Limited space operation monitoring method and system
By acquiring and processing the operational data and infrared image data of the workers, generating target infrared image data and conducting safety assessments, the problem of difficulty in making accurate early warnings in confined spaces in existing technologies is solved, thus improving operational safety.
Patent Information
- Application Number
- CN202511096595.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-06
- Publication Date
- 2025-11-07
AI Technical Summary
Existing technologies mainly rely on fixed sensors to monitor workers in confined spaces, but they primarily monitor environmental data and are difficult to provide accurate early warnings before accidents occur, thus reducing the safety of confined space operations.
By acquiring the operational data and infrared image data of the workers, the operational data is preprocessed, target separation is performed, target infrared image data is generated, and a safety assessment is conducted in conjunction with the target operational data to generate operational monitoring results.
It improves the safety of confined space operations, ensures the accuracy of operational safety assessments, and enables precise early warnings before accidents occur.
Smart Images
Figure CN120912926A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of limited space monitoring, and in particular to a limited space operation monitoring method and system. BACKGROUND
[0002] In the power industry, especially in limited space operation environments (such as underground cable wells), the safety risk is high, and the consequences of accidents are serious. Therefore, how to monitor the operating personnel in the limited space is crucial.
[0003] At present, the prior art mainly relies on fixed sensors to monitor the operating personnel in the limited space, but mainly monitors environmental data. Due to the complexity of the limited space operation environment, it is difficult to accurately warn before an accident occurs, which reduces the safety of limited space operation. SUMMARY
[0004] The present application provides a limited space operation monitoring method and system, which solves the technical problem that the prior art mainly relies on fixed sensors to monitor the operating personnel in the limited space, but mainly monitors environmental data. Due to the complexity of the limited space operation environment, it is difficult to accurately warn before an accident occurs, which reduces the safety of limited space operation.
[0005] The first aspect of the present application provides a limited space operation monitoring method, comprising:
[0006] Obtaining operation data and infrared image data of operating personnel, data preprocessing the operation data to obtain corresponding target operation data;
[0007] Target separation processing the infrared image data to obtain corresponding target infrared image data;
[0008] Performing operation safety evaluation according to the target infrared image data and the target operation data to generate a corresponding operation monitoring result.
[0009] Optionally, the step of data preprocessing the operation data to obtain corresponding target operation data comprises:
[0010] Performing missing data elimination processing on the operation data to obtain filtered data;
[0011] Performing standardization processing on the filtered data to obtain target operation data.
[0012] Optionally, the step of target separation processing the infrared image data to obtain corresponding target infrared image data comprises:
[0013] Performing linear image enhancement processing on the infrared image data to obtain corresponding to-be-separated image data;
[0014] randomly selecting two pixel points from the image data to be separated as class center points;
[0015] inputting each of the class center points and the image data to be separated into a preset target separation function to obtain a target separation value;
[0016] updating each of the class center points according to the target separation value to obtain a plurality of updated class center points;
[0017] jumping to execute the step of inputting each of the class center points and the image data to be separated into a preset target separation function to obtain a target separation value until a preset iteration condition is met, and generating corresponding target infrared image data.
[0018] Optionally, the step of performing linear image enhancement processing on the infrared image data to obtain corresponding image data to be separated includes:
[0019] inputting the infrared image data into a preset linear enhancement function to obtain enhanced image data;
[0020] when the brightness in the enhanced image data is less than a preset brightness threshold, increasing the brightness of the enhanced image data to obtain corresponding image data to be separated;
[0021] when the brightness in the enhanced image data is greater than or equal to the preset brightness threshold, reducing the brightness of the enhanced image data to obtain corresponding image data to be separated.
[0022] Optionally, the step of performing operation safety evaluation according to the target infrared image data and the target operation data to generate corresponding operation monitoring results includes:
[0023] calculating the similarity between the target infrared image data and a plurality of standard infrared image data obtained in advance by a preset similarity function to obtain a plurality of similarity values;
[0024] determining whether each of the similarity values is greater than a preset similarity threshold;
[0025] if any of the similarity values is greater than the similarity threshold, determining that an operation anomaly is the corresponding operation monitoring result;
[0026] if each of the similarity values is less than or equal to the similarity threshold, inputting the target operation data into a pre-trained safety warning decision model to obtain a corresponding risk value, wherein the target operation data includes environmental data and vital sign data;
[0027] When the risk value is greater than a preset risk threshold value, the job abnormality is determined as the corresponding job monitoring result.
[0028] When the risk value is less than or equal to the risk threshold value, the job normality is determined as the corresponding job monitoring result.
[0029] Optionally, the training process of the safety warning decision model specifically comprises:
[0030] A plurality of historical job data are acquired, and each historical job data is subjected to data preprocessing to generate a job feature set;
[0031] The job feature set is input into a preset initial safety warning decision model for training to output training job data, wherein the initial safety warning decision model is a BP neural network;
[0032] A training loss value of the training job feature set is calculated according to the training job data;
[0033] When the training loss value is greater than or equal to a preset error threshold value, a first model parameter and a second model parameter of the initial safety warning decision model are adjusted in sequence, and the step of inputting the job feature set into the preset initial safety warning decision model for training to output training job data is executed until the training loss value is less than the error threshold value;
[0034] When the training loss value is less than the error threshold value, a safety warning decision model is generated.
[0035] The second aspect of the present application provides a limited space job monitoring system, comprising:
[0036] A collection module is configured to acquire job data and infrared image data of a job worker, and to perform data preprocessing on the job data to obtain corresponding target job data;
[0037] A separation module is configured to perform target separation processing on the infrared image data to obtain corresponding target infrared image data;
[0038] A warning module is configured to perform job safety evaluation according to the target infrared image data and the target job data to generate a corresponding job monitoring result.
[0039] The third aspect of the present application provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the limited space job monitoring method according to any one of the above aspects.
[0040] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed to implement the limited space operation monitoring method according to any one of the preceding aspects.
[0041] The fifth aspect of the present application provides a computer program product, which comprises a computer program stored on a non-transitory computer readable storage medium, and the computer program comprises program instructions, wherein when the program instructions are executed by a computer, the computer executes the limited space operation monitoring method according to any one of the preceding aspects.
[0042] From the above technical solutions, the present application has the following advantages:
[0043] The present application obtains the operation data and the infrared image data of the operation personnel, pre-processes the operation data to obtain corresponding target operation data, performs target separation processing on the infrared image data to obtain corresponding target infrared image data, performs operation safety evaluation according to the target infrared image data and the target operation data, and generates corresponding operation monitoring results. The present application overcomes the technical problem that the prior art mainly relies on fixed sensors to monitor the operation personnel in the limited space, but mainly monitors environmental data, and it is difficult to accurately warn before an accident occurs due to the complex limited space operation environment, thereby reducing the safety of the limited space operation. Compared with the traditional limited space operation monitoring method, the present application performs target separation processing on the infrared image data to obtain target infrared image data with higher data quality, and then performs operation safety evaluation by fusing the target infrared image data and the target operation data to generate corresponding operation monitoring results, thereby ensuring the accuracy of the operation safety evaluation and improving the safety of the limited space operation. BRIEF DESCRIPTION OF DRAWINGS
[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0045] Figure 1 A step flow chart of a limited space operation monitoring method provided for the first embodiment of the present application is shown in the figure.
[0046] Figure 2 A step flow chart of a limited space operation monitoring method provided for the second embodiment of the present application is shown in the figure.
[0047] Figure 3 A structure block diagram of a limited space operation monitoring system provided for the third embodiment of the present application is shown in the figure.
[0048] Figure 4 A structural block diagram of an electronic device provided for the fourth embodiment of the present application. DETAILED DESCRIPTION
[0049] The embodiment of the present application provides a limited space operation monitoring method and system, which is used for solving the technical problem that the prior art mainly depends on fixed sensors to monitor the operation personnel in the limited space, but mainly monitors environmental data, and it is difficult to accurately warn before an accident occurs due to the complex limited space operation environment, and the safety of the limited space operation is reduced.
[0050] In order to make the invention purpose, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the following described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.
[0051] Please refer to Figure 1 , Figure 1 A step flow chart of a limited space operation monitoring method provided for the first embodiment of the present application.
[0052] The limited space operation monitoring method provided by the present application comprises:
[0053] In step 101, operation data and infrared image data of the operation personnel are acquired, the operation data is preprocessed, and corresponding target operation data is obtained.
[0054] The operation data refers to positioning data, environmental data and human body sign data of the operation personnel in the limited space, wherein the positioning data refers to real-time position information of the operation personnel and equipment in a three-dimensional space. The environmental data refers to a set of quantitative parameters reflecting the physical / chemical state inside the operation space, which are collected by a sensor network in real time. The sign data refers to a set of biomedical parameters reflecting the physiological state of the operation personnel, which are acquired by wearable devices or non-contact sensing technology in real time.
[0055] The infrared image data refers to visualized data reflecting temperature distribution in the limited space, which are collected by an infrared image detection device (such as an infrared thermal imager) in real time. The function is to identify targets (such as operation personnel and equipment) and backgrounds through thermal radiation differences.
[0056] The target operation data refers to the operation data after data preprocessing.
[0057] In the embodiment of the present application, the working data and infrared image data of the workers are collected in real time by the plurality of Internet of Things devices arranged in the limited space, and the working data is filtered and labeled in sequence to obtain the corresponding target working data.
[0058] It should be noted that a plurality of monitoring points are arranged in the limited space, and an Internet of Things device is installed on each monitoring point. The Internet of Things device includes an infrared image detection device, a positioning sensor, an environmental online monitoring device, and a human body sign detection device. The environmental online monitoring device is used to collect the physical / chemical parameters (such as oxygen concentration, toxic gas (such as H2S), flammable gas concentration, and dust content) of the working environment in real time. The human body sign detection device is a wearable Internet of Things sensing device arranged on the workers in the limited space, which is used to collect the physiological parameters (such as heart rate, blood oxygen saturation, body temperature, acceleration (fall detection), attitude angle, skin surface humidity, and breathing rate) of the workers in real time.
[0059] It is worth mentioning that in the positioning data collection process, a preset reference number of monitoring points are randomly selected from the plurality of monitoring points as reference points, the positioning sensor signals around each reference point are collected in real time by a wireless positioning method, and the collection range of the wireless positioning is determined based on the received positioning sensor signals. After receiving the positioning sensor signals, the wireless positioning device generates a set of triplet data composed of the position coordinates of the positioning sensor, the positioning sensor signals, and the device number of the positioning sensor as the positioning data.
[0060] Step 102, the infrared image data is subjected to target separation processing to obtain corresponding target infrared image data.
[0061] The target infrared image data refers to the target (such as workers, equipment) in the infrared image data.
[0062] In the embodiment of the present application, the infrared image data is subjected to linear image enhancement processing to obtain corresponding to-be-separated image data. Two pixel points are randomly selected from the to-be-separated image data as class center points. Each class center point and the to-be-separated image data are input into a preset target separation function to obtain a target separation value. Each class center point is updated according to the target separation value to obtain a plurality of updated class center points. The step of inputting each class center point and the to-be-separated image data into the preset target separation function to obtain the target separation value is executed repeatedly until a preset iteration condition is met, and the corresponding target infrared image data is generated.
[0063] Step 103, performing working safety evaluation according to the target infrared image data and the target working data to generate corresponding working monitoring results.
[0064] In the embodiment of the present application, the similarity between the target infrared image data and the plurality of standard infrared image data obtained in advance is calculated by a preset similarity function, and a plurality of similarity values are obtained. If any similarity value is greater than a preset similarity threshold, the work abnormality is determined as the corresponding work monitoring result. If each similarity value is less than or equal to the similarity threshold, the target work data is input into the pre-trained safety warning decision model, and a corresponding risk value is obtained, wherein the target work data includes environmental data and vital sign data. When the risk value is greater than a preset risk threshold, the work abnormality is determined as the corresponding work monitoring result. When the risk value is less than or equal to the risk threshold, the work is normal, and the corresponding work monitoring result is determined.
[0065] In the embodiment of the present application, by obtaining the work data and the infrared image data of the work personnel, the work data is pre-processed to obtain the corresponding target work data, the infrared image data is target-separated to obtain the corresponding target infrared image data, and the work safety is evaluated according to the target infrared image data and the target work data to generate the corresponding work monitoring result. Overcome the technical problem that the prior art mainly relies on fixed sensors to monitor the work personnel in a limited space, but mainly monitors environmental data, and due to the complex limited space work environment, it is difficult to accurately warn before an accident occurs, and the safety of the limited space work is reduced. Compared with the traditional limited space work monitoring method, the present application separates the infrared image data to obtain target infrared image data with higher data quality, and then evaluates the work safety by fusing the target infrared image data and the target work data to generate the corresponding work monitoring result, ensuring the accuracy of the work safety evaluation and improving the safety of the limited space work.
[0066] Please refer to Figure 2 , Figure 2 The step flow chart of a limited space work monitoring method provided in the second embodiment of the present application.
[0067] The present application provides a limited space work monitoring method, comprising:
[0068] Step 201, obtaining work data and infrared image data of work personnel, and pre-processing the work data to obtain corresponding target work data.
[0069] Further, step 201 comprises the following sub-steps:
[0070] S11, the missing data of the work data is removed to obtain filtered data.
[0071] The filtered data refers to the work data after the missing data removal processing.
[0072] In the embodiment of the present application, the missing data of the work data is removed to obtain filtered data. For example, the work data includes a plurality of positioning data, a plurality of environmental data and a plurality of physical data. Whether each positioning data, each environmental data and each physical data is complete is judged based on a preset data format (the data format is: data number, data content and data time). If not, the corresponding data is removed to obtain the filtered data.
[0073] S12, the filtered data is standardized to obtain target work data.
[0074] In the embodiment of the present application, the filtered data is input into a preset standardization function to obtain corresponding target work data.
[0075] It should be noted that the standardization function is specifically:
[0076]
[0077] Wherein, The z-th set of standardized positioning data, environmental data and physical data, The z-th set of positioning data, environmental data and physical data, The mean of the positioning data, environmental data and physical data, The variance of the positioning data, environmental data and physical data, and z is the index of the work data.
[0078] Step 202, the infrared image data is linearly enhanced to obtain corresponding image data to be separated.
[0079] Further, step 202 includes the following sub-steps:
[0080] S21, the infrared image data is input into a preset linear enhancement function to obtain enhanced image data.
[0081] Enhanced image data refers to linearly enhanced infrared image data.
[0082] In the embodiment of the present application, the infrared image data is linearly enhanced by the preset linear enhancement function to obtain the enhanced image data.
[0083] It should be noted that the linear enhancement function is specifically:
[0084]
[0085]
[0086] Wherein, The pixel value of the local pixel block in the contrast-enhanced infrared image data, contrast gain parameter, pixel value of a local pixel block in infrared image data before contrast enhancement, brightness mean value of a local pixel block in infrared image data, abscissa of a local pixel block, ordinate of a local pixel block.
[0087] S22, when the brightness in the enhanced image data is less than the preset brightness threshold value, the brightness of the enhanced image data is increased to obtain corresponding to-be-separated image data.
[0088] to-be-separated image data refers to the enhanced image data after brightness adjustment.
[0089] brightness threshold value refers to a critical brightness value used for judging whether a pixel block needs to be enhanced in the infrared image processing process.
[0090] In the embodiment of the present application, whether the brightness in the enhanced image data is less than the preset brightness threshold value is judged. The output brightness of the enhanced image data is increased to obtain corresponding to-be-separated image data.
[0091] S23, when the brightness in the enhanced image data is greater than or equal to the preset brightness threshold value, the brightness of the enhanced image data is decreased to obtain corresponding to-be-separated image data.
[0092] In the embodiment of the present application, if the brightness in the enhanced image data is greater than or equal to the preset brightness threshold value, the output brightness of the enhanced image data is decreased to obtain corresponding to-be-separated image data.
[0093] Step 203, two pixel points in the to-be-separated image data are randomly selected as class center points.
[0094] class center point refers to a center point of a class in the to-be-separated image data.
[0095] In the embodiment of the present application, one pixel point in the to-be-separated image data is randomly selected as a class center point of a target, and another pixel point in the to-be-separated image data is randomly selected as a class center point of a background.
[0096] Step 204, each class center point and the to-be-separated image data are input into a preset target separation function to obtain a target separation value.
[0097] target separation value refers to a membership value calculated by a fuzzy C-means clustering (FCM) algorithm.
[0098] In the embodiment of the present application, each class center point and the to-be-separated image data are calculated by the preset target separation function to obtain corresponding target separation values.
[0099] It should be noted that the target separation function is specifically:
[0100]
[0101]
[0102] wherein, is a target separation value, is the jth pixel point, is the ith class center point, is the pixel point j belonging to the ith class, is the total number of pixel points, a is the center point, i is the index of the class, j is the index of the pixel point, m is the fuzzy index, the fuzzy index is the degree of blur in the captured image, and A is the total number of class center points.
[0103] Step 205, updating each class center point according to the target separation value to obtain a plurality of updated class center points.
[0104] In the embodiment of the present application, the class center points of the target and the class center points of the background are updated according to the target separation value to obtain the updated class center points of the target and the class center points of the background.
[0105] Step 206, jump to execute the step of inputting each class center point and the to-be-separated image data into the preset target separation function to obtain the target separation value until the preset iteration condition is met, and generate corresponding target infrared image data.
[0106] In the embodiment of the present application, steps 204-206 are executed in a jump manner until the pixel points of the target and the pixel points of the background no longer change, and the pixel points of the target are used to generate corresponding target infrared image data.
[0107] Step 207, performing work safety evaluation according to the target infrared image data and the target work data to generate corresponding work monitoring results.
[0108] Further, step 207 includes the following sub-steps:
[0109] S31, calculating the similarity between the target infrared image data and the plurality of standard infrared image data pre-acquired by the preset similarity function respectively to obtain a plurality of similarity values.
[0110] The standard infrared image data refers to the infrared image data with dangerous behavior.
[0111] In the embodiment of the present application, the target infrared image data and the plurality of standard infrared image data pre-acquired are input into the preset similarity function respectively to obtain a plurality of similarity values.
[0112] It should be noted that the similarity function is specifically:
[0113]
[0114] wherein, is a similarity value, is the area of the overlapping part of the target infrared image data and the standard infrared image data, is the area of the target infrared image data, is the area of the standard infrared image data, is the target infrared image data, is the standard infrared image data.
[0115] S32, judge whether each similarity value is greater than a preset similarity threshold value.
[0116] The similarity threshold value refers to a critical value for judging the matching degree of the worker's infrared image behavior and the pre-stored dangerous behavior image.
[0117] S33, if any similarity value is greater than the similarity threshold value, the work abnormality is determined as the corresponding work monitoring result.
[0118] In the embodiment of the application, it is judged whether each similarity value is greater than a preset similarity threshold value, and when any similarity value is greater than the similarity threshold value, it indicates that the worker has dangerous operation, the work abnormality is determined as the corresponding work monitoring result, and the positioning data of the worker and the work monitoring result are uploaded to the terminal for early warning.
[0119] S34, if each similarity value is less than or equal to the similarity threshold value, the target work data is input into a pre-trained safety warning decision model to obtain a corresponding risk value, wherein the target work data includes environmental data and physical sign data.
[0120] In the embodiment of the application, when each similarity value is less than or equal to the similarity threshold value, the target work data is input into a pre-trained safety warning decision model to obtain a corresponding risk value, wherein the target work data includes environmental data and physical sign data.
[0121] It should be noted that the training process of the safety warning decision model is specifically:
[0122] A1, obtain a plurality of historical work data, and perform data preprocessing on each historical work data to generate a work feature set.
[0123] The historical work data refers to historical environmental data and historical physical sign data.
[0124] It should be noted that the data preprocessing refers to the cleaning and smoothing operation on the data, so as to generate the historical environment data and the historical sign data for model training, and to label the historical environment data and the historical sign data, so as to obtain the job feature set composed of a plurality of labeled historical environment data and historical sign data.
[0125] In the embodiment of the application, a plurality of historical environment data and a plurality of historical sign data are obtained, and each historical environment data and each historical sign data are subjected to data preprocessing to generate a job feature set.
[0126] A2, the job feature set is input into a preset initial safety warning decision model for training, and training job data is output, wherein the initial safety warning decision model is a BP neural network.
[0127] In the embodiment of the application, the job feature set is input into a preset initial safety warning decision model for training, and corresponding training job data is obtained.
[0128] It should be noted that the BP neural network (i.e. back propagation neural network) is composed of an input layer, a hidden layer and an output layer. In the training process, the input of the input layer is , wherein, is the job feature set, is the first set of historical environment data and historical sign data after being combined, is the second set of historical environment data and historical sign data after being combined, is the gth set of historical environment data and historical sign data after being combined, and g is the index of the historical environment data and the historical sign data. The output of the output layer is , wherein, is the training job data, is the first set of predicted risk values, is the second set of predicted risk values, is the s th set of predicted risk values, and s is the index of the risk value in the training process.
[0129] A3, the training loss value of the training job feature set is calculated according to the training job data.
[0130] In the embodiment of the application, the difference between each predicted risk value in the training job data and the corresponding expected value in the training job feature set is calculated to obtain a plurality of first differences. The absolute values of each first difference are processed by mean value to obtain the corresponding training loss value.
[0131] A4、when the training loss value is greater than or equal to the preset error threshold value, then the first model parameter and the second model parameter of the initial safety warning decision model are adjusted in sequence, and the step of inputting the preset initial safety warning decision model with the job feature set to perform training and outputting the training job data is executed until the training loss value is less than the error threshold value.
[0132] The first model parameter refers to the weight of the input layer to the hidden layer in the initial safety warning decision model.
[0133] The second model parameter refers to the weight of the hidden layer to the output layer in the initial safety warning decision model.
[0134] In the embodiment of the present application, when the training loss value is greater than or equal to the preset error threshold value, then the weight of the input layer to the hidden layer and the weight of the hidden layer to the output layer in the initial safety warning decision model are adjusted in sequence, and the steps of A2-A5 are executed until the training loss value is less than the error threshold value.
[0135] A5、when the training loss value is less than the error threshold value, then the safety warning decision model is generated.
[0136] The error threshold value refers to a critical value for judging whether the model output reaches the expected accuracy in the BP neural network (back propagation neural network) training process.
[0137] In the embodiment of the present application, when the training loss value is less than the error threshold value, then the trained BP neural network is used as the safety warning decision model.
[0138] S35、when the risk value is greater than the preset risk threshold value, then the job abnormality is determined as the corresponding job monitoring result.
[0139] The risk threshold value refers to a critical numerical value for judging whether the environmental data or the personnel behavior reaches a dangerous level.
[0140] In the embodiment of the present application, when the risk value is greater than the preset risk threshold value, then it indicates that the current job personnel has a job risk, the job abnormality is determined as the corresponding job monitoring result, and the positioning data of the job personnel and the job monitoring result are uploaded to the terminal for early warning.
[0141] S36、when the risk value is less than or equal to the risk threshold value, then the job normality is determined as the corresponding job monitoring result.
[0142] In the embodiment of the present application, when the risk value is less than or equal to the risk threshold value, then it indicates that the job personnel operates normally and has no job risk, and the job normality is determined as the corresponding job monitoring result.
[0143] In the embodiment of the present application, by acquiring the work data and infrared image data of the work personnel, the work data is pre-processed to obtain corresponding target work data, the infrared image data is target separation processed to obtain corresponding target infrared image data, the work safety is evaluated according to the target infrared image data and the target work data, and the corresponding work monitoring result is generated. Overcome the technical problem that the prior art mainly relies on fixed sensors to monitor the work personnel in the limited space, but mainly monitors the environmental data, and due to the complex limited space working environment, it is difficult to accurately warn before the accident occurs, and the safety of the limited space work is reduced. Compared with the traditional limited space work monitoring method, the present application processes the infrared image data by target separation to obtain target infrared image data with higher data quality, and then evaluates the work safety by fusing the target infrared image data and the target work data to generate the corresponding work monitoring result, thereby ensuring the accuracy of the work safety evaluation and improving the safety of the limited space work.
[0144] Please refer to Figure 3 , Figure 3 The structure block diagram of a limited space work monitoring system provided in the third embodiment of the present application is shown.
[0145] The limited space work monitoring system provided by the present application comprises:
[0146] The acquisition module 301 is configured to acquire the work data and infrared image data of the work personnel, and pre-process the work data to obtain corresponding target work data.
[0147] The separation module 302 is configured to perform target separation processing on the infrared image data to obtain corresponding target infrared image data.
[0148] The warning module 303 is configured to evaluate the work safety according to the target infrared image data and the target work data, and generate a corresponding work monitoring result.
[0149] Further, the acquisition module 301 comprises:
[0150] The filtering submodule is configured to perform missing data elimination processing on the work data to obtain filtered data.
[0151] The standardization submodule is configured to perform standardization processing on the filtered data to obtain target work data.
[0152] Further, the separation module 302 comprises:
[0153] The enhancement submodule is configured to perform linear image enhancement processing on the infrared image data to obtain corresponding to-be-separated image data.
[0154] The first analysis submodule is configured to randomly select two pixel points as class center points from the image data to be separated;
[0155] The class center points and the image data to be separated are input into a preset target separation function to obtain target separation values;
[0156] The class center points are updated according to the target separation values to obtain a plurality of updated class center points;
[0157] The step of inputting the class center points and the image data to be separated into the preset target separation function to obtain the target separation values is executed until a preset iteration condition is met, and the corresponding target infrared image data is generated.
[0158] Further, the enhancement submodule includes:
[0159] The linear enhancement unit is configured to input the infrared image data into a preset linear enhancement function to obtain enhanced image data;
[0160] The brightness adjustment unit is configured to increase the brightness of the enhanced image data when the brightness in the enhanced image data is less than a preset brightness threshold to obtain corresponding image data to be separated;
[0161] The brightness adjustment unit is configured to decrease the brightness of the enhanced image data when the brightness in the enhanced image data is greater than or equal to the preset brightness threshold to obtain corresponding image data to be separated.
[0162] Further, the early warning module 303 includes:
[0163] The first early warning submodule is configured to calculate the similarity between the target infrared image data and a plurality of standard infrared image data pre-acquired by using a preset similarity function to obtain a plurality of similarity values;
[0164] It is determined whether each similarity value is greater than a preset similarity threshold;
[0165] If any similarity value is greater than the similarity threshold, the work abnormality is determined as the corresponding work monitoring result;
[0166] The second early warning submodule is configured to input the target work data into a safety early warning decision model pre-trained if each similarity value is less than or equal to the similarity threshold to obtain a corresponding risk value, wherein the target work data includes environmental data and vital sign data;
[0167] When the risk value is greater than a preset risk threshold, the work abnormality is determined as the corresponding work monitoring result;
[0168] When the risk value is less than or equal to the risk threshold, the work is normal and is determined as the corresponding work monitoring result.
[0169] Further, the training process of the safety warning decision model is specifically:
[0170] Obtaining a plurality of historical operation data, performing data preprocessing on each historical operation data, and generating an operation feature set;
[0171] The operation feature set is input into a preset initial safety warning decision model for training, and training operation data is output, wherein the initial safety warning decision model is a BP neural network;
[0172] According to the training operation data, a training loss value of the training operation feature set is calculated;
[0173] When the training loss value is greater than or equal to a preset error threshold value, the first model parameter and the second model parameter of the initial safety warning decision model are adjusted in turn, and the step of inputting the operation feature set into the preset initial safety warning decision model for training and outputting the training operation data is executed, until the training loss value is less than the error threshold value;
[0174] When the training loss value is less than the error threshold value, a safety warning decision model is generated.
[0175] Please refer to Figure 4 , Figure 4 A structural block diagram of an electronic device provided for the fourth embodiment of the present application.
[0176] An electronic device of an embodiment of the present application, the electronic device comprising: a memory 401 and a processor 402, the memory 401 storing a computer program; when the computer program is executed by the processor 402, the processor 402 executes the limited space operation monitoring method of any one of the above embodiments.
[0177] The memory 401 can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. The memory 401 has a storage space 403 for program codes 413 for performing any of the method steps in the above-described methods. For example, the storage space 403 for program codes can include individual program codes 413 for implementing various steps in the above-described methods, respectively. These program codes can be read from or written to one or more computer program products. These computer program products include program code carriers such as a hard disk, a compact disc (CD), a memory card, or a floppy disk. The program codes can be compressed in a suitable form, for example. These codes, when run by a computing processing device, cause the computing processing device to perform the individual steps in the above-described methods. These program codes can be read from or written to one or more computer program products. These computer program products include program code carriers such as a hard disk, a compact disc (CD), a memory card, or a floppy disk. The program codes can be compressed in a suitable form, for example. These codes, when run by a computing processing device, cause the computing processing device to perform the individual steps in the above-described limited space operation monitoring method.
[0178] The embodiment five of the present application further provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the limited space operation monitoring method according to any one of the above-described embodiments.
[0179] The embodiment six of the present application further provides a computer program product, comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions that, when executed by a computer, cause the computer to perform the limited space operation monitoring method according to any one of the above-described embodiments.
[0180] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described herein again.
[0181] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the division of the units is only a logical function division, and there can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or in other forms.
[0182] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.
[0183] In addition, each functional unit in the various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software functional unit.
[0184] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such an understanding, the technical solutions of the present application essentially, or the part that contributes to the prior art, or all or a part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, and various media that can store program codes.
[0185] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A method of monitoring work in a confined space, characterized by, The method comprises the following steps: acquiring work data and infrared image data of a worker, performing data preprocessing on the work data to obtain corresponding target work data; performing target separation processing on the infrared image data to obtain corresponding target infrared image data; performing work safety evaluation according to the target infrared image data and the target work data to generate corresponding work monitoring results.
2. The method of monitoring of confined space operations as claimed in claim 1, wherein, The step of performing data preprocessing on the work data to obtain corresponding target work data comprises the following steps: performing missing data elimination processing on the work data to obtain filtered data; performing standardization processing on the filtered data to obtain target work data.
3. The method of monitoring of confined space operations as claimed in claim 1 wherein, The step of performing target separation processing on the infrared image data to obtain corresponding target infrared image data comprises the following steps: performing linear image enhancement processing on the infrared image data to obtain corresponding to-be-separated image data; randomly selecting two pixel points from the to-be-separated image data as class center points; inputting each class center point and the to-be-separated image data into a preset target separation function to obtain a target separation value; updating each class center point according to the target separation value to obtain multiple updated class center points; jumping to execute the step of inputting each class center point and the to-be-separated image data into a preset target separation function to obtain a target separation value until a preset iteration condition is met to generate corresponding target infrared image data.
4. The method of monitoring of confined space operations as claimed in claim 3 wherein, The step of performing linear image enhancement processing on the infrared image data to obtain corresponding to-be-separated image data comprises the following steps: inputting the infrared image data into a preset linear enhancement function to obtain enhanced image data; when the brightness in the enhanced image data is less than a preset brightness threshold, increasing the brightness of the enhanced image data to obtain corresponding to-be-separated image data; when the brightness in the enhanced image data is greater than or equal to the preset brightness threshold, reducing the brightness of the enhanced image data to obtain corresponding to-be-separated image data.
5. The method of confined space work monitoring according to claim 1, wherein, The step of performing work safety evaluation according to the target infrared image data and the target work data to generate corresponding work monitoring results comprises the following steps: calculating the similarity between the target infrared image data and multiple standard infrared image data pre-acquired through a preset similarity function to obtain multiple similarity values; determining whether each similarity value is greater than a preset similarity threshold; if any similarity value is greater than the similarity threshold, determining work abnormality as the corresponding work monitoring result; if each similarity value is less than or equal to the similarity threshold, inputting the target work data, which comprises environmental data and vital sign data, into a pre-trained safety warning decision model to obtain a corresponding risk value; when the risk value is greater than a preset risk threshold, determining work abnormality as the corresponding work monitoring result; when the risk value is less than or equal to the risk threshold, determining work normality as the corresponding work monitoring result.
6. The method of confined space work monitoring according to claim 5, wherein, The training process of the safety warning decision model is specifically as follows: Obtaining a plurality of historical operation data, performing data preprocessing on each of the historical operation data to generate an operation feature set; The operation feature set is input into a preset initial safety warning decision model for training, and training operation data is output, wherein the initial safety warning decision model is a BP neural network; According to the training operation data, a training loss value of the training operation feature set is calculated; When the training loss value is greater than or equal to a preset error threshold, the first model parameter and the second model parameter of the initial safety warning decision model are adjusted in sequence, and the step of inputting the operation feature set into the preset initial safety warning decision model for training and outputting the training operation data is executed until the training loss value is less than the error threshold; When the training loss value is less than the error threshold, a safety warning decision model is generated.
7. A confined space work monitoring system characterized by, Comprising: The acquisition module is used for acquiring operation data and infrared image data of an operator, performing data preprocessing on the operation data to obtain corresponding target operation data; The separation module is used for performing target separation processing on the infrared image data to obtain corresponding target infrared image data; The warning module is used for performing operation safety evaluation according to the target infrared image data and the target operation data to generate corresponding operation monitoring results.
8. An electronic device, comprising: The computer program is executed by the processor to make the processor execute the steps of the limited space operation monitoring method according to any one of claims 1-6.
9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed to realize the limited space operation monitoring method according to any one of claims 1-6.
10. A computer program product, characterised in that, The computer program product comprises a computer program stored on a non-transitory computer readable storage medium, and the computer program comprises program instructions, wherein when the program instructions are executed by a computer, the computer executes the limited space operation monitoring method according to any one of claims 1-6.