency distribution analysis on the head movement frequency to generate movement frequency distribution data;

[0076] Step S43: quantifying the left and right swing amplitudes of the smart helmet posture data to generate left and right swing amplitude values;

[0077] Step S44: performing a swing symmetry analysis on the left and right swing amplitude values ​​to obtain swing amplitude symmetry data;

[0078] Step S45: Evaluate the head muscle relaxation based on the motion frequency distribution data and the swing amplitude symmetry data, so as to obtain a head muscle relaxation evaluation value.

[0079] In this embodiment, by setting a threshold, the time period of head movement is identified, valid movement data is screened out, the number of movements within the movement period is counted, the number of movements is divided by the time period, the overall head movement frequency is calculated, and recorded as time series data, the calculated head movement frequency is sorted into frequency data within a time period (such as every minute, every 5 minutes), and a statistical method (such as a histogram or density estimation) is used to analyze the distribution of the movement frequency, generate movement frequency distribution data, and display the frequency distribution through a chart to help identify movement patterns and behavioral trends. The left and right swing angle data of the head is extracted from the smart helmet posture data, the maximum and minimum angles of each left and right swing are calculated, and the swing amplitude value is determined , forming left and right swing amplitude data, calculating the symmetry of the swing amplitude by comparing the values ​​of the left and right swing amplitudes, and using absolute difference, ratio or other appropriate methods to organize the symmetry analysis results and generate swing amplitude symmetry data for subsequent comprehensive evaluation, graphically displaying the swing symmetry data to facilitate identification of whether there are obvious deviations, integrating the movement frequency distribution data and the swing amplitude symmetry data to form a comprehensive evaluation data set, and establishing a head muscle relaxation evaluation model, based on the movement frequency and swing symmetry analysis results, using weighted calculation or machine learning methods to evaluate muscle relaxation, and calculating the head muscle relaxation evaluation value, which is recorded and stored for subsequent analysis and reference.

[0080] In this embodiment, the specific steps of step S5 are:

[0081] Step S51: performing deep correlation analysis on the user's eye state evolution representation and head muscle relaxation evaluation value to obtain eye state-muscle relaxation correlation data;

[0082] Step S52: predicting the user's short-term fatigue degree based on the eyeball state-muscle relaxation correlation data, thereby generating user's short-term fatigue degree prediction data;

[0083] Step S53: Perform long-term sports fatigue evolution on the user's short-term fatigue prediction data to generate long-term fatigue situation prediction data.

[0084] In this embodiment, the user's eye state evolution representation data (such as optical flow trajectory, pupil size change, etc.) and head muscle relaxation evaluation value are integrated to form a joint data set, and statistical analysis methods (such as Pearson correlation coefficient, Spearman rank correlation) or machine learning algorithms (such as multivariate linear regression) are used to perform deep correlation analysis on the integrated data, calculate and record the correlation between the eye state and muscle relaxation, generate eye state-muscle relaxation correlation data, reveal the mutual influence relationship between the two, and construct a short-term fatigue prediction model based on the eye state-muscle relaxation correlation data. Machine learning algorithms (such as decision trees, random forests, or support vector machines) can be used to select important data from the correlation data. Features (such as the amplitude of eye state changes, muscle relaxation, etc.), and perform necessary data preprocessing (such as standardization), use historical data to train the prediction model, and evaluate the accuracy and stability of the model through cross-validation methods, input real-time user data into the model, generate user short-term fatigue prediction data, and record the results, collect users' short-term fatigue prediction data and historical motion data to form a long-term monitoring data set, use time series analysis methods (such as ARIMA, LSTM) to perform long-term evolution analysis on short-term fatigue, identify the trend of fatigue changes over time, visualize the long-term fatigue situation prediction results, and display the fatigue change curve to help users intuitively understand their fatigue status.

[0085] In this embodiment, the specific steps of step S6 are:

[0086] Step S61: Performing a comprehensive health status assessment on the physiological status trend prediction data and the long-term fatigue status prediction data of multiple time periods to generate a comprehensive health status assessment value of the user;

[0087] Step S62: performing health risk analysis on the user's comprehensive health status assessment value based on the preset health status assessment value to generate user health risk data;

[0088] Step S63: Make health risk decisions on the user's health risk data and build a health monitoring risk decision engine.

[0089] In this embodiment, physiological status trend prediction data and long-term fatigue situation prediction data for multiple time periods are collected to form a comprehensive data set, and a comprehensive health status assessment model is designed. Key physiological indicators (such as heart rate, body temperature, exercise duration, etc.) and fatigue are considered, and weighted average or multivariate analysis methods are used to calculate the integrated data according to the model to generate a comprehensive health status assessment value for the user, reflecting the user's overall health status, preset risk thresholds and standards for the health status assessment value, and define different health risk levels (such as low, medium, and high risks). Based on the user's comprehensive health status assessment value, a risk analysis model (such as logistic regression, decision tree) is applied to perform health risk analysis, identify the user's health risk level, generate user health risk data, and mark the corresponding risk level. Based on the health risk data, decision rules are designed to determine health intervention measures at different risk levels (such as recommended exercise, dietary adjustments, medical advice, etc.), and a health monitoring risk decision engine is developed to automatically generate personalized health intervention recommendations based on user health data and health risk analysis results.

[0090] In this embodiment, a health monitoring device based on a smart helmet is also provided, including:

[0091] The state portrait module is used to obtain the user's real-time physiological parameters based on the biological monitoring chip; the user's real-time physiological parameters are mined for personalized physiological state characteristics, thereby generating a personalized physiological state portrait of the user;

[0092] The state trend prediction module is used to analyze the time series state characteristic changes of the user's personalized physiological state portrait and perform multi-time state trend prediction, thereby obtaining physiological state trend prediction data for multiple time periods;

[0093] The eye state evolution module is used to obtain the user's eye visual image and smart helmet posture data based on multimodal sensors; perform optical flow tracking of eye feature points and pupil center location on the user's eye visual image to generate a representation of the user's eye state evolution;

[0094] The relaxation evaluation module is used to calculate the overall head movement frequency of the smart helmet posture data and evaluate the relaxation of the head muscles, thereby obtaining the evaluation value of the relaxation of the head muscles;

[0095] The fatigue state prediction module is used to conduct in-depth correlation analysis on the user's eye state evolution representation and head muscle relaxation evaluation value, and to perform long-term motion fatigue evolution to generate long-term fatigue state prediction data;

[0096] The health risk decision module is used to conduct a comprehensive health status assessment on the physiological status trend prediction data of multiple time periods and the long-term fatigue status prediction data, make health risk decisions, and build a health monitoring risk decision engine.

[0097] The present invention also provides a computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the health monitoring method for the smart helmet described in any one of the above items are implemented.

[0098] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the health monitoring method for the smart helmet described in any one of the above items are implemented.

[0099] Those skilled in the art clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0100] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0101] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0102] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A health monitoring method based on a smart helmet, characterized in that: The smart helmet is embedded with a bio-monitoring sensor chip and a multi-modal sensor, which includes the following steps: Step S1: obtaining the user's real-time physiological parameters based on the biological monitoring chip; mining the personalized physiological state characteristics of the user's real-time physiological parameters, thereby generating a personalized physiological state portrait of the user; Step S2: Analyze the time series state characteristic changes of the user's personalized physiological state portrait, and perform multi-time period state trend prediction, so as to obtain physiological state trend prediction data for multiple time periods; Step S3: obtaining the user's eyeball visual image and the smart helmet posture data based on the multimodal sensor; performing optical flow tracking of eye feature points and pupil center location on the user's eyeball visual image to generate a representation of the user's eyeball state evolution; Step S4: calculating the overall head movement frequency of the smart helmet posture data, and evaluating the head muscle relaxation, thereby obtaining a head muscle relaxation evaluation value; Step S5: Perform deep correlation analysis on the user's eyeball state evolution representation and head muscle relaxation evaluation value, and perform long-term motion fatigue evolution to generate long-term fatigue situation prediction data; the specific steps of step S5 are: Step S51: performing deep correlation analysis on the eye state evolution representation and the head muscle relaxation evaluation value of the user to obtain eye state-muscle relaxation correlation data; the deep correlation analysis specifically comprises: performing deep correlation analysis on the integrated data using a statistical analysis method or a machine learning algorithm, calculating and recording the correlation between the eye state and the muscle relaxation, generating eye state-muscle relaxation correlation data, and revealing the mutual influence relationship between the two; Step S52: predicting the user's short-term fatigue degree based on the eyeball state-muscle relaxation correlation data, thereby generating user's short-term fatigue degree prediction data; Step S53: performing long-term sports fatigue evolution on the user's short-term fatigue prediction data to generate long-term fatigue situation prediction data; the long-term sports fatigue evolution specifically comprises: collecting the user's short-term fatigue prediction data and historical sports data to form a long-term monitoring data set, using a time series analysis method to perform a long-term evolution analysis on the short-term fatigue, identifying the trend of fatigue changes over time, visualizing the long-term fatigue situation prediction results, and displaying the fatigue change curve; Step S6: Conduct a comprehensive health status assessment on the physiological status trend prediction data and long-term fatigue status prediction data of multiple time periods, make health risk decisions, and build a health monitoring risk decision engine.

2. The health monitoring method based on the smart helmet according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: Real-time monitoring of the user's physiological signals based on the biological monitoring chip, and digital processing to generate real-time physiological parameters of the user, wherein the real-time physiological parameters of the user include heart rate parameters, body temperature parameters and sweat composition data; Step S12: Calculating the real-time heart rate fluctuation to generate the user's heart rate fluctuation data; Step S13: performing statistics on the variation range of the body temperature parameters to generate a body temperature variation range curve; Step S14: performing electrolyte balance analysis on the sweat component data to generate electrolyte concentration data; Step S15: Perform personalized physiological state feature mining on the user's heart rate fluctuation data, body temperature variation curve and electrolyte concentration data, so as to generate a personalized physiological state portrait of the user.

3. The health monitoring method based on the smart helmet according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: extracting multiple timestamps based on the user's real-time physiological parameters; Step S22: Calculate the exercise time of multiple timestamps to obtain the user's exercise duration; Step S23: performing multi-time window division processing according to the user's exercise duration, thereby obtaining a multi-stage exercise time window; Step S24: performing temporal state characteristic change analysis on the user's personalized physiological state portrait based on the multi-stage exercise time window to generate user temporal state characteristic change data; Step S25: performing deep convolution learning processing on the user's temporal state feature change data to generate a temporal state change law; Step S26: Utilizing the time series state change rule to perform multi-time period state trend prediction on the user's real-time physiological parameters, thereby obtaining physiological state trend prediction data for multiple time periods.

4. The health monitoring method based on the smart helmet according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: Acquire the user's eye visual image and smart helmet posture data based on the multimodal sensor; Step S32: performing optical flow tracking of eye feature points on the user's eyeball visual image to generate an eyeball optical flow movement trajectory; Step S33: calculating the eye movement range of the eye optical flow movement trajectory to obtain the movement range of the optical flow trajectory; Step S34: performing dynamic trajectory movement speed analysis on the eyeball optical flow movement trajectory to obtain the eyeball dynamic trajectory speed; Step S35: locating the pupil center point of the user's eyeball visual image and marking the pupil center point; Step S36: Analyze pupil size change characteristics of the pupil center point to generate pupil size change characteristics; Step S37: Dynamically evolve the eye state of the optical flow trajectory moving range, the eye dynamic trajectory speed and the pupil size change characteristics to generate a representation of the user's eye state evolution.

5. The health monitoring method based on the smart helmet according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: Calculate the overall head movement frequency of the smart helmet posture data to generate the head movement frequency; Step S42: performing time series frequency distribution analysis on the head movement frequency to generate movement frequency distribution data; Step S43: quantifying the left and right swing amplitudes of the smart helmet posture data to generate left and right swing amplitude values; Step S44: performing a swing symmetry analysis on the left and right swing amplitude values ​​to obtain swing amplitude symmetry data; Step S45: Evaluate the head muscle relaxation based on the motion frequency distribution data and the swing amplitude symmetry data, so as to obtain a head muscle relaxation evaluation value.

6. The health monitoring method based on the smart helmet according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: Performing a comprehensive health status assessment on the physiological status trend prediction data and the long-term fatigue status prediction data of multiple time periods to generate a comprehensive health status assessment value of the user; Step S62: performing health risk analysis on the user's comprehensive health status assessment value based on the preset health status assessment value to generate user health risk data; Step S63: Make health risk decisions on the user's health risk data and build a health monitoring risk decision engine.

7. A health monitoring device based on a smart helmet, characterized in that: The method for performing the health monitoring method based on the smart helmet as claimed in claim 1 comprises: The state portrait module is used to obtain the user's real-time physiological parameters based on the biological monitoring chip; the user's real-time physiological parameters are mined for personalized physiological state characteristics, thereby generating a personalized physiological state portrait of the user; The state trend prediction module is used to analyze the time series state characteristic changes of the user's personalized physiological state portrait and perform multi-time state trend prediction, thereby obtaining physiological state trend prediction data for multiple time periods; The eye state evolution module is used to obtain the user's eye visual image and smart helmet posture data based on multimodal sensors; perform optical flow tracking of eye feature points and pupil center location on the user's eye visual image to generate a representation of the user's eye state evolution; The relaxation evaluation module is used to calculate the overall head movement frequency of the smart helmet posture data and evaluate the relaxation of the head muscles, thereby obtaining the evaluation value of the relaxation of the head muscles; The fatigue state prediction module is used to conduct in-depth correlation analysis on the user's eye state evolution representation and head muscle relaxation evaluation value, and to perform long-term motion fatigue evolution to generate long-term fatigue state prediction data; The health risk decision module is used to conduct a comprehensive health status assessment on the physiological status trend prediction data of multiple time periods and the long-term fatigue status prediction data, make health risk decisions, and build a health monitoring risk decision engine.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the smart helmet-based health monitoring method described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the smart helmet-based health monitoring method described in any one of claims 1 to 6 are implemented.

Images