User operation risk early warning method and device, computer equipment and storage medium

Abstract

The invention relates to a user operation risk early warning method and device, computer equipment and a storage medium. The method comprises the following steps: a server obtains physiological signal data and operation data of a to-be-discriminated user, then inputs the physiological signal data into a pre-constructed hidden danger risk assessment model to obtain a hidden danger risk assessment result of the to-be-discriminated user, and inputs the operation data into a pre-constructed posture risk assessment model to obtain a posture risk assessment result of the to-be-discriminated user; and obtaining a posture risk assessment result of the to-be-discriminated user, then obtaining a risk discrimination result of the to-be-discriminated user according to the hidden danger risk assessment result and the posture risk assessment result, and finally obtaining user operation risk early warning information of the to-be-discriminated user according to the risk discrimination result. And returning to the terminal of the to-be-discriminated user to carry out user operation risk early warning. By adopting the method, the objectivity of risk judgment is realized, and meanwhile, the accuracy and credibility of risk judgment are also enhanced.

Description

Technical Field

[0001] The present application relates to the technical field of personal risk identification, and in particular to a user operation risk warning method, device, computer equipment, computer-readable storage medium and computer program product. Background Art

[0002] In modern industrial production and on-site operations, personal safety is always a crucial issue. In the past, the identification of personal risks mainly relied on empirical judgment or simple safety rules. This method can make decisions quickly, has low implementation costs and strong flexibility, but the identification results obtained by this method are relatively rough, the scope of application is extremely limited, and there are also problems such as strong subjectivity and lack of precision. Summary of the invention

[0003] Based on this, it is necessary to provide a user operation risk warning method, device, computer equipment, computer-readable storage medium and computer program product that can improve the objectivity and accuracy of judgment in response to the above technical problems.

[0004] In a first aspect, the present application provides a user operation risk early warning method, comprising:

[0005] Acquiring physiological signal data of the user to be identified, and acquiring operation data of the user to be identified;

[0006] Inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified;

[0007] According to the hidden danger risk assessment results and posture risk assessment results, the risk identification results of the user to be identified are obtained;

[0008] Based on the risk identification result, the user operation risk warning information of the user to be identified is obtained, and the user operation risk warning is returned to the terminal of the user to be identified.

[0009] In one embodiment, obtaining physiological signal data of a user to be identified includes:

[0010] Acquire original physiological signal data of the user to be identified; the original physiological signal data includes first original physiological signal data and second original physiological signal data; the first original physiological signal data is disturbed original physiological signal data; the second physiological signal data is undisturbed original physiological signal data;

[0011] Multiplying the first original physiological signal data and a pre-constructed random matrix to obtain transformed first original physiological signal data;

[0012] Physiological signal data is obtained according to the transformed first original physiological signal data and the second original physiological signal data.

[0013] In one embodiment, the transformed first original physiological signal data includes third original physiological signal data, fourth original physiological signal data and fifth original physiological signal data; wherein the third original physiological signal data is original physiological signal data subject to electromechanical interference, the fourth original physiological signal data is original physiological signal data subject to motion artifact interference, and the fifth original physiological signal data is original physiological signal data without interference;

[0014] Obtaining physiological signal data according to the transformed first original physiological signal data and the second original physiological signal data includes:

[0015] Using a preset band-stop filter to filter the third original physiological signal data to obtain filtered third original physiological signal data, and using a preset low-pass filter to filter the fourth original physiological signal data to obtain filtered fourth original physiological signal data;

[0016] performing sum processing on the second original physiological signal data, the filtered third original physiological signal data, the filtered fourth original physiological signal data, and the fifth original physiological signal data to obtain physiological signal data to be processed;

[0017] The physiological signal data to be processed is subjected to denoising and normalization processing to obtain the physiological signal data.

[0018] In an exemplary embodiment, the working data includes working video data and acceleration data;

[0019] Get the job data of the user to be identified, including:

[0020] Performing downsampling, normalization and denoising processing on the working video data to obtain processed working video data, and performing denoising and normalization processing on the acceleration data to obtain processed acceleration data;

[0021] Acquire a corresponding working video vector based on the processed working video data, and acquire a corresponding acceleration feature vector based on the processed acceleration data;

[0022] Obtaining a first weight corresponding to the working video feature vector, and obtaining a second weight corresponding to the acceleration feature vector;

[0023] Based on the first weight and the second weight, weighted summation is performed on the work video feature vector and the acceleration feature vector to obtain the work data feature vector of the user to be identified;

[0024] Input the operation data into the pre-built posture risk assessment model to obtain the posture risk assessment results of the user to be identified, including:

[0025] The operation data feature vector is input into a pre-built posture risk assessment model to obtain a high-order feature vector corresponding to the operation data feature vector, and the posture risk assessment result of the user to be identified is obtained based on the high-order feature vector; the high-order feature vector is used to distinguish different posture information of the user to be identified.

[0026] In one embodiment, based on the hidden danger risk assessment results and the posture risk assessment results, a risk assessment result of the user to be assessed is obtained, including:

[0027] Obtaining a third weight corresponding to the hidden danger risk assessment result, and obtaining a fourth weight corresponding to the posture risk assessment result;

[0028] The third weight and the fourth weight are used to perform weighted processing on the hidden danger risk assessment result and the posture risk assessment result to obtain the risk identification result of the user to be identified.

[0029] In one of the embodiments, the risk determination result is characterized by a risk value;

[0030] Based on the risk identification results, the risk warning information of the user to be identified is obtained, including:

[0031] When the risk value is greater than a preset risk value threshold, determining the personnel status information of the user to be identified based on the physiological signal data and the operation data;

[0032] Determine the corresponding work ticket content and on-site operation risk information based on the personnel status information, and obtain the corresponding safety warning information and response measures information according to the work ticket content and on-site operation risk information;

[0033] Generate user operation risk warning information based on safety warning information and response measures information.

[0034] In a second aspect, the present application also provides a user operation risk warning device, comprising:

[0035] A data acquisition module, used to acquire physiological signal data of the user to be identified, and to acquire operation data of the user to be identified;

[0036] An assessment result acquisition module is used to input physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and to input operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified;

[0037] A discrimination result determination module is used to obtain a risk discrimination result of the user to be discriminated based on the hidden danger risk assessment result and the posture risk assessment result;

[0038] The warning information determination module is used to obtain the user operation risk warning information of the user to be identified based on the risk identification result, and return the user operation risk warning to the terminal of the user to be identified.

[0039] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0040] Acquiring physiological signal data of the user to be identified, and acquiring operation data of the user to be identified;

[0041] Inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified;

[0042] According to the hidden danger risk assessment results and posture risk assessment results, the risk identification results of the user to be identified are obtained;

[0043] Based on the risk identification result, the user operation risk warning information of the user to be identified is obtained, and the user operation risk warning is returned to the terminal of the user to be identified.

[0044] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0045] Acquiring physiological signal data of the user to be identified, and acquiring operation data of the user to be identified;

[0046] Inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified;

[0047] According to the hidden danger risk assessment results and posture risk assessment results, the risk identification results of the user to be identified are obtained;

[0048] Based on the risk identification result, the user operation risk warning information of the user to be identified is obtained, and the user operation risk warning is returned to the terminal of the user to be identified.

[0049] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0050] Acquiring physiological signal data of the user to be identified, and acquiring operation data of the user to be identified;

[0051] Inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified;

[0052] According to the hidden danger risk assessment results and posture risk assessment results, the risk identification results of the user to be identified are obtained;

[0053] Based on the risk identification result, the user operation risk warning information of the user to be identified is obtained, and the user operation risk warning is returned to the terminal of the user to be identified.

[0054] In the above-mentioned user operation risk warning method, device, computer equipment, computer-readable storage medium and computer program product, the server obtains physiological signal data and operation data of the user to be identified, then inputs the physiological signal data into a pre-constructed hidden danger risk assessment model to obtain the hidden danger risk assessment result of the user to be identified, inputs the operation data into a pre-constructed posture risk assessment model to obtain the posture risk assessment result of the user to be identified, and then obtains the risk discrimination result of the user to be identified based on the hidden danger risk assessment result and the posture risk assessment result, and finally obtains the user operation risk warning information of the user to be identified based on the risk discrimination result, and returns to the terminal of the user to be identified for user operation risk warning. The pre-constructed model is used to process the obtained relevant data respectively to obtain the corresponding risk assessment result, and finally the corresponding risk discrimination result is determined based on the risk assessment result, and the risk warning information is further obtained based on the risk discrimination result, so as to introduce the model to process the data, thereby avoiding the subjectivity caused by human processing, and comprehensively considering the hidden danger risk and posture risk to enhance the accuracy and credibility of risk discrimination. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0056] FIG1 is an application environment diagram of a user operation risk early warning method in one embodiment;

[0057] FIG2 is a schematic diagram of a process flow of a user operation risk early warning method in one embodiment;

[0058] FIG3 is a schematic diagram of a process for obtaining risk identification results in one embodiment;

[0059] FIG4 is a schematic diagram of a process for generating risk warning information in another embodiment;

[0060] FIG5 is a schematic flow chart of a method for early warning of user operation risks in another embodiment;

[0061] FIG6 is a schematic diagram of the structure of an electric power knowledge parsing tree in one embodiment;

[0062] FIG7 is a structural block diagram of a user operation risk warning device in one embodiment;

[0063] FIG. 8 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] The user operation risk warning method provided in the embodiment of the present application can be applied to the application environment shown in Figure 1. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 obtains the physiological signal data and operation data of the user to be identified from the terminal 102, and then inputs the physiological signal data into the pre-built hidden danger risk assessment model to obtain the hidden danger risk assessment result of the user to be identified, and inputs the operation data into the pre-built posture risk assessment model to obtain the posture risk assessment result of the user to be identified, and then obtains the risk discrimination result of the user to be identified based on the hidden danger risk assessment result and the posture risk assessment result, and finally obtains the user operation risk warning information of the user to be identified based on the risk discrimination result, and returns to the terminal of the user to be identified for user operation risk warning. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device may be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0066] In an exemplary embodiment, as shown in FIG2 , a user operation risk warning method is provided, which is described by taking the method applied to the server 104 in FIG1 as an example, and includes the following steps S201 to S204. Among them:

[0067] Step S201 , obtaining physiological signal data of a user to be identified, and obtaining operation data of the user to be identified.

[0068] Step S202, inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified.

[0069] Among them, physiological signal data can be understood as measurable information generated by internal or external physiological activities of the organism, such as a series of time series data such as electrocardiogram, electroencephalogram, electromyogram, body temperature, pulse and respiratory rate; operation data can be understood as measurable information that reflects the user's working status in the current operation scenario, which includes multiple types of measurable information.

[0070] Among them, the hidden danger risk assessment model is disclosed in the present application but is not limited to a time series network. The time series network can be understood as a network model specifically used to process time series data; the hidden danger risk assessment result can be understood as the risk level of the user suffering from an acute disease; the posture risk assessment model is disclosed in the present application but is not limited to a multimodal neural network. The multimodal neural network can be understood as a network model that comprehensively considers different types of measurable information; the posture risk assessment result can be understood as the assessment result of whether the user's current posture in the current working scenario will bring safety risks.

[0071] For example, the server 104 obtains the physiological signal data and operation data of the user to be identified from the terminal 102 through the network, inputs the physiological signal data into the pre-built hidden danger risk assessment model to obtain the hidden danger risk assessment result of the user to be identified, and inputs the operation data into the pre-built posture risk assessment model to obtain the posture risk assessment result of the user to be identified. By introducing the risk assessment model to perform risk assessment on the acquired physiological signal data and operation data, the objectivity of the risk assessment result is improved, and a data foundation is laid for the subsequent acquisition of risk identification results.

[0072] Step S203, obtaining a risk identification result of the user to be identified according to the hidden danger risk assessment result and the posture risk assessment result.

[0073] Step S204, based on the risk identification result, obtain the user operation risk warning information of the user to be identified, and return it to the terminal of the user to be identified for user operation risk warning.

[0074] Optionally, the server 104 obtains preset weight information for the hidden danger risk assessment result and the posture risk assessment result, and performs weighted summation on the hidden danger risk assessment result and the posture risk assessment result based on the corresponding weight information to obtain the risk identification result of the user to be identified, and further determines whether to generate user operation risk warning information based on the risk identification result. In the case of determining to generate, the user operation risk warning information of the user to be identified is obtained based on the physiological signal data and operation data of the user to be identified previously obtained, and returns to the terminal of the user to be identified for user operation risk warning. The risk identification result is obtained by performing weighted summation on the hidden danger risk assessment result and the posture risk result, so as to improve the accuracy of the risk identification result, and the corresponding user operation risk warning information is obtained based on the physiological signal data and operation data when the preset conditions are met, so as to ensure the adaptability of the warning information, and facilitate timely warning to ensure the personal safety of the user.

[0075] In the above-mentioned user operation risk warning method, the server obtains the physiological signal data and operation data of the user to be identified, and then inputs the physiological signal data into the pre-constructed hidden danger risk assessment model to obtain the hidden danger risk assessment result of the user to be identified, and inputs the operation data into the pre-constructed posture risk assessment model to obtain the posture risk assessment result of the user to be identified, and then obtains the risk discrimination result of the user to be identified based on the hidden danger risk assessment result and the posture risk assessment result, and finally obtains the user operation risk warning information of the user to be identified based on the risk discrimination result, and returns to the terminal of the user to be identified for user operation risk warning. The pre-constructed model is used to process the obtained relevant data respectively to obtain the corresponding risk assessment result, and finally the corresponding risk discrimination result is determined based on the risk assessment result, and the risk warning information is further obtained based on the risk discrimination result, so as to introduce the model to process the data, thereby avoiding the subjectivity caused by human processing, and comprehensively considering the hidden danger risk and posture risk to enhance the accuracy and credibility of risk discrimination.

[0076] In one embodiment, obtaining physiological signal data of a user to be identified includes: obtaining original physiological signal data of the user to be identified; the original physiological signal data includes first original physiological signal data and second original physiological signal data; the first original physiological signal data is disturbed original physiological signal data; the second physiological signal data is undisturbed original physiological signal data; multiplying the first original physiological signal data and a pre-constructed random matrix to obtain transformed first original physiological signal data; and obtaining physiological signal data based on the transformed first original physiological signal data and the second original physiological signal data.

[0077] The random matrix can be understood as a matrix A of m*n dimensions, where m is the dimension of the first original physiological signal data collected, and n is the dimension of the pure physiological signal part that needs to be separated.

[0078] For example, assuming that the first original physiological signal data is x and the second original physiological signal data is s; the collected first original signal data x is multiplied by the constructed random matrix A to obtain the transformed first original signal data y=Ax, and then the physiological signal data is obtained according to y and s. The random matrix technology is used to reduce the data dimension of the first original physiological signal data, which enhances the distinction between different signals, namely, pure physiological signal data and disturbed physiological signal data, enhances the validity and availability of the original physiological signal data, and thus improves the model output effect of the hidden danger risk assessment model.

[0079] In one embodiment, the transformed first original physiological signal data includes third original physiological signal data, fourth original physiological signal data and fifth original physiological signal data; wherein the third original physiological signal data is original physiological signal data subject to electromechanical interference, the fourth original physiological signal data is original physiological signal data subject to motion artifact interference, and the fifth original physiological signal data is original physiological signal data without interference;

[0080] Obtaining physiological signal data according to the transformed first original physiological signal data and the second original physiological signal data includes:

[0081] The third original physiological signal data is filtered using a preset band-stop filter to obtain filtered third original physiological signal data, and the fourth original physiological signal data is filtered using a preset low-pass filter to obtain filtered fourth original physiological signal data; the second original physiological signal data, the filtered third original physiological signal data, the filtered fourth original physiological signal data and the fifth original physiological signal data are added together to obtain the physiological signal data to be processed; the physiological signal data to be processed are denoised and normalized to obtain the physiological signal data.

[0082] Optionally, the transformed first original physiological signal data y is decomposed into two parts, one part is the disturbed part, namely the third original physiological signal data and the fourth original signal data x', and the other part is the pure physiological signal part, namely the fifth original physiological signal data s', then y=x'+s', the disturbed part x' includes x 1 and x 2 , where x 1 represents the part affected by electromechanical interference, i.e., the third original physiological signal data, x 2The fourth original physiological signal data is represented by the part affected by motion artifacts, and the part affected by electromechanical interference x is filtered by a band-stop filter. 1 Separate to get x 1 ', use a low-pass filter to filter the part x affected by motion artifacts 2 Separate to get x 2 '. 1 ' and x 2 ' Perform summation to obtain the physiological signal data x without interference''=x 1 '+x 2 '.

[0083] The interfering part x will be removed 1 '' and the pure physiological signal part s' are synthesized to obtain the final physiological signal data s''=x''+s' after interference removal, and finally the physiological signal data s+s'' to be processed is obtained. The physiological signal data s+s'' to be processed is preprocessed, including denoising and normalization, and the data is organized into a format suitable for the input of the time series network model, namely, physiological signal data. In the above manner, the interference of the disturbed physiological signal data is removed, and the physiological signal data to be processed after the interference is removed is preprocessed, which further improves the effectiveness and availability of the physiological signal data, thereby improving the model output effect of the hidden danger risk assessment model.

[0084] In one embodiment, the working data includes working video data and acceleration data;

[0085] Get the job data of the user to be identified, including:

[0086] Perform downsampling, normalization and denoising on the working video data to obtain processed working video data, and perform denoising and normalization on the acceleration data to obtain processed acceleration data; obtain a corresponding working video vector based on the processed working video data, and obtain a corresponding acceleration feature vector based on the processed acceleration data; obtain a first weight corresponding to the working video feature vector, and obtain a second weight corresponding to the acceleration feature vector; perform weighted summation on the working video feature vector and the acceleration feature vector based on the first weight and the second weight to obtain a work data feature vector of the user to be identified;

[0087] Input the operation data into the pre-built posture risk assessment model to obtain the posture risk assessment results of the user to be identified, including:

[0088] The operation data feature vector is input into a pre-built posture risk assessment model to obtain a high-order feature vector corresponding to the operation data feature vector, and the posture risk assessment result of the user to be identified is obtained based on the high-order feature vector; the high-order feature vector is used to distinguish different posture information of the user to be identified.

[0089] Among them, work video data can be understood as video content that records work scenes, processes or activities; acceleration data can be understood as the change in speed per unit time, which is expressed as changes in speed and direction, and is used to indicate the user's current motion state and information about the next possible action to be taken; high-order feature vectors can be understood as more abstract and complex feature information extracted from the original data, and are also feature vectors that can better distinguish different posture information of the user; the first weight can be understood as the importance of the influence of the work video feature vector; the second weight can be understood as the importance of the influence of the acceleration feature vector.

[0090] For example, the server 104 performs downsampling, normalization and denoising on the work video data to obtain the processed work video data, and performs denoising and normalization on the acceleration data to obtain the processed acceleration data, and uses a convolutional neural network to extract features from the processed work video data to obtain the corresponding work video feature vector, and uses frequency domain analysis to extract features from the processed acceleration data to obtain the corresponding acceleration feature vector. Then, a first weight corresponding to the work video feature vector is obtained, and a second weight corresponding to the acceleration feature vector is obtained. Based on the first weight and the second weight, the work video feature vector and the acceleration feature vector are weighted and summed to obtain the operation data feature vector of the user to be identified.

[0091] Taking the posture risk assessment model as a multimodal neural network as an example, the server 104 inputs the operation data feature vector into the multimodal neural network, and uses the hidden layer of the multimodal neural network to propose a high-order feature vector in the operation data feature vector. The output layer of the multimodal neural network then performs posture classification prediction on the high-order feature vector and outputs the corresponding posture risk assessment result.

[0092] By preprocessing the operation data to be input into the posture risk assessment model as described above, and using different data processing methods for different operation data, the validity and availability of various types of operation data are guaranteed to a great extent, thereby improving the model output effect of the posture risk assessment model.

[0093] In an exemplary embodiment, as shown in FIG3 , a risk identification result of a user to be identified is obtained according to a hidden danger risk assessment result and a posture risk assessment result, including steps S301 to S302 . Among them:

[0094] Step S301, obtaining a third weight corresponding to a hidden danger risk assessment result, and obtaining a fourth weight corresponding to a posture risk assessment result.

[0095] Step S302: weighting the hidden danger risk assessment result and the posture risk assessment result using the third weight and the fourth weight to obtain a risk identification result of the user to be identified.

[0096] Among them, the third weight can be understood as the weight set according to the degree of influence of historical hidden danger risk assessment results on the operation risk judgment results by experts in this field. The fourth weight can be understood as the weight set by experts in this field on the operation risk judgment results based on historical posture risk assessment results. The setting of the above weights takes into account multiple factors.

[0097] Optionally, the server 104 obtains a third weight corresponding to the hidden danger risk assessment result and a fourth weight corresponding to the posture risk assessment result from the data storage system, and uses the third weight and the fourth weight to weight the hidden danger risk assessment result and the posture risk assessment result to obtain a risk identification result of the user to be identified. By comprehensively considering the hidden danger risk assessment result and the posture risk assessment result, the accuracy and objectivity of the risk identification result are improved.

[0098] In one embodiment, as shown in FIG. 4 , the risk identification result is characterized by a risk value; based on the risk identification result, risk warning information of the user to be identified is obtained, including steps S401 to S403, wherein:

[0099] Step S401, when the risk value is greater than a preset risk value threshold, determining the personnel status information of the user to be identified according to the physiological signal data and the operation data.

[0100] The status information may be understood as the situation information of the user in different working environments or corresponding risk environments, including the user's own physiological information and corresponding working status information.

[0101] For example, when the server 104 determines that the risk value representing the risk identification result is greater than a preset risk data threshold, it determines to initiate a risk warning, and further determines the personnel status information of the user to be identified based on the physiological signal data and the operation data, laying a data foundation for the subsequent generation of user operation risk warning information.

[0102] Step S402, determining the corresponding work ticket content and on-site operation risk information based on the personnel status information, and obtaining corresponding safety warning information and response measures information according to the work ticket content and on-site operation risk information.

[0103] Step S403: Generate user operation risk warning information based on the safety warning information and the countermeasure information.

[0104] Among them, the content of the work ticket can be understood as basic work information, including work tasks, work locations, work time, operating personnel, etc. On-site work risk information can be understood as the assessment and record of possible dangers and risks when performing work in a specific working environment, including electrical shock, work height, and equipment failure; safety warning information can be understood as reminding staff to pay attention to potential risks during task execution, such as special equipment operation, material use and other reminder information; response measures information can be understood as ensuring a safe environment at the work site, such as whether fire-fighting facilities are complete and whether the work area is clearly marked.

[0105] Optionally, the personnel status information corresponds to the work ticket content and the on-site operation risk information. The server 104 determines the corresponding work ticket content and the on-site operation risk information based on the personnel status information, and obtains the corresponding safety warning information and countermeasure information according to the work ticket content and the on-site operation risk information, and finally combines the safety warning information and the countermeasure information to generate the user operation risk warning information. The user operation risk matching the user to be identified is generated according to the personnel status information, which is conducive to protecting the personal safety of the user to be identified.

[0106] In an exemplary embodiment, as shown in FIG5 , a specific embodiment of a user operation risk early warning method is provided, including steps S501 to S507, wherein:

[0107] Step S501a, obtaining the original physiological signal data of the user to be identified:

[0108] 1. Workers wear sensors or wearable devices to collect physiological signal data in real time to form a time series. The data includes heart rate, blood pressure, and body temperature, which is expressed as {HR, BP, T}, where t represents the time point, HR represents heart rate, BP represents blood pressure, and T represents body temperature.

[0109] 2. Preprocess the collected physiological signal data, including denoising and normalization, and organize the data into a format suitable for input into the time series network model.

[0110] Step S502a, using random matrix technology to process the original physiological signal data to obtain processed original physiological signal data:

[0111] Specifically, the specific process of removing electromechanical interference and motion artifact interference from the physiological signals of the operator using random matrix technology is as follows:

[0112] 1. Assume that the collected interfered signal is x and the actual physiological signal is s;

[0113] 2. Construct a random matrix A, which is an m×n dimensional matrix that matches the collected physiological signal data, where m is the dimension of the collected signal data and n is the dimension of the pure physiological signal part that needs to be separated;

[0114] 3. Multiply the collected interfered signal x with the constructed random matrix A to obtain the transformed signal data y=Ax;

[0115] 4. Decompose the transformed signal data y into two parts, one is the disturbed part x', and the other is the pure physiological signal part s', then y=x'+s', the disturbed part x' includes x 1 and x 2 , where x 1 Indicates the part affected by electromechanical interference, x 2 represents the part affected by motion artifacts, and the part affected by electromechanical interference x is filtered by 1 and the part affected by motion artifacts x 2 Separate and obtain the physiological signal x'' with the interference part removed.

[0116] 5. The interference-removed part x'' and the pure physiological signal part s' are synthesized to obtain the final physiological signal s''=x''+s' after removing electromechanical interference.

[0117] Step S503a, using a filter to process the processed original physiological signal data to obtain physiological signal data:

[0118] Specifically, in step S24, filtering is used to filter the part x affected by electromechanical interference. 1 and the part affected by motion artifacts x 2 The process of separating and obtaining the physiological signal x'' without interference includes:

[0119] Use band-stop filter to filter the part affected by electromechanical interference 1 Separate to get x 1 ', use a low-pass filter to filter the part x affected by motion artifacts 2 Separate to get x 2 ';

[0120] The x 1 ' and x 2 'Synthesize to obtain the interference removal part x'', wherein x''=x 1 '+x 2 '.

[0121] Step S504a, input the physiological signal data into the time series network to obtain the hidden danger risk identification result of the user to be identified:

[0122] 1. Use the time series network model to identify potential acute disease risks by learning the rules and characteristics in the time series data. The acute disease risks include A, B, and C.

[0123] I_disease = f(BP_t,T_t,HR_t,H,M)

[0124] Among them, I_disease represents the acute disease hidden danger assessment result, f represents the acute disease hidden danger assessment function, BP_t represents blood pressure, T_t represents body temperature, HR_t represents heart rate, H represents disease history, and M represents medication status;

[0125] Step S501b, obtaining the operation data of the user to be identified; the operation data includes work video data and acceleration data.

[0126] Step S502b, performing data preprocessing on the working video data and the acceleration data, and obtaining a working video feature vector corresponding to the preprocessed working video data and an acceleration feature vector corresponding to the preprocessed acceleration data.

[0127] Step S503b, performing weighted summation on the work video feature vector and the acceleration feature vector according to the first weight corresponding to the work video feature vector and the second weight corresponding to the acceleration feature vector, to obtain the operation data feature vector.

[0128] Step S504b, input the operation feature vector into the multimodal neural network to obtain the posture risk identification result of the user to be identified:

[0129] 1. Use the hidden layers of the multimodal neural network to extract high-order features from the input data;

[0130] 2. Use the output layer of the multimodal neural network to perform posture classification prediction on the high-order features in the input data and output the risk assessment results.

[0131] I_posture = g(S,F,A)

[0132] Among them, I_posture represents the evaluation result of the staff posture, g represents the staff posture evaluation function, S represents the posture stability, F represents the action frequency, and A represents the action amplitude;

[0133] Step S505, obtaining a third weight corresponding to the hidden danger risk assessment result, and obtaining a fourth weight corresponding to the posture risk assessment result:

[0134] According to expert experience and domain knowledge, the weight of acute disease risk is determined as w_disease and the weight of staff posture is determined as w_posture;

[0135] Step S506: weighting the hidden danger risk assessment result and the posture risk assessment result using the third weight and the fourth weight to obtain a risk identification result of the user to be identified:

[0136] Risk = w_disease*I_disease + w_posture*I_posture

[0137] Among them, Risk represents the comprehensive risk assessment result.

[0138] Step S507, when the risk value represented by the risk identification result is greater than the preset risk value threshold, the user operation risk warning information is obtained by combining the physiological signal data, the operation data and the preset power knowledge safety analysis tree, and returned to the terminal of the user to be identified for user operation risk warning.

[0139] The power knowledge safety parsing tree is shown in FIG5 . Referring to FIG5 , the step S507 specifically includes:

[0140] The content of the work ticket includes work tasks, work location, work time, and workers. The on-site work risks include electrical shock and work height. The personnel status information includes health status and safety warning measures for each on-site work risk to construct an electric power safety knowledge parsing tree.

[0141] For example, the work ticket includes the work task, countermeasures: ensure that the staff clearly understands the specific requirements and steps of the work task, as well as the corresponding safety operating procedures. Safety warning: remind the staff to pay attention to potential risks during the execution of the task, such as special equipment operation, material use, etc.

[0142] Work site, countermeasures: Ensure the safety environment of the work site, such as whether the fire-fighting facilities are complete, whether the work area is clearly marked, etc. Safety warning: remind staff to pay attention to potential safety hazards at the work site, such as slipping on the ground, high temperature and high pressure areas, etc.

[0143] Working time, countermeasures: arrange working time reasonably, avoid fatigue and overtime. Safety warning: remind workers to pay attention to the physical and mental stress that long working hours may bring.

[0144] On-site operation risks include electric shock. Countermeasures: Use insulating tools, wear insulating gloves and other protective measures to ensure safe operation. Safety warning: Remind workers to pay attention to the safe distance and operating procedures around high-voltage equipment.

[0145] Equipment failure, countermeasures: Regularly check the equipment status, and promptly maintain and repair the equipment with problems. Safety warning: Remind the staff to pay attention to abnormal sounds, vibrations and other signs during the operation of the equipment.

[0146] Working height, countermeasures: Wear safety belts and use appropriate climbing tools to ensure the personal safety of workers. Safety warning: remind workers to pay attention to the risk of falling when working at heights.

[0147] Compared with the prior art, this application has the following advantages:

[0148] 1. The introduction of the model avoids the subjectivity brought about by risk judgment based on experience and enhances the objectivity of the risk judgment results.

[0149] 2. Comprehensively considering the hidden danger risk identification results and posture risk identification results enhances the accuracy of the final risk identification results.

[0150] 3. Determine whether to initiate a risk warning based on the risk identification result, and determine the corresponding personnel status information based on the physiological signal data and operation data. Determine the corresponding safety warning information and response measures information based on the personnel status information. Combine the above safety warning information and response measures information to generate the corresponding user operation risk warning information, thereby ensuring the availability of the generated warning information, helping users to stay away from or deal with possible risks, and protecting the personal safety of users.

[0151] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0152] Based on the same inventive concept, the embodiment of the present application also provides a user operation risk warning device for implementing the user operation risk warning method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more user operation risk warning device embodiments provided below can refer to the limitations of the user operation risk warning method above, and will not be repeated here.

[0153] In an exemplary embodiment, as shown in FIG. 7 , a user operation risk warning device is provided, comprising: a data acquisition module 701, an evaluation result acquisition module 702, a discrimination result determination module 703 and a warning information determination module 704, wherein:

[0154] The data acquisition module 701 is used to acquire the physiological signal data of the user to be identified, and acquire the operation data of the user to be identified;

[0155] An assessment result acquisition module 702 is used to input the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and to input the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified;

[0156] The identification result determination module 703 is used to obtain the risk identification result of the user to be identified according to the hidden danger risk assessment result and the posture risk assessment result;

[0157] The warning information determination module 704 is used to obtain user operation risk warning information of the user to be identified based on the risk identification result, and return the user operation risk warning to the terminal of the user to be identified.

[0158] In one embodiment, the data acquisition module 701 further includes:

[0159] The acquisition submodule is used to acquire the original physiological signal data of the user to be identified; the original physiological signal data includes first original physiological signal data and second original physiological signal data; the first original physiological signal data is the disturbed original physiological signal data; the second physiological signal data is the undisturbed original physiological signal data.

[0160] The multiplication submodule is used to perform multiplication processing on the first original physiological signal data and the pre-constructed random matrix to obtain the transformed first original physiological signal data.

[0161] The physiological signal data acquisition submodule is used to obtain physiological signal data according to the transformed first original physiological signal data and the second original physiological signal data.

[0162] In one embodiment, the transformed first original physiological signal data includes third original physiological signal data, fourth original physiological signal data and fifth original physiological signal data; wherein the third original physiological signal data is original physiological signal data affected by electromechanical interference, the fourth original physiological signal data is original physiological signal data affected by motion artifacts, and the fifth original physiological signal data is original physiological signal data not affected by interference.

[0163] The physiological signal data acquisition submodule is also used to filter the third original physiological signal data using a preset band-stop filter to obtain the filtered third original physiological signal data, and to filter the fourth original physiological signal data using a preset low-pass filter to obtain the filtered fourth original physiological signal data; to add the second original physiological signal data, the filtered third original physiological signal data, the filtered fourth original physiological signal data, and the fifth original physiological signal data to obtain the physiological signal data to be processed; and to denoise and normalize the physiological signal data to be processed to obtain the physiological signal data.

[0164] In an exemplary embodiment, the working data includes working video data and acceleration data; the data acquisition module 701 is also used to downsample, normalize and denoise the working video data to obtain processed working video data, and to denoise and normalize the acceleration data to obtain processed acceleration data; obtain the corresponding working video vector based on the processed working video data, and obtain the corresponding acceleration feature vector based on the processed acceleration data; obtain the first weight corresponding to the working video feature vector, and obtain the second weight corresponding to the acceleration feature vector; based on the first weight and the second weight, perform weighted summation on the working video feature vector and the acceleration feature vector to obtain the operation data feature vector of the user to be identified.

[0165] The assessment result acquisition module 702 is also used to input the operation data feature vector into a pre-built posture risk assessment model to obtain a high-order feature vector corresponding to the operation data feature vector, and obtain the posture risk assessment result of the user to be identified based on the high-order feature vector; the high-order feature vector is used to distinguish different posture information of the user to be identified.

[0166] In one embodiment, the discrimination result determination module 703 is also used to obtain a third weight corresponding to the hidden danger risk assessment result, and to obtain a fourth weight corresponding to the posture risk assessment result; using the third weight and the fourth weight, the hidden danger risk assessment result and the posture risk assessment result are weighted to obtain a risk discrimination result of the user to be discriminated.

[0167] In one of the embodiments, the risk identification result is characterized by a risk value; the warning information determination module 704 is also used to determine the personnel status information of the user to be identified based on the physiological signal data and the operation data when the risk value is greater than a preset risk value threshold; determine the corresponding work ticket content and on-site operation risk information based on the personnel status information, and obtain the corresponding safety warning information and response measures information according to the work ticket content and the on-site operation risk information; generate user operation risk warning information according to the safety warning information and response measures information.

[0168] Each module in the above user operation risk warning device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0169] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG8. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store physiological signal data and operation data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a user operation risk warning method is implemented.

[0170] Technical personnel can understand that the structure shown in Figure 8 is only a block diagram of a partial structure related to the present application scheme, and does not constitute a limitation on the computer device to which the present application scheme is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0171] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the user operation risk warning method of the above embodiment is implemented.

[0172] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the user operation risk warning method of the above embodiment is implemented.

[0173] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the user operation risk warning method of the above embodiment.

[0174] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0175] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0176] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0177] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A user operation risk early warning method, characterized in that: The method comprises: Acquiring physiological signal data of a user to be identified, and acquiring operation data of the user to be identified; Inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified; Obtaining a risk identification result of the user to be identified according to the hidden danger risk assessment result and the posture risk assessment result; Based on the risk identification result, the user operation risk warning information of the user to be identified is obtained, and the user operation risk warning is returned to the terminal of the user to be identified.

2. The method according to claim 1, characterized in that: The step of obtaining physiological signal data of the user to be identified includes: Acquire original physiological signal data of the user to be identified; the original physiological signal data includes first original physiological signal data and second original physiological signal data; the first original physiological signal data is disturbed original physiological signal data; the second physiological signal data is undisturbed original physiological signal data; Multiplying the first original physiological signal data and a pre-constructed random matrix to obtain transformed first original physiological signal data; The physiological signal data is obtained according to the transformed first original physiological signal data and the second original physiological signal data.

3. The method according to claim 2, characterized in that The transformed first original physiological signal data includes third original physiological signal data, fourth original physiological signal data and fifth original physiological signal data; wherein the third original physiological signal data is original physiological signal data subject to electromechanical interference, the fourth original physiological signal data is original physiological signal data subject to motion artifact interference, and the fifth original physiological signal data is original physiological signal data free from interference; The step of obtaining the physiological signal data according to the transformed first original physiological signal data and the second original physiological signal data includes: Using a preset band-stop filter to filter the third original physiological signal data to obtain filtered third original physiological signal data, and using a preset low-pass filter to filter the fourth original physiological signal data to obtain filtered fourth original physiological signal data; performing sum processing on the second original physiological signal data, the filtered third original physiological signal data, the filtered fourth original physiological signal data, and the fifth original physiological signal data to obtain physiological signal data to be processed; The physiological signal data to be processed is subjected to denoising and normalization processing to obtain the physiological signal data.

4. The method according to claim 1, characterized in that: The working data includes working video data and acceleration data; The step of obtaining the operation data of the user to be identified includes: Performing downsampling, normalization and denoising processing on the working video data to obtain processed working video data, and performing denoising and normalization processing on the acceleration data to obtain processed acceleration data; Acquire a corresponding working video vector based on the processed working video data, and acquire a corresponding acceleration feature vector based on the processed acceleration data; Obtaining a first weight corresponding to the working video feature vector, and obtaining a second weight corresponding to the acceleration feature vector; Based on the first weight and the second weight, weighted summing is performed on the work video feature vector and the acceleration feature vector to obtain the work data feature vector of the user to be identified; Inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified, including: The operation data feature vector is input into a pre-constructed posture risk assessment model to obtain a high-order feature vector corresponding to the operation data feature vector, and the posture risk assessment result of the user to be identified is obtained based on the high-order feature vector; the high-order feature vector is used to distinguish different posture information of the user to be identified.

5. The method according to claim 1, characterized in that The step of obtaining a risk identification result of the user to be identified based on the hidden danger risk assessment result and the posture risk assessment result includes: Obtaining a third weight corresponding to the hidden danger risk assessment result, and obtaining a fourth weight corresponding to the posture risk assessment result; The hidden danger risk assessment result and the posture risk assessment result are weighted by using the third weight and the fourth weight to obtain a risk identification result of the user to be identified.

6. The method according to claim 1, characterized in that The risk identification result is characterized by a risk value; The step of obtaining risk warning information of the user to be identified based on the risk identification result includes: In the case where the risk value is greater than a preset risk value threshold, determining the personnel status information of the user to be identified according to the physiological signal data and the operation data; Determine the corresponding work ticket content and on-site operation risk information based on the personnel status information, and obtain corresponding safety warning information and response measures information according to the work ticket content and the on-site operation risk information; User operation risk warning information is generated according to the safety warning information and the countermeasure information.

7. A user operation risk warning device, characterized in that: The device comprises: A data acquisition module, used to acquire physiological signal data of a user to be identified, and to acquire operation data of the user to be identified; An assessment result acquisition module, used for inputting the physiological signal data into a pre-built hidden danger risk assessment model to obtain a hidden danger risk assessment result of the user to be identified, and inputting the operation data into a pre-built posture risk assessment model to obtain a posture risk assessment result of the user to be identified; A discrimination result determination module, used for obtaining a risk discrimination result of the user to be discriminated according to the hidden danger risk assessment result and the posture risk assessment result; The warning information determination module is used to obtain the user operation risk warning information of the user to be identified based on the risk identification result, and return the user operation risk warning to the terminal of the user to be identified.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to 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 method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images