Automatic labeling method and system for standardization of low-voltage distribution safety operation

Abstract

A normative automatic labeling method for low-voltage power distribution safety operation comprises the following steps: S1, acquiring a power distribution operation picture, and inputting the power distribution operation picture into an identification model; s2, the recognition model recognizes a worker and a working state according to the power distribution operation picture, wherein the working state comprises the position of the worker and clothes assembly of the worker; s3, obtaining a risk score of the current worker according to the working state of the worker, and obtaining a risk level of the worker according to a preset level range in which the risk score falls; a risk level is attached to the vicinity of a worker in a power distribution operation picture. The identification model is based on an advanced algorithm and technology, can accurately identify key information such as the position of a worker, clothing assembly and the like, and reduces errors possibly caused by manual observation. Therefore, preventive measures are taken in advance, the possibility of accidents is reduced, and the operation safety is improved.

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a method and system for automatically labeling standard low-voltage power distribution safety operations. Background Art

[0002] The low-voltage distribution system is an important part of my country's power system and has a wide range of applications. In recent years, my country's economy has developed rapidly, and the electricity consumption of various industries has continued to increase, which has made the application scope of the low-voltage distribution system more and more extensive. In order to ensure the safe use of electricity in the low-voltage distribution system, we should carefully understand the anti-electric shock protection methods of the low-voltage distribution system, such as repeated grounding, protective grounding, protective grounding, etc., combined with the grounding type of the low-voltage distribution system, make relevant safety settings, give full play to the application advantages of the low-voltage distribution system, and ensure people's life safety. In view of this, this article analyzes and discusses the safe use of electricity in the low-voltage distribution system. The use environment of low-voltage switches is complex and the frequency of use is high. In actual work, faults often occur, and power grid operation and maintenance personnel need to troubleshoot in time to ensure the use of electricity by residents and businesses. In the actual operation and maintenance process, the low-voltage switchgear is of different degrees of age, which makes low-voltage operations have certain risks.

[0003] In terms of regulating the safe operation behavior of power distribution personnel, although there are relatively complete technical regulations for power distribution safety, there are still cases where power distribution personnel ignore or even disregard safety regulations and perform dangerous power distribution operations. This behavior will greatly increase the risk of power distribution operations by power distribution personnel. In order to reduce the irregular operation behavior of power distribution personnel, many companies still need to monitor the power distribution operation site through monitoring equipment to ensure the safe operation of power distribution personnel. However, this method is very wasteful of manpower and material resources, and requires someone to observe the screen carefully for a long time to check, which greatly consumes the mental state of the monitoring personnel and is not conducive to the discovery of potential risks. Summary of the invention

[0004] In view of the above-mentioned defects, the purpose of the present invention is to propose a method and system for automatic labeling of low-voltage power distribution safety operation norms, so as to realize automatic monitoring of risk levels, reduce the possibility of accidents and improve operation safety.

[0005] To achieve this purpose, the present invention adopts the following technical solution: a method for automatically labeling the safety operation of low-voltage power distribution, comprising the following steps:

[0006] Step S1: obtaining a power distribution operation screen, and inputting the power distribution operation screen into a recognition model;

[0007] Step S2: the recognition model recognizes the worker and the working status according to the power distribution operation screen, wherein the working status includes: the position of the worker and the clothing of the worker;

[0008] Step S3: Obtain the risk score of the current staff member according to the staff member's work status, and obtain the risk level of the staff member according to the preset level range within which the risk score falls;

[0009] In the power distribution operation screen, a risk level is assigned to the vicinity of the worker.

[0010] Preferably, before executing step S1, the following steps need to be performed:

[0011] Obtain image data of clothing, hard hats, insulating gloves, bamboo ladders, tool boxes, and workers, and use annotation software to generate different data sets;

[0012] Performing an enhancement operation on the data set to obtain a first data set;

[0013] The first data set is input into the YOLOV9 neural network for training to obtain a trained recognition model.

[0014] Preferably, the risk score in step S3 includes a height work score, a clothing score, a safety helmet score, and an insulating gloves score.

[0015] Preferably, the climbing operation score is obtained as follows:

[0016] Identify the bamboo ladders in the power distribution operation screen and determine whether the number of bamboo ladders is greater than two. If so, crop the detection picture with only a single bamboo ladder through image segmentation. If not, use the power distribution operation screen as the detection picture.

[0017] Determine whether the height of the bamboo ladder in the detection image is greater than a threshold. If it is greater than the height threshold, assign a value to the height parameter; if it is not greater than the height threshold, do not assign a value to the height parameter;

[0018] Obtain the working conditions of workers climbing up bamboo ladders, where the working conditions include the climbing height of the workers, whether they are wearing safety measures, whether they have placed power distribution tools, and hand information;

[0019] Calculate the climbing work score according to the working conditions and height parameters;

[0020] The calculation formula for the climbing work score is as follows:

[0021]

[0022] Among them, ω1, ω2, ω3, and ω4 are project weights respectively, h is the height parameter, when assigned, h=1, when not assigned, h=0, F is the safety parameter for escalator personnel, Z1 is the safety parameter for the climbing height of the staff on the bamboo ladder, Z2 is the safety parameter for whether the staff wears safety measures, Z3 is the safety parameter for whether the staff places the distribution work tools, Z4 is the safety parameter for the distribution of the staff's hands, Z1′ is the reference safety parameter for the safe distance from the top of the ladder to climb, Z2′ is the reference safety parameter for fully implementing safety measures, Z3′ is the reference safety parameter for the correct way to use distribution work related tools, and Z4′ is the reference safety parameter for using both hands when going up and down the escalator.

[0023] Preferably, the steps for obtaining the work clothes rating are as follows:

[0024] Obtain the working scene of the power distribution operation screen and determine the working scene;

[0025] Determine whether the staff is wearing work clothes and / or reflective clothing. If they are wearing work clothes, assign a value to ε1; if they are wearing reflective clothing, assign a value to ε2;

[0026] Obtain the working clothes wearing situation of the staff, where the working clothes wearing situation includes wearing the working clothes in compliance, wearing the working clothes in non-compliance, and being unable to identify whether the working clothes are worn in compliance, and select the corresponding working clothes safety parameter C1 according to the working clothes wearing situation of the staff;

[0027] The reflective clothing wearing situation of the staff is obtained, where the work clothes wearing situation includes wearing type I reflective clothing, wearing type II reflective clothing, wearing type III reflective clothing, and not wearing reflective clothing, and the corresponding reflective clothing safety parameter C2 is selected according to the reflective clothing wearing situation of the staff and the working scene;

[0028] The work clothes score is calculated according to the reflective clothing safety parameters, the work clothes safety parameters, ε1 and ε2;

[0029] The calculation formula for the work clothes score is as follows:

[0030]

[0031] When ε1 is assigned a value, ε1=1, when ε1 is not assigned a value, ε1=0, when ε2 is assigned a value, ε2=1, when ε2 is not assigned a value, ε2=0;

[0032] C1 is the safety parameter of work clothes, C2 is the safety parameter of reflective clothing, C1′ is the reference safety parameter of work clothes worn in compliance with regulations, and C2′ is the reference safety parameter of reflective clothing worn in accordance with work scenarios.

[0033] Preferably, the steps for obtaining the helmet score are as follows:

[0034] Obtain the condition of workers wearing helmets in the distribution operation screen, and assign a value to the helmet safety parameter K1 according to the condition of wearing helmets, wherein the condition of wearing helmets includes: not wearing helmets, not wearing helmets correctly, wearing helmets correctly, and being unable to detect whether helmets are worn;

[0035] After the helmet is checked out, the helmet is tested to determine whether there are any damage points on the key points of the helmet, and the helmet safety parameter K2 is assigned according to the number of damage points;

[0036] The score of the helmet is calculated according to the helmet safety parameter K1 and the helmet safety parameter K2; wherein the score calculation formula of the helmet is as follows:

[0037]

[0038] in and is the weight coefficient, K1′ is the reference safety parameter for correctly wearing a helmet, and K is the reference safety parameter for an available helmet.

[0039] Preferably, the steps for obtaining the insulating gloves rating are as follows:

[0040] Obtain the wearing of insulating gloves by the worker in the distribution operation screen, and assign a value to the insulating gloves safety parameter M1 according to the wearing of insulating gloves, wherein the wearing of insulating gloves includes: not wearing insulating gloves, wearing insulating gloves, and being unable to detect whether gloves are worn;

[0041] After the insulating gloves are checked, the insulating gloves are tested to determine whether there are damaged points at the key points of the insulating gloves, and the safety parameter M2 of the insulating gloves is assigned according to the number of damaged points;

[0042] The score of the safety helmet is calculated according to the insulating gloves safety parameter M1 and the insulating gloves safety parameter M2; wherein the scoring calculation formula of the insulating gloves is as follows:

[0043]

[0044] Among them, τ1 and τ2 are weight coefficients, M1′ is the reference safety parameter for correctly wearing insulating gloves, and M is the reference safety parameter for available insulating gloves.

[0045] Preferably, the calculation formula of the risk level is as follows:

[0046] N=e -[α1*δ+α2*γ+α3*μ+α4*θ] ;

[0047] Among them, α1, α2, α3 and α4 are weight factors.

[0048] A low-voltage power distribution safety operation standard automatic labeling system, using the low-voltage power distribution safety operation standard automatic labeling method, including an input module, an identification module and a labeling module;

[0049] The input module is used to obtain the power distribution operation picture and input the power distribution operation picture into the recognition model;

[0050] The identification module is used to identify the staff and the working status according to the power distribution operation screen using the identification model;

[0051] The label module is used to obtain the risk score of the current staff member according to the work status of the staff member, and obtain the risk level of the staff member according to the preset level range within which the risk score falls;

[0052] In the power distribution operation screen, a risk level is assigned to the vicinity of the worker.

[0053] One of the above technical solutions has the following advantages or beneficial effects: the recognition model of the present invention is based on advanced algorithms and technologies, and can accurately identify key information such as the position of the staff, clothing assembly, etc., reducing the errors that may be caused by manual observation. Thus, preventive measures can be taken in advance to reduce the possibility of accidents and improve work safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flow chart of an embodiment of the method of the present invention.

[0055] Figure 2 This is a diagram of automatic labeling results for a low-voltage power distribution safety operation site in one embodiment of the present invention.

[0056] Figure 3 It is a schematic diagram of labels for different situations of wearing a helmet in one embodiment of the present invention.

[0057] Figure 4 It is a schematic diagram of the structure of an embodiment of the system of the present invention. DETAILED DESCRIPTION

[0058] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0059] In the description of the embodiments of the present invention, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0060] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "plurality" is two or more. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] like Figures 1 to 4 As shown, a method for automatically labeling the safety operation of low-voltage power distribution includes the following steps:

[0062] Step S1: obtaining a power distribution operation screen, and inputting the power distribution operation screen into a recognition model;

[0063] Step S2: the recognition model recognizes the worker and the working status according to the power distribution operation screen, wherein the working status includes: the position of the worker and the clothing of the worker;

[0064] Step S3: Obtain the risk score of the current staff member according to the staff member's work status, and obtain the risk level of the staff member according to the preset level range within which the risk score falls;

[0065] In the power distribution operation screen, a risk level is assigned to the vicinity of the worker.

[0066] In the present invention, a recognition model is used to identify the power distribution operation screen, wherein the recognition model can be used to identify the workers, the positions of the workers, and the clothing and accessories of the workers in the power distribution operation screen. Because in the power distribution work, the workers are required to wear insulating gloves, safety helmets, work clothes, etc. in accordance with regulations to ensure the safety of the power distribution workers. After the recognition model identifies the working status, it will score the risk of the workers in the screen according to the preset risk scoring standards, and then determine the risk level of the workers according to the value of the risk score, and attach the risk level to the workers in the power distribution operation screen in real time. Figure 2As shown, after the picture is input, the monitoring personnel can determine whether there are staff members who are not wearing the correct clothing in the picture according to the risk level in the picture. Of course, they can also add plug-ins to identify staff members whose risk level is lower than the risk level, and then issue warnings and other operations.

[0067] The recognition model of the present invention is based on advanced algorithms and technologies, and can accurately identify key information such as the position of the staff, clothing assembly, etc., reducing the errors that may be caused by manual observation. Preventive measures can be taken in advance to reduce the possibility of accidents and improve work safety.

[0068] Preferably, before executing step S1, the following steps need to be performed:

[0069] Obtain image data of clothing, hard hats, insulating gloves, bamboo ladders, tool boxes, and workers, and use annotation software to generate different data sets;

[0070] In one embodiment of the present invention, a professional labeling software is used to label the historical low-voltage power distribution safety operation site images according to the conditions of the detection targets, and generate data sets corresponding to different conditions of different detection targets, such as Figure 3 As shown in the figure, labeling tools can help operators accurately locate and mark targets. By carefully analyzing the inspection images, it is ensured that each target area is correctly labeled, thus forming a manually labeled target dataset.

[0071] Performing an enhancement operation on the data set to obtain a first data set;

[0072] The data enhancement operation can be: adding random pixels, Gaussian noise, random rectangular occlusion, random pixel zeroing, salt and pepper noise, animation effect, gradient mask blending, Gaussian blur, motion blur, changing color temperature, adaptive histogram equalization, horizontal flip, vertical flip, equal scaling, unequal scaling, random translation, HSV transformation, perspective transformation, random contrast, edge enhancement, random brightness, maximum pooling, average pooling, random cropping and filling and other image enhancement methods; the expanded first data set covers a wider range of power distribution safety operation scenarios and work wear conditions. The expanded scenarios are conducive to improving the generalization ability of low-voltage power distribution safety operation standardization detection, so that it can adapt to more different scenarios and optical fibers, and detection tasks under angles. In addition, by adding noisy data images, it is also helpful to improve the noise resistance of model detection.

[0073] The first data set is input into the YOLOV9 neural network for training to obtain a trained recognition model.

[0074] The recognition model described in the present invention is obtained by training the YOLOV9 neural network. YOLOV9 solves problems such as gradient disappearance or gradient saturation while maintaining the original neural network convergence speed and convergence effect.

[0075] Preferably, the risk score in step S3 includes a height work score, a clothing score, a safety helmet score, and an insulating gloves score.

[0076] In the present invention, since clothing, safety helmets, insulating gloves, bamboo ladders, tool boxes and image data of workers are added as training data when training the model, the recognition model can accurately identify information such as clothing, safety helmets, insulating gloves, bamboo ladders, tool boxes, etc. in the power distribution work screen; the identified data can cover all safety behavior norms of the power distribution workers, thereby comprehensively scoring whether the workers are working safely.

[0077] Preferably, the climbing operation score is obtained as follows:

[0078] Identify the bamboo ladders in the power distribution operation screen and determine whether the number of bamboo ladders is greater than two. If so, crop the detection picture with only a single bamboo ladder through image segmentation. If not, use the power distribution operation screen as the detection picture.

[0079] Determine whether the height of the bamboo ladder in the detection image is greater than a threshold. If it is greater than the height threshold, assign a value to the height parameter; if it is not greater than the height threshold, do not assign a value to the height parameter;

[0080] Obtain the working conditions of workers climbing up bamboo ladders, where the working conditions include the climbing height of the workers, whether they are wearing safety measures, whether they have placed power distribution tools, and hand information;

[0081] Calculate the climbing work score according to the working conditions and height parameters;

[0082] The calculation formula for the climbing work score is as follows:

[0083]

[0084] Among them, ω1, ω2, ω3, and ω4 are project weights respectively, h is the height parameter, when assigned, h=1, when not assigned, h=0, F is the safety parameter for escalator personnel, Z1 is the safety parameter for the climbing height of the staff on the bamboo ladder, Z2 is the safety parameter for whether the staff wears safety measures, Z3 is the safety parameter for whether the staff places the distribution work tools, Z4 is the safety parameter for the distribution of the staff's hands, Z1′ is the reference safety parameter for the safe distance from the top of the ladder to climb, Z2′ is the reference safety parameter for fully implementing safety measures, Z3′ is the reference safety parameter for the correct way to use distribution work related tools, and Z4′ is the reference safety parameter for using both hands when going up and down the escalator.

[0085] In the present invention, when calculating the score of equal-height operations, the staff climbing a single bamboo ladder will be used as the object of the score. With the help of image segmentation and cropping technology, the single bamboo ladder in the picture can be more accurately identified to reduce misjudgment. After the bamboo ladder is identified, the height of the bamboo ladder will be identified. When the height of the bamboo ladder is higher than the height threshold, danger will arise. Therefore, only when the height of the bamboo ladder is higher than the height threshold, the height parameter h is assigned. The scoring standards of Z1, Z2, Z3, Z4, Z1', Z2', Z3', and Z4' can be assigned accordingly according to the technical specifications issued by the state, among which the "National Electrical Equipment Safety Technical Specifications" GB 19517-2023 can be referred to. Z1, Z2, Z3, and Z4 are automatically assigned according to the height of the staff in the operation picture, whether they wear safety measures, whether they place distribution operation tools, and whether their hands fit the ladder. Z1', Z2', Z3', and Z4' are fixed values.

[0086] Preferably, the steps for obtaining the work clothes rating are as follows:

[0087] Obtain the working scene of the power distribution operation screen and determine the working scene;

[0088] Determine whether the staff is wearing work clothes and / or reflective clothing. If they are wearing work clothes, assign a value to ε1; if they are wearing reflective clothing, assign a value to ε2;

[0089] Obtain the working clothes wearing situation of the staff, where the working clothes wearing situation includes wearing the working clothes in compliance, wearing the working clothes in non-compliance, and being unable to identify whether the working clothes are worn in compliance, and select the corresponding working clothes safety parameter C1 according to the working clothes wearing situation of the staff;

[0090] The reflective clothing wearing situation of the staff is obtained, where the work clothes wearing situation includes wearing type I reflective clothing, wearing type II reflective clothing, wearing type III reflective clothing, and not wearing reflective clothing, and the corresponding reflective clothing safety parameter C2 is selected according to the reflective clothing wearing situation of the staff and the working scene;

[0091] The work clothes score is calculated according to the reflective clothing safety parameters, the work clothes safety parameters, ε1 and ε2;

[0092] The calculation formula for the work clothes score is as follows:

[0093]

[0094] When ε1 is assigned a value, ε1=1, when ε1 is not assigned a value, ε1=0, when ε2 is assigned a value, ε2=1, when ε2 is not assigned a value, ε2=0;

[0095] C1 is the safety parameter of work clothes, C2 is the safety parameter of reflective clothing, C1′ is the reference safety parameter of work clothes worn in compliance with regulations, and C2′ is the reference safety parameter of reflective clothing worn in accordance with work scenarios.

[0096] When it is identified that the staff is wearing work clothes, ε1 is assigned a value. In actual work, different reflective clothes need to be worn in different environmental scenes. At this time, the reflective clothing safety parameter C2 is assigned according to the different environments and reflective clothes worn by the staff. For example, when working in service areas, warehouses and roads within the factory area, according to the electrical equipment safety technical specifications, type I reflective clothes should be worn. If the staff is detected to be wearing type I reflective clothes, the C2 can be taken as 1 and ε2 is 1. If the staff is detected to be wearing type II or type III reflective clothes, the C2 can be taken as 0.5 and ε2 is still 1. If the staff is not detected to be wearing reflective clothes, ε2 can be taken as 0.1 and C2 can be taken as 0. Similarly, the work clothes safety parameter C1 can also be obtained by this method. And the C1′ and C2′ are fixed values.

[0097] Preferably, the steps for obtaining the helmet score are as follows:

[0098] Obtain the condition of workers wearing helmets in the distribution operation screen, and assign a value to the helmet safety parameter K1 according to the condition of wearing helmets, wherein the condition of wearing helmets includes: not wearing helmets, not wearing helmets correctly, wearing helmets correctly, and being unable to detect whether helmets are worn;

[0099] After the helmet is checked out, the helmet is tested to determine whether there are any damage points on the key points of the helmet, and the helmet safety parameter K2 is assigned according to the number of damage points;

[0100] The score of the helmet is calculated according to the helmet safety parameter K1 and the helmet safety parameter K2; wherein the score calculation formula of the helmet is as follows:

[0101]

[0102] in and is the weight coefficient, K1′ is the reference safety parameter for correctly wearing a helmet, and K is the reference safety parameter for an available helmet.

[0103] In the present invention, K1′ and K are fixed values, and K1 can be assigned a value according to the wearing condition of the helmet, for example, when the helmet is not worn, K1=0. When the helmet is not worn correctly, K1=0.5, when the helmet is worn correctly, K1=1. When it is impossible to detect whether the helmet is worn, K1=0.25.

[0104] The helmet safety parameter K2 is also assigned a corresponding value according to the number of damage points. For example, when 0 damage points are identified, K2=1, when 1 damage point is identified, K2=0.9, and so on, thereby obtaining the assignment of the helmet safety parameter K2.

[0105] Preferably, the steps for obtaining the insulating gloves rating are as follows:

[0106] Obtain the wearing of insulating gloves by the worker in the distribution operation screen, and assign a value to the insulating gloves safety parameter M1 according to the wearing of insulating gloves, wherein the wearing of insulating gloves includes: not wearing insulating gloves, wearing insulating gloves, and being unable to detect whether gloves are worn;

[0107] After the insulating gloves are checked, the insulating gloves are tested to determine whether there are damaged points at the key points of the insulating gloves, and the safety parameter M2 of the insulating gloves is assigned according to the number of damaged points;

[0108] The score of the safety helmet is calculated according to the insulating gloves safety parameter M1 and the insulating gloves safety parameter M2; wherein the scoring calculation formula of the insulating gloves is as follows:

[0109]

[0110] Among them, τ1 and τ2 are weight coefficients, M1′ is the reference safety parameter for correctly wearing insulating gloves, and M is the reference safety parameter for available insulating gloves.

[0111] In the present invention, M1′ and M are fixed values, and M1 can be assigned a value according to the wearing of insulating gloves. For example, when insulating gloves are not worn, M1=0. When insulating gloves are not worn correctly, M1=0.5, and when insulating gloves are worn correctly, M1=1. When it is impossible to detect whether insulating gloves are worn, M1=0.25.

[0112] The insulating glove safety parameter M2 is also assigned a corresponding value according to the number of damage points. For example, when 0 damage points are identified, M2=1, when 1 damage point is identified, M2=0.9, and so on, thereby obtaining the assignment of the insulating glove safety parameter M2.

[0113] Preferably, the calculation formula of the risk level is as follows:

[0114] N=e -[α1*δ+α2*γ+α3*μ+α4*θ] ;

[0115] Among them, α1, α2, α3 and α4 are weight factors.

[0116] A low-voltage power distribution safety operation standard automatic labeling system, using the low-voltage power distribution safety operation standard automatic labeling method, including an input module, an identification module and a labeling module;

[0117] The input module is used to obtain the power distribution operation picture and input the power distribution operation picture into the recognition model;

[0118] The identification module is used to identify the staff and the working status according to the power distribution operation screen using the identification model;

[0119] The label module is used to obtain the risk score of the current staff member according to the work status of the staff member, and obtain the risk level of the staff member according to the preset level range within which the risk score falls;

[0120] In the power distribution operation screen, a risk level is assigned to the vicinity of the worker.

[0121] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0122] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for automatically labeling the safety operation of low-voltage power distribution, characterized in that: The steps include: Step S1: obtaining a power distribution operation screen, and inputting the power distribution operation screen into a recognition model; Step S2: the recognition model recognizes the worker and the working status according to the power distribution operation screen, wherein the working status includes: the position of the worker and the clothing of the worker; Step S3: Obtain the risk score of the current staff member according to the staff member's work status, and obtain the risk level of the staff member according to the preset level range within which the risk score falls; In the power distribution operation screen, a risk level is assigned to the vicinity of the worker.

2. According to claim 1, a method for automatically labeling the safety operation of low-voltage power distribution is characterized in that: Before executing step S1, you need to perform the following steps: Obtain image data of clothing, hard hats, insulating gloves, bamboo ladders, tool boxes, and workers, and use annotation software to generate different data sets; Performing an enhancement operation on the data set to obtain a first data set; The first data set is input into the YOLOV9 neural network for training to obtain a trained recognition model.

3. A method for automatically labeling the safety operation of low-voltage power distribution according to claim 2, characterized in that: The risk score in step S3 includes a height work score, a clothing score, a safety helmet score, and an insulating glove score.

4. A method for automatically labeling the safety operation of low-voltage power distribution according to claim 3, characterized in that: The specific steps to obtain the climbing work score are as follows: Identify the bamboo ladders in the power distribution operation screen and determine whether the number of bamboo ladders is greater than two. If so, crop the detection picture with only a single bamboo ladder through image segmentation. If not, use the power distribution operation screen as the detection picture. Determine whether the height of the bamboo ladder in the detection image is greater than a threshold. If it is greater than the height threshold, assign a value to the height parameter; if it is not greater than the height threshold, do not assign a value to the height parameter; Obtain the working conditions of workers climbing up bamboo ladders, where the working conditions include the climbing height of the workers, whether they are wearing safety measures, whether they have placed power distribution tools, and hand information; Calculate the climbing work score according to the working conditions and height parameters; The calculation formula for the climbing work score is as follows: Among them, ω1, ω2, ω3, and ω4 are project weights respectively, h is the height parameter, when assigned, h=1, when not assigned, h=0, F is the safety parameter for escalator personnel, Z1 is the safety parameter for the climbing height of the staff on the bamboo ladder, Z2 is the safety parameter for whether the staff wears safety measures, Z3 is the safety parameter for whether the staff places the distribution work tools, Z4 is the safety parameter for the distribution of the staff's hands, Z1′ is the reference safety parameter for the safe distance from the top of the ladder to climb, Z2′ is the reference safety parameter for fully implementing safety measures, Z3′ is the reference safety parameter for the correct way to use distribution work related tools, and Z4′ is the reference safety parameter for using both hands when going up and down the escalator.

5. A method for automatically labeling the safety operation of low-voltage power distribution according to claim 4, characterized in that: The steps to obtain a workwear rating are as follows: Obtain the working scene of the power distribution operation screen and determine the working scene; Determine whether the staff is wearing work clothes and / or reflective clothing. If they are wearing work clothes, assign a value to ε1; if they are wearing reflective clothing, assign a value to ε2; Obtain the working clothes wearing situation of the staff, where the working clothes wearing situation includes wearing the working clothes in compliance, wearing the working clothes in non-compliance, and being unable to identify whether the working clothes are worn in compliance, and select the corresponding working clothes safety parameter C1 according to the working clothes wearing situation of the staff; The reflective clothing wearing situation of the staff is obtained, where the work clothes wearing situation includes wearing type I reflective clothing, wearing type II reflective clothing, wearing type III reflective clothing, and not wearing reflective clothing, and the corresponding reflective clothing safety parameter C2 is selected according to the reflective clothing wearing situation of the staff and the working scene; The work clothes score is calculated according to the reflective clothing safety parameters, the work clothes safety parameters, ε1 and ε2; The calculation formula for the work clothes score is as follows: When ε1 is assigned a value, ε1=1, when ε1 is not assigned a value, ε1=0, when ε2 is assigned a value, ε2=1, when ε2 is not assigned a value, ε2=0; C1 is the safety parameter of work clothes, C2 is the safety parameter of reflective clothing, C1′ is the reference safety parameter of work clothes worn in compliance with regulations, and C2′ is the reference safety parameter of reflective clothing worn in accordance with work scenarios.

6. A method for automatically labeling the safety operation of low-voltage power distribution according to claim 5, characterized in that: The steps to obtain a hard hat rating are as follows: Obtain the condition of workers wearing helmets in the distribution operation screen, and assign a value to the helmet safety parameter K1 according to the condition of wearing helmets, wherein the condition of wearing helmets includes: not wearing helmets, not wearing helmets correctly, wearing helmets correctly, and being unable to detect whether helmets are worn; After the helmet is checked out, the helmet is tested to determine whether there are any damage points on the key points of the helmet, and the helmet safety parameter K2 is assigned according to the number of damage points; The score of the helmet is calculated according to the helmet safety parameter K1 and the helmet safety parameter K2; wherein the score calculation formula of the helmet is as follows: in and is the weight coefficient, K1′ is the reference safety parameter for correctly wearing a helmet, and K is the reference safety parameter for an available helmet.

7. A method for automatically labeling the safety operation of low-voltage power distribution according to claim 6, characterized in that: The steps to obtain an insulating glove rating are as follows: Obtain the wearing of insulating gloves by the worker in the distribution operation screen, and assign a value to the insulating gloves safety parameter M1 according to the wearing of insulating gloves, wherein the wearing of insulating gloves includes: not wearing insulating gloves, wearing insulating gloves, and being unable to detect whether gloves are worn; After the insulating gloves are checked, the insulating gloves are tested to determine whether there are damaged points at the key points of the insulating gloves, and the safety parameter M2 of the insulating gloves is assigned according to the number of damaged points; The score of the safety helmet is calculated according to the insulating gloves safety parameter M1 and the insulating gloves safety parameter M2; wherein the scoring calculation formula of the insulating gloves is as follows: Among them, τ1 and τ2 are weight coefficients, M1′ is the reference safety parameter for correctly wearing insulating gloves, and M is the reference safety parameter for available insulating gloves.

8. A method for automatically labeling the safety operation of low-voltage power distribution according to claim 7, characterized in that: The risk level is calculated as follows: N=e -[α1*δ+α2*γ+α3*μ+α4*θ] ; Among them, α1, α2, α3 and α4 are weight factors.

9. A low-voltage power distribution safety operation standard automatic labeling system, characterized in that: A method for automatically labeling low-voltage power distribution safety operations as described in any one of claims 1 to 8, comprising an input module, an identification module and a labeling module; The input module is used to obtain the power distribution operation picture and input the power distribution operation picture into the recognition model; The identification module is used to identify the staff and the working status according to the power distribution operation screen using the identification model; The label module is used to obtain the risk score of the current staff member according to the work status of the staff member, and obtain the risk level of the staff member according to the preset level range within which the risk score falls; In the power distribution operation screen, a risk level is assigned to the vicinity of the worker.

Images