Miner fatigue state monitoring system based on multi-source data fusion

Abstract

The invention discloses a miner fatigue state monitoring system based on multi-source data fusion. The miner fatigue state monitoring system comprises a data acquisition module, a ground scheduling platform and a background management platform, the data acquisition module acquires multi-source data; the ground scheduling platform receives and stores the multi-source data, analyzes the multi-source data to obtain fatigue state feature information, determines the fatigue degree based on the multiple pieces of fatigue state feature information in combination with a fatigue degree monitoring model, and triggers early warning when the fatigue degree exceeds a set threshold value; and the background management platform receives the early warning information and the fatigue degree, and performs operation intervention processing based on the early warning information and the fatigue degree. Physiological data and video data are acquired through the data acquisition module, data transmission is performed through the 10-gigabit looped network, the reliability and safety of data transmission are improved, the 10-gigabit looped network transmits the physiological data and the video data to the ground scheduling platform for processing, and the data transmission efficiency is improved. Based on the physiological data and the video data, the fatigue state evaluation accuracy of the ground scheduling platform is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health data processing, and in particular to a monitoring system for miners' fatigue state based on multi-source data fusion. Background Art

[0002] In the mining operation environment, due to factors such as complex geological conditions, high labor intensity, and harsh environment, miners are prone to fatigue during operation; this fatigue state not only affects work efficiency but may also lead to operational errors, resulting in equipment injury accidents, posing a great threat to the personal safety of miners; currently, fatigue monitoring is mainly applied in the field of fatigue driving, and it is determined whether there is a fatigue driving problem through image monitoring. However, fatigue driving is on the ground, with good data transmission and no problem of transmission interruption; while the underground coal mine environment is complex, there are various interference factors, which may cause the collected data to be lost, affecting the fatigue state assessment result. Moreover, fatigue driving collects a single data source, and this single data source may introduce errors in the data processing and analysis process, affecting the accuracy of the fatigue state; therefore, it is necessary to design a fatigue state monitoring system for the underground coal mine environment. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies in the above-mentioned prior art, and provide a monitoring system for miners' fatigue state based on multi-source data fusion. The physiological data and video data are collected through a data collection module, and the data is transmitted through a 10 Gigabit Ring Network, improving the reliability and security of data transmission. The 10 Gigabit Ring Network transmits the physiological data and video data to the ground dispatching platform for processing, and based on the physiological data and video data, the accuracy of the fatigue state assessment by the ground dispatching platform is improved.

[0004] To achieve the above purpose, the technical solution adopted by the present invention is: a monitoring system for miners' fatigue state based on multi-source data fusion, including a data collection module, a ground dispatching platform, and a background management platform; the data collection module is used to collect multi-source data; the ground dispatching platform receives and stores the multi-source data, analyzes the multi-source data respectively to obtain corresponding fatigue state characteristic information, determines the fatigue degree based on multiple pieces of the fatigue state characteristic information in combination with a fatigue degree monitoring model, and triggers an early warning reminder when the fatigue degree exceeds a set threshold; the background management platform is used to receive the early warning information and the fatigue degree, and perform operation intervention processing based on the early warning information and the fatigue degree.

[0005] Preferably, the multi-source data includes physiological data and video data, the fatigue state includes physiological fatigue and visual fatigue, the physiological fatigue is obtained based on the physiological data, and the visual fatigue is obtained based on the video data.

[0006] Preferably, the data acquisition module includes a smart wearable device and a video monitoring device. The smart wearable device is used to continuously collect physiological data, including body temperature, heart rate, and blood pressure. The video monitoring device is used to continuously collect video data, including operation behavior and environmental data.

[0007] Preferably, the fatigue status monitoring system further includes a user terminal, which is used to receive warning information and operation intervention information.

[0008] Preferably, the ground dispatching platform includes a data storage module, a data preprocessing module, and a data statistical analysis module. The data storage module is used to receive and store the physiological data and video data transmitted by the data transmission module. The data preprocessing module is used to clean and organize the physiological data and video data received by the data storage module to obtain feature information related to the fatigue status. The data statistical analysis module analyzes and processes the feature information to extract key features and determine the fatigue degree.

[0009] Preferably, the data preprocessing module includes a physiological data processing unit and a video data processing unit. The physiological data processing unit uses wavelet transform to process the physiological data to obtain the feature information of the physiological data. The video data processing unit uses a convolutional neural network to process the video data to obtain the feature information of the video data.

[0010] Preferably, the data statistical analysis module includes a first data processing unit and a second data processing unit. The first data processing unit uses an LSTM neural network structure to determine the physiological fatigue degree based on the feature information of the physiological data. The second data processing unit uses the YOLO algorithm for identification and analysis to determine the visual fatigue degree based on the feature information of the video data.

[0011] Preferably, the fatigue degree monitoring model is determined by physiological fatigue and visual fatigue, and is specifically expressed as:

[0012] Fatigue degree = 0.7 * physiological fatigue + 0.3 * visual fatigue

[0013] Preferably, the operation intervention measures include mild fatigue management strategies, moderate fatigue intervention measures, severe fatigue emergency response measures, and extremely severe fatigue comprehensive adjustment measures.

[0014] Preferably, the background management platform includes a data storage module and a data display module. The data storage module is used to store the received warning information and fatigue degree data. The data display module is used to display the fatigue degree. At least one user is registered on the background management platform, and the background management platform is used to manage the fatigue status information of each user.

[0015] The present invention has the following advantages compared with the prior art:

[0016] 1. The present invention collects physiological data and video data through a data acquisition module, and transmits the data through a 10 Gigabit Ring Network, improving the reliability and security of data transmission. The 10 Gigabit Ring Network transmits the physiological data and video data to the ground dispatching platform for processing, and improves the accuracy of the ground dispatching platform's assessment of the fatigue state based on the physiological data and video data.

[0017] 2. The data acquisition module of the present invention includes intelligent wearable devices and video monitoring devices. The intelligent wearable devices collect multiple physiological data, and the video monitoring devices collect operation behaviors and environmental data, improving the accuracy of the data through multi-data acquisition.

[0018] 3. The present invention guides the miners' operation conditions through operation intervention measures, combines the miners' fatigue state and work requirements, automatically adjusts the work plan, reduces the miners' work load, and improves the personal safety of the miners.

[0019] The following further describes the present invention in detail through the drawings and embodiments. Description of the Drawings

[0020] Figure 1 is a schematic diagram of the modules of the present invention;

[0021] Figure 2 is a schematic diagram of the data processing flow of the present invention;

[0022] Figure 3 is a schematic diagram of the data acquisition and processing process of the present invention. Detailed Embodiments

[0023] As Figures 1 to 3 shown, the present invention discloses a miners' fatigue state monitoring system based on multi-source data fusion, including a data acquisition module, a ground dispatching platform, and a background management platform; the data acquisition module is used to collect multi-source data; the ground dispatching platform receives and stores the multi-source data, analyzes the multi-source data respectively to obtain corresponding fatigue state characteristic information, determines the fatigue degree based on multiple fatigue state characteristic information in combination with a fatigue degree monitoring model, and triggers a warning reminder when the fatigue degree exceeds a set threshold; the background management platform receives the warning information and the fatigue degree, and performs operation intervention processing based on the warning information and the fatigue degree.

[0024] The multi-source data includes physiological data and video data, the fatigue state includes physiological fatigue and visual fatigue, the physiological fatigue is obtained based on the physiological data, and the visual fatigue is obtained based on the video data.

[0025] During the operation of the system, the data acquisition module continuously collects the physiological data and video data of miners during their work. The physiological data and video data are respectively encapsulated and encrypted through the data transmission module to ensure that the physiological data and video data of miners are quickly and securely transmitted to the ground dispatching platform. After receiving the physiological data and video data of miners, the ground dispatching platform first stores the data, and then processes the physiological data and video data respectively, extracts the feature information related to the fatigue state from the physiological data and the feature information related to the fatigue state from the video data, determines the physiological fatigue and video fatigue respectively based on the physiological data feature information and video data feature information, combines the fatigue degree monitoring model to determine the fatigue degree of miners. The ground dispatching platform limits the fatigue degree of miners. If the current fatigue degree of miners exceeds the set threshold, the ground dispatching platform triggers an alarm and sends the alarm information to the background management platform. The background management platform adjusts the next work task of this miner according to the alarm information and the fatigue degree of the miner, reduces the operation risk of this miner, and through the combination of physiological data and video data, improves the accuracy of miners' fatigue data and the accuracy of monitoring the fatigue degree of miners.

[0026] The data transmission module includes a first switch, multiple second switches and a wireless base station. The first switch, as the core switch, is connected to the ground dispatching platform. The first switch is connected to one of the second switches. Multiple second switches and the wireless base station are connected through fiber optic links to form a 10 Gigabit ring network, improving the transmission rate and accuracy of data.

[0027] The data acquisition module includes a smart wearable device and a video monitoring device. The smart wearable device is used to continuously collect physiological data, and the physiological data includes body temperature, heart rate and blood pressure. The video monitoring device is used to continuously collect video data, and the video data includes operation behavior and environmental data.

[0028] The smart wearable device real-time detects the body temperature, heart rate and blood pressure of miners, and sends the detected data to the ground dispatching platform for processing through the data transmission module. The smart wearable device includes a smart safety helmet, a mine intrinsically safe bracelet and a smart vest, improving the accuracy of data through multi-device data acquisition.

[0029] Miners correctly wear the smart safety helmet. The smart safety helmet continuously monitors the core physiological indicators of miners such as body temperature and heart rate. Data such as body temperature and heart rate are transmitted to the ground dispatching platform for processing through the data transmission module. The smart safety helmet also integrates a smart voice interaction device, which can instantly broadcast the current physiological health status of miners. At the same time, the smart voice interaction device, as an audio communication tool, enables miners to maintain real-time voice contact with the ground general dispatching platform and the ground management platform, improving the personal safety of miners.

[0030] Miners correctly wear intrinsically safe mine bracelets, which facilitates the continuous monitoring of miners' heart rate, blood pressure, blood oxygen saturation, and body temperature by the intrinsically safe mine bracelets. At the same time, the intrinsically safe mine bracelets can also monitor the temperature and humidity underground. The intrinsically safe mine bracelets transmit the monitored data to the ground dispatching platform through the data transmission module for data processing.

[0031] Miners wear intelligent vests equipped with electrocardiogram sensors. The electrocardiogram sensors continuously monitor the heart rate variability of miners and transmit the data to the ground dispatching platform, providing an important basis for evaluating the heart health status of miners. By collecting multiple data through multiple devices, the accuracy of the data is improved.

[0032] The video monitoring device uses a high-definition camera. The high-definition camera is installed in front of the miners' working positions, facilitating the collection of miners' facial expression data and underground environmental data. The high-definition camera has high resolution, night vision function, and remote control ability. The high-definition camera continuously collects the working behaviors of miners underground and transmits the collected data to the ground dispatching platform. The ground dispatching platform analyzes the video data and physiological data respectively to determine the fatigue degree of miners, improving the accuracy of judging the fatigue degree of miners.

[0033] The fatigue status monitoring system also includes a user terminal, which is used to receive warning information and operation intervention information.

[0034] The user terminal receives the warning information sent by the ground dispatching platform, reminding the miner that they are currently in a fatigued state. At the same time, the user terminal receives the subsequent operation intervention information sent by the background management platform. The user adjusts the subsequent operation tasks according to the operation intervention information to avoid physical damage caused by continuous high-intensity operations.

[0035] The user terminal can be the APP of the user's mobile phone terminal or any mine communication device that meets the information transmission and display functions.

[0036] The ground dispatching platform includes a data storage module, a data preprocessing module, and a data statistical analysis module. The data storage module is used to receive and store the physiological data and video data transmitted by the data transmission module; the data preprocessing module is used to clean and organize the physiological data and video data received by the data storage module to obtain feature information related to the fatigue state; the data statistical analysis module analyzes and processes the feature information to extract key features and determine the fatigue degree.

[0037] Physiological data and video data are transmitted to the data storage module through the data transmission module for storage. At the same time, the data storage module transmits the received physiological data and video data to the data preprocessing module. The data preprocessing module preliminarily processes the physiological data and video data respectively, removing outliers, smoothing noise, filling missing values, etc., to ensure the accuracy and reliability of the data. Then the data preprocessing module transmits the physiological data and video data to the data statistical analysis module. The data statistical analysis module extracts feature information related to the fatigue state from the physiological data and video data respectively, which is convenient for determining the fatigue state of the miners.

[0038] The data preprocessing module includes a physiological data processing unit and a video data processing unit. The physiological data processing unit processes the physiological data using wavelet transform to obtain the feature information of the physiological data; through operations such as stretching and translation, wavelet transform performs multi-scale focused analysis on the physiological data, so as to accurately extract useful information, that is, the feature information of the physiological data; the video data processing unit processes the video data using a convolutional neural network to obtain the feature information of the video data; the feature information includes heart rate variability, body temperature fluctuation pattern, periodic changes in blink frequency, etc.

[0039] The physiological data processing unit uses wavelet transform to clean the physiological data. The wavelet transform can be expressed by the following formula:

[0040]

[0041] In the formula: f(t) is the physiological signal to be analyzed; ψ a,b (t) is the wavelet basis function, is the complex conjugate of ψ a,b (t); a is the scale factor, which determines the stretching degree of the wavelet function; b is the translation factor, which determines the position of the wavelet function on the time axis; W(a, b) is the result of the wavelet transform, representing the feature of the physiological signal f(t) at scale a and position b.

[0042] The wavelet basis function ψ a,b (t) is defined as:

[0043]

[0044] By selecting an appropriate Haar wavelet basis function, that is, ψ a,b (t), wavelet transform performs multi-scale decomposition on the preprocessed physiological data to achieve time-frequency localization analysis of the signal, effectively extracting key features such as heart rate variability and body temperature fluctuation. At the same time, by discarding unimportant wavelet coefficients, noise reduction and data compression processing are completed.

[0045] The video data processing unit processes video data using a convolutional neural network. The video data processing unit iteratively optimizes network parameters, which include weights, biases, etc. These parameters are adjusted through an iterative optimization method to reduce the loss function value and improve the model's processing ability for video data, enabling the convolutional neural network to accurately classify video data. A loss function and an optimizer are used in the convolutional neural network. The loss function uses the entropy loss function. The cross-entropy loss function can measure the difference between the predicted value and the true value and is suitable for fatigue state problems. The optimizer uses the Adam optimizer. Based on the adaptive learning rate adjustment strategy of the Adam optimizer, it can handle non-convex optimization problems and has a fast convergence speed. Feature information is obtained from the video data through the convolutional neural network.

[0046] The predicted value is the output result after the convolutional neural network in the video data processing unit processes the video data, that is, the prediction result of the model for classifying the video data. This predicted value is usually a probability distribution representing the probabilities of the video belonging to each category; the true value is the actual category label of the video data, that is, the true classification result of the video data. During the training process, this true value is used to compare with the predicted value to calculate the loss function and optimize the model parameters.

[0047] The data statistical analysis module includes a first data processing unit and a second data processing unit. The first data processing unit uses an LSTM neural network structure to determine the physiological fatigue degree based on the feature information of physiological data; the second data processing unit uses the YOLO algorithm for recognition and analysis to determine the visual fatigue degree based on the feature information of video data.

[0048] LSTM (Long Short-Term Memory) is a special recurrent neural network structure. LSTM automatically extracts time series features in the data, including heart rate variability and blood pressure fluctuation patterns. These features are deeply learned using the LSTM recurrent neural network to capture the long-term dependencies and patterns in their time series, and a highly accurate prediction model for miners' fatigue states is trained.

[0049] LSTM introduces a forget gate, an input gate, an output gate, and a memory cell update. The forget gate determines the physiological data feature information to be discarded, the input gate determines the physiological data feature information to be stored, the output gate determines the physiological data feature information to be output, and the memory cell update continuously updates the state at each time step. The memory cell update is expressed by the formula:

[0050]

[0051] In the formula: C t is the memory cell state at the current time step; C t-1is the memory cell state at the previous time step; f t represents the forget gate, f t determines how much information in the previous cell state needs to be retained; i t represents the input gate, i t determines how much new information at the current time step is added to the memory cell; represents the candidate memory cell for updating the memory cell; the memory cell is updated by comparison to determine the degree of physiological fatigue.

[0052] The hidden state update formula is expressed as follows:

[0053] h t = o t * tanh(C t )

[0054] where: o t is the output of the output gate, which determines the output information; C t is the memory cell state at the current time step; h t is the hidden state at the current time step.

[0055] By continuously updating the hidden state to capture short-term and long-term changes in the physiological fatigue state in time series data, and using the gating mechanism to selectively focus on relevant features, accurate prediction and detection of an individual's fatigue state are achieved.

[0056] The hidden state update formula enables the LSTM network to capture long-term dependencies in sequence data by controlling the flow of information, updating the cell state, and outputting the hidden state, and is suitable for processing physiological data with temporality.

[0057] The YOLO algorithm identifies and analyzes video data, decomposes the video data into individual image frames, uses CNN (Convolutional Neural Networks) to extract features from the images. The CNN extracts hierarchical edge, texture, and shape features from the input images through multiple convolutional and pooling operations. To improve the accuracy and robustness of object detection, different scales of convolutional kernels and the FPN (Feature Pyramid Network) structure are used. The YOLO algorithm can capture feature information at different scales and fuse this information to determine the fatigue state of the miner.

[0058] In this embodiment, video data of miners at work is collected, the facial image data of miners at work is preprocessed by a video data processing unit, and then the YOLO algorithm is used to extract facial features and train a model to identify fatigue signs such as closing eyes and yawning, to achieve real-time monitoring and evaluation of the fatigue state of miners.

[0059] The fatigue level monitoring model is determined by physiological fatigue and visual fatigue, and is specifically expressed as:

[0060] Fatigue level = 0.7 * physiological fatigue + 0.3 * visual fatigue

[0061] The fatigue level formula is used to quantitatively evaluate the fatigue state of miners to ensure the objectivity and accuracy of the evaluation results; the physiological fatigue level is obtained through physiological index detection. The physiological fatigue level is divided into four levels: mild fatigue, moderate fatigue, severe fatigue, and extremely severe fatigue, with values of 1, 2, 3, and 4 respectively; the visual fatigue level is obtained through video data analysis. The visual fatigue level is divided into four levels: mild fatigue, moderate fatigue, severe fatigue, and extremely severe fatigue, with values of 1, 2, 3, and 4 respectively. The current fatigue level of miners can be determined by combining the physiological fatigue level and the visual fatigue level.

[0062] The criteria for dividing fatigue levels are shown in Table 1.

[0063] Table 1

[0064] 0 < Fatigue level <= 1 Mild fatigue 1 < degree of fatigue <= 2 Moderate fatigue 2 < degree of fatigue <= 3 Severe fatigue 3 < fatigue level ≤ 4 Extremely severe fatigue

[0065] The job intervention measures include mild fatigue management strategies, moderate fatigue intervention measures, severe fatigue emergency response measures, and extremely severe fatigue comprehensive adjustment measures.

[0066] The mild fatigue management strategy means that for miners who frequently experience mild fatigue within a month, it is necessary to strengthen the health monitoring system of the miner group, implement regular physical examinations and comprehensive mental health assessments, and at the same time, carry out systematic fatigue management and prevention education and training programs, aiming to improve the miners' awareness of the hazards of fatigue and their self-protection ability, and reduce the risk of mild fatigue transforming into a higher level.

[0067] Moderate fatigue intervention measures: When it is monitored that a miner is in a moderate fatigue state, immediately start the tracking and observation mechanism to work for 20 minutes, observe whether the physiological data fatigue level of the miner is obvious and whether the facial recognition fatigue frequency increases, and evaluate whether there is a possibility of further deterioration to severe fatigue, so as to take more active intervention measures in a timely manner.

[0068] Severe fatigue emergency response: For miners with severe fatigue who reach the warning threshold, immediately arrange a mandatory rest period to ensure that they can return to work after fully recovering their physical strength and mental state, so as to avoid potential health risks and decreased work efficiency caused by fatigue accumulation.

[0069] Comprehensive adjustment for extremely severe fatigue: If a miner shows extremely severe fatigue multiple times within a month, it is necessary to flexibly adjust the assignment of their work tasks based on their specific fatigue level and the nature and intensity of the current job; specifically, tasks involving high-risk operations or high physical exertion should be avoided for miners in an extremely severe fatigue state, and instead, they should be arranged to engage in low-intensity and low-risk work, or provided with a necessary rest period to promote the full recovery of their physical and mental state.

[0070] The background management platform includes a data storage module and a data display module. The data storage module is used to store the received warning information and fatigue level data, and the data display module is used to display the fatigue level; at least one user is registered on the background management platform, and the background management platform is used to manage the fatigue state information of each user.

[0071] After receiving the warning information and fatigue level data transmitted by the ground dispatching platform, the background management platform stores the data in the storage module, and the storage module sends the received data to the display module for page display, facilitating data viewing by management personnel.

[0072] Through the user terminal and multiple users registered on the background management platform, the fatigue state characteristic information of each user can be managed on the background management platform, facilitating centralized processing and big data analysis of user data, and facilitating the promotion of application scenarios. For example, it can be used for monitoring the fatigue state of high-altitude workers, monitoring driver fatigue driving, monitoring the physical health of the elderly, and so on.

[0073] The above is only a preferred embodiment of the present invention and does not impose any limitation on the present invention. Any simple modification, change, and equivalent structural transformation made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A miner fatigue state monitoring system based on multi-source data fusion, characterized in that: It includes a data acquisition module, a ground scheduling platform, and a background management platform; The data acquisition module is used to acquire multi-source data; The ground scheduling platform receives and stores the multi-source data, analyzes the multi-source data respectively to obtain corresponding fatigue state characteristic information, determines the fatigue degree based on multiple pieces of the fatigue state characteristic information in combination with a fatigue degree monitoring model, and triggers a warning reminder when the fatigue degree exceeds a set threshold; The background management platform is used to receive warning information and the fatigue degree, and perform operation intervention processing based on the warning information and the fatigue degree.

2. The miner fatigue state monitoring system based on multi-source data fusion according to claim 1, characterized in that: The multi-source data includes physiological data and video data, the fatigue state includes physiological fatigue and visual fatigue, the physiological fatigue is obtained based on the physiological data, and the visual fatigue is obtained based on the video data.

3. The miner fatigue state monitoring system based on multi-source data fusion according to claim 2, characterized in that: The data acquisition module includes a smart wearable device and a video monitoring device, The smart wearable device is used to continuously acquire physiological data, and the physiological data includes body temperature, heart rate, and blood pressure; The video monitoring device is used to continuously acquire video data, and the video data includes operation behaviors and environmental data.

4. A miner fatigue state monitoring system based on multi-source data fusion according to claim 1, characterized in that: The fatigue state monitoring system further includes a user terminal, and the user terminal is used to receive warning information and operation intervention information.

5. The miner fatigue state monitoring system based on multi-source data fusion according to claim 2, characterized in that: The ground scheduling platform includes a data storage module, a data preprocessing module, and a data statistical analysis module, The data storage module is used to receive and store the physiological data and video data transmitted by the data transmission module; The data preprocessing module is used to clean and sort the physiological data and video data received by the data storage module to obtain characteristic information related to the fatigue state; The data statistical analysis module analyzes and processes the characteristic information, and extracts key features to determine the fatigue degree.

6. The miner fatigue state monitoring system based on multi-source data fusion according to claim 5, characterized in that: The data preprocessing module includes a physiological data processing unit and a video data processing unit, The physiological data processing unit processes the physiological data using wavelet transform to obtain characteristic information of the physiological data; The video data processing unit processes the video data using a convolutional neural network to obtain characteristic information of the video data.

7. A miner fatigue state monitoring system based on multi-source data fusion according to claim 6, characterized in that: The data statistical analysis module includes a first data processing unit and a second data processing unit, The first data processing unit uses an LSTM neural network structure to determine the physiological fatigue degree based on the characteristic information of the physiological data; The second data processing unit uses the YOLO algorithm for recognition and analysis to determine the visual fatigue degree based on the characteristic information of the video data.

8. A miner fatigue state monitoring system based on multi-source data fusion according to claim 2, characterized in that: The fatigue degree monitoring model is determined by physiological fatigue and visual fatigue, and is specifically expressed as: Fatigue degree = physiological fatigue * 0.7 + visual fatigue * 0.

3.

9. The miner fatigue state monitoring system based on multi-source data fusion according to claim 1, characterized in that: The operation intervention measures include mild fatigue management strategies, moderate fatigue intervention measures, severe fatigue emergency response measures, and extremely severe fatigue comprehensive adjustment measures.

10. A miner fatigue state monitoring system based on multi-source data fusion according to any one of claims 1-9, characterized in that: The background management platform includes a data storage module and a data display module. The data storage module is used to store the received warning information and fatigue degree data, and the data display module is used to display the fatigue degree; At least one user is registered on the background management platform, and the background management platform is used to manage the fatigue state information of each user.

Images