Power transformation operation safety distance monitoring method and device, electronic equipment and storage medium

Abstract

The invention discloses a power transformation operation safety distance monitoring method and device, electronic equipment and a storage medium, and relates to the technical field of power operation. The method comprises the following steps: acquiring a left eye operation image acquired by a left eye camera and a right eye operation image acquired by a right eye camera in a binocular camera, wherein the left eye operation image and the right eye operation image both comprise operating personnel and electrified equipment; inputting the left-eye operation image and the right-eye operation image into a pre-constructed stereo matching model, obtaining a disparity map corresponding to the operator and the live-line equipment, and converting the disparity map into a depth map; inputting the left-eye operation image or the right-eye operation image into a pre-trained target detection model, identifying operating personnel and electrified equipment, and extracting three-dimensional point cloud information corresponding to the operating personnel and the electrified equipment from the depth map; the distance between the operator and the electrified equipment is calculated according to the three-dimensional point cloud information, and an alarm is given when the distance is smaller than or equal to a preset threshold value, so that the distance can be efficiently and accurately detected.

Description

Technical Field

[0001] The present application relates to the technical field of electric power operations, and in particular to a method, device, electronic equipment, and storage medium for monitoring safety distances in substation operations. Background Art

[0002] As global energy demand continues to grow and power systems expand in complexity and scale, the operational safety of high-voltage substations, critical nodes in the power grid, is directly linked to the stability of the entire network and the smooth functioning of socioeconomic activities. Substation operations, operating in the specialized environment of high voltage and high current, place extremely high demands on personnel safety. Improper operation or lax adherence to safety regulations can lead to serious safety incidents such as electric shock, equipment damage, and even large-scale power outages. These threats not only threaten personal safety but also result in significant economic losses and negative social impacts.

[0003] When detecting the distance of substation operations, relevant technologies mainly use electric field monitoring and positioning technologies for operation. Among them, the method based on electric field monitoring measures the electric field strength of personnel and equipment and estimates the distance between personnel and live objects. However, due to factors such as the easy distortion of high-voltage electric fields and the difficulty in accurately measuring electric field strength, this method has large detection errors. The method based on positioning technology uses Beidou and ultra-wideband wireless communication technology (Ultra-Wide Band, UWB) to monitor the location of personnel and equipment in real time and calculate the distance between personnel and live equipment. However, the signal in this method is easily affected by weather changes, and the construction cost, cycle and maintenance cost are high. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the relevant technology, it is desired to provide a method, device, electronic equipment and storage medium for monitoring the safe distance of substation operations, which can efficiently and accurately detect the distance between operators and live equipment, reduce costs and ensure operation safety.

[0005] In a first aspect, the present application provides a method for monitoring a safe distance for a substation operation, the method comprising:

[0006] Obtaining a left-eye operation image captured by a left camera and a right-eye operation image captured by a right camera in a binocular camera, wherein the binocular camera includes the left camera and the right camera arranged horizontally, and both the left-eye operation image and the right-eye operation image include an operator and live equipment;

[0007] Inputting the left-eye working image and the right-eye working image into a pre-built stereo matching model to obtain a disparity map corresponding to the worker and the energized equipment, and converting the disparity map into a depth map;

[0008] Inputting the left-eye operation image or the right-eye operation image into a pre-trained target detection model, identifying the operator and the live equipment, and extracting three-dimensional point cloud information corresponding to the operator and the live equipment respectively from the depth map;

[0009] The distance between the operator and the energized equipment is calculated based on the three-dimensional point cloud information, and an alarm is issued when the distance is less than or equal to a preset threshold.

[0010] Optionally, in some embodiments of the present application, before converting the disparity map into a depth map, the method for monitoring the safe distance of substation operations further includes performing hole filling, disparity smoothing, and left-right consistency detection on the disparity map.

[0011] Optionally, in some embodiments of the present application, calculating the distance between the operator and the energized equipment based on the three-dimensional point cloud information includes:

[0012] Obtaining a first coordinate value set corresponding to a detection bounding box of the operator and a first depth value set within the box, and a second coordinate value set corresponding to a detection bounding box of the energized device and a second depth value set within the box;

[0013] Using an average value of coordinates in the first coordinate value set as the coordinate value of the operator, using an average value of depths in the first depth value set as the depth value of the operator, and using an average value of coordinates in the second coordinate value set as the coordinate value of the energized device, and using an average value of depths in the second depth value set as the depth value of the energized device;

[0014] The distance between the worker and the live equipment is calculated based on the coordinate value of the worker and the depth value of the worker and the coordinate value of the live equipment and the depth value of the live equipment.

[0015] Optionally, in some embodiments of the present application, the stereo matching model is a CREStereo model, and the target detection model is a YOLOv8 model.

[0016] Optionally, in some embodiments of the present application, the substation operation safety distance monitoring method further includes pruning and lightweighting the YOLOv8 model.

[0017] Optionally, in some embodiments of the present application, before obtaining the left-eye operation image captured by the left camera and the right-eye operation image captured by the right camera in the binocular camera, the substation operation safety distance monitoring method also includes calibrating and correcting the binocular camera.

[0018] In a second aspect, the present application provides a power substation operation safety distance monitoring device, the power substation operation safety distance monitoring device comprising:

[0019] an acquisition module, configured to acquire a left-eye operation image captured by a left camera and a right-eye operation image captured by a right camera in a binocular camera, wherein the binocular camera includes the left camera and the right camera arranged horizontally, and the left-eye operation image and the right-eye operation image both include an operator and live equipment;

[0020] A stereo matching module is configured to input the left-eye working image and the right-eye working image into a pre-built stereo matching model to obtain a disparity map corresponding to the working person and the energized equipment, and convert the disparity map into a depth map;

[0021] a target detection module, configured to input the left-eye operation image or the right-eye operation image into a pre-trained target detection model, identify the operator and the energized equipment, and extract three-dimensional point cloud information corresponding to the operator and the energized equipment respectively from the depth map;

[0022] An alarm module is used to calculate the distance between the operator and the energized equipment based on the three-dimensional point cloud information, and to issue an alarm when the distance is less than or equal to a preset threshold.

[0023] Optionally, in some embodiments of the present application, the alarm module includes:

[0024] an acquisition unit, specifically configured to acquire a first coordinate value set corresponding to a detection bounding box of the operator and a first depth value set within the box, and a second coordinate value set corresponding to a detection bounding box of the energized device and a second depth value set within the box;

[0025] a first calculation unit, specifically configured to use an average coordinate value in the first coordinate value set as the coordinate value of the operator, an average depth value in the first depth value set as the depth value of the operator, and use an average coordinate value in the second coordinate value set as the coordinate value of the energized device, and use an average depth value in the second depth value set as the depth value of the energized device;

[0026] The second calculation unit is specifically configured to calculate the distance between the operator and the live device based on the coordinate value of the operator and the depth value of the operator and the coordinate value of the live device and the depth value of the live device.

[0027] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the instruction, program, code set or instruction set is loaded and executed by the processor to implement the steps of the substation operation safety distance monitoring method described in any one of the first aspects.

[0028] In a fourth aspect, the present application provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the substation operation safety distance monitoring method described in any one of the first aspects.

[0029] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0030] The embodiments of the present application provide a method, device, electronic device and storage medium for monitoring the safe distance of substation operations. The left camera and the right camera in the binocular camera respectively collect high-resolution left-eye operation images and right-eye operation images, and then use the stereo matching model to process the left-eye operation image and the right-eye operation image to obtain an accurate disparity map corresponding to the operator and the live equipment, and convert the disparity map into a depth map to understand the spatial distribution information. At the same time, the target detection model is used to process the left-eye operation image or the right-eye operation image to accurately identify the operator and the live equipment, and extract the three-dimensional point cloud information corresponding to the operator and the live equipment from the depth map. Then, the distance between the operator and the live equipment can be quickly calculated based on the three-dimensional point cloud information. The required monitoring cost is low, and an alarm is issued in time when the distance is less than or equal to the preset threshold, thereby improving the overall safety and operational efficiency of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 A flow chart of a method for monitoring safe distances in substation operations provided in an embodiment of the present application;

[0033] Figure 2 A schematic diagram of an imaging model and geometric model of a binocular camera provided in an embodiment of the present application;

[0034] Figure 3A structural block diagram of a safety distance monitoring device for substation operations provided in an embodiment of the present application;

[0035] Figure 4 A structural block diagram of another power transformation operation safety distance monitoring device provided in an embodiment of the present application;

[0036] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0038] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0039] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other. Figures 1 to 5 The invention describes in detail the method, device, electronic device and storage medium for monitoring the safe distance of substation operations provided by the embodiments of the present application.

[0040] Please refer to Figure 1 , which is a flow chart of a method for monitoring safe distance of substation operation provided by an embodiment of the present application, the method specifically comprises the following steps:

[0041] S101, obtaining a left-eye operation image captured by a left camera and a right-eye operation image captured by a right camera in a binocular camera, wherein the binocular camera includes a left camera and a right camera arranged horizontally, and both the left-eye operation image and the right-eye operation image include an operator and live equipment.

[0042] For example, Figure 2 As shown, it is a schematic diagram of an imaging model and geometric model of a binocular camera provided in an embodiment of the present application, where the optical center O of the left camera is L and the optical center O of the right camera R Both are located on the x-axis. The line segment between them is called the baseline of the binocular camera, denoted as b. The image of a point P in space on the left camera is denoted as P. L , the image of the right camera is denoted as P R , f represents the focal length, u L and u R is the coordinate of the imaging plane, where uR is a negative number. According to the triangle similarity principle, ΔPP L P R ~ΔPO L O R ,but

[0043]

[0044] Further, we can get

[0045]

[0046] In formula (2), d represents the difference between the horizontal coordinates of the left eye working image and the right eye working image, that is, the disparity, and its value is u L -u R ; z represents the depth value.

[0047] For example, before obtaining the left-eye working image captured by the left camera and the right-eye working image captured by the right camera, embodiments of the present application can also calibrate and correct the binocular cameras. For example, a standard checkerboard is first selected; then, the binocular cameras are fixed and multi-view images are simultaneously captured, with the number of images ranging from 20 to 30. The checkerboard corner points are extracted from the images using OpenCV; then, the binocular camera intrinsic and extrinsic parameters are obtained, and the distortion coefficients are calculated; and then, stereo correction is performed on the left and right views.

[0048] S102 : Input the left-eye working image and the right-eye working image into a pre-built stereo matching model to obtain a disparity map corresponding to the working personnel and the energized equipment, and convert the disparity map into a depth map.

[0049] For example, the stereo matching model can be a CREStereo model, that is, a cascaded cyclic stereo matching network, which uses a recursively refined hierarchical network to update the disparity from coarse to fine, and uses a stacked cascade architecture for high-resolution reasoning. At the same time, an adaptive group association layer is designed to reduce the impact of error correction. The embodiment of the present application uses the CREStereo model training weights to perform deep reasoning on the substation scene through the method of transfer learning, thereby better adapting to the complex environment and diverse texture conditions of the scene, integrating a wider range of contextual information, helping to solve problems such as occlusion, repeated textures and discontinuities, and then making accurate disparity estimates. It has strong generalization capabilities, better matching effects for unseen data and new scenes, and strong flexibility and scalability.

[0050] For example, before converting the disparity map into a depth map, the embodiment of the present application can also perform hole filling, disparity smoothing, and left-right consistency detection on the disparity map. For example, hole filling includes operations such as hole detection, neighborhood selection, adaptive threshold calculation, pixel value selection, and iterative filling. Hole detection refers to detecting the hole area in the disparity map, neighborhood selection refers to selecting a suitable neighborhood range for each hole pixel, adaptive threshold calculation refers to calculating an adaptive threshold for the selected neighborhood, pixel value selection refers to selecting the pixel value that meets the conditions in the neighborhood for filling the hole according to the adaptive threshold, and iterative filling means that if the hole is large, it may take multiple iterations to gradually fill the hole area until all the holes are filled, thereby filling the missing areas caused by occlusion and other reasons in the disparity map, thereby improving the integrity and practicality of the disparity map. For example, disparity smoothing uses a mean filter for smoothing, which replaces the value of the pixel by calculating the average value of the neighbors around each pixel to achieve a smoothing effect, thereby reducing the noise in the disparity map and improving the quality of the disparity map. For example, left-right consistency detection first uses a stereo matching algorithm to calculate an initial disparity map from the left-eye working image to the right-eye working image, then performs stereo matching again from the right-eye working image to the left-eye working image to obtain a reverse disparity map, and then compares the disparity values ​​obtained in the two directions to check whether they are consistent. If there are inconsistencies, the final disparity value can be determined by majority voting within the local window, and these pixels can be corrected or excluded.

[0051] S103: Input the left-eye operation image or the right-eye operation image into a pre-trained target detection model to identify the operator and the live equipment, and extract the three-dimensional point cloud information corresponding to the operator and the live equipment from the depth map.

[0052] Exemplarily, the target detection model in the embodiment of the present application can be a YOLOv8 model, and the training process of the model is conventional, which will not be described in detail. Furthermore, some embodiments of the present application can also prune and lightweight the YOLOv8 model, thereby reducing latency and ensuring that the system can respond in a timely manner in a highly dynamic operating environment. For example, when pruning, the L2 norm is first calculated for each filter. The L2 norm refers to taking the square root of the sum of the squares of the filter weights. Secondly, the filters are sorted according to the L2 norm, and then, based on the sorting results, the filters with the lowest index are selectively removed, that is, those filters that contribute the least to the model performance. For example, lightweighting is weight lightweighting, that is, converting the floating-point number float32 into int8.

[0053] S104: Calculate the distance between the operator and the energized equipment based on the three-dimensional point cloud information, and issue an alarm when the distance is less than or equal to a preset threshold.

[0054] For example, first obtain the first coordinate value set and the first depth value set in the detection bounding box of the operator, and the second coordinate value set and the second depth value set in the detection bounding box of the live equipment; secondly, use the average coordinate value in the first coordinate value set as the coordinate value of the operator, the average depth value in the first depth value set as the depth value of the operator, and the average coordinate value in the second coordinate value set as the coordinate value of the live equipment, and the average depth value in the second depth value set as the depth value of the live equipment. The reason for this setting is that the detection bounding box surrounds the outside of the target, and the depth value of the center point is not necessarily the depth value of the target. Therefore, the average depth value in the detection bounding box is selected as the depth value of the target. For example, the detection bounding box box = [x1, y1, x2, y2], [x1, y1] represents the coordinate of the lower left corner, [x2, y2] represents the coordinate of the upper right corner, the coordinate average x_middle = (x1+x2) / 2, y_middle = (y1+y2) / 2, and the value of the three-dimensional point Point3D containing x, y and z is

[0055]

[0056] In formula (3), [u, v] represents the pixel coordinate value on the two-dimensional depth map, [c x ,c y ] represents the coordinate value of the camera optical center in the image coordinate system.

[0057] Finally, the distance between the operator and the live equipment is calculated based on the operator's coordinate values ​​and depth values ​​as well as the coordinate values ​​and depth values ​​of the live equipment. That is, the average depth value is used to replace z in this process.

[0058] For example, a preset threshold can be set according to safety regulations and historical data. If the distance is less than or equal to the preset threshold, an alarm will be sent through sound, light or directly to the monitoring center to quickly inform on-site personnel and management personnel to take measures to ensure operational safety.

[0059] The embodiment of the present application provides a method for monitoring the safe distance of substation operations. The left camera and the right camera in the binocular camera respectively collect high-resolution left-eye operation images and right-eye operation images, and then use the stereo matching model to process the left-eye operation image and the right-eye operation image to obtain an accurate disparity map corresponding to the operator and the live equipment, and convert the disparity map into a depth map to understand the spatial distribution information. At the same time, the target detection model is used to process the left-eye operation image or the right-eye operation image to accurately identify the operator and the live equipment, and extract the three-dimensional point cloud information corresponding to the operator and the live equipment from the depth map. Then, the distance between the operator and the live equipment can be quickly calculated based on the three-dimensional point cloud information. The required monitoring cost is low, and an alarm is issued in time when the distance is less than or equal to the preset threshold, thereby improving the overall safety and operational efficiency of the power system.

[0060] Based on the above embodiments, the present application provides a device for monitoring the safe distance of substation operation. The device 100 can be applied to Figures 1 and 2 In the corresponding embodiment of the power transformation operation safety distance monitoring method. Please refer to Figure 3 The substation operation safety distance monitoring device 100 includes:

[0061] An acquisition module 101 is configured to acquire a left-eye operation image captured by a left camera and a right-eye operation image captured by a right camera of a binocular camera, wherein the binocular camera includes a left camera and a right camera arranged horizontally, and both the left-eye operation image and the right-eye operation image include an operator and live equipment;

[0062] The stereo matching module 102 is used to input the left-eye working image and the right-eye working image into a pre-built stereo matching model to obtain a disparity map corresponding to the working person and the live equipment, and convert the disparity map into a depth map;

[0063] The target detection module 103 is used to input the left-eye operation image or the right-eye operation image into a pre-trained target detection model, identify the operator and the live equipment, and extract the three-dimensional point cloud information corresponding to the operator and the live equipment from the depth map;

[0064] The alarm module 104 is used to calculate the distance between the operator and the energized equipment based on the three-dimensional point cloud information, and issue an alarm when the distance is less than or equal to a preset threshold.

[0065] Optionally, in some embodiments of the present application, the stereo matching module 102 is further configured to perform hole filling, disparity smoothing, and left-right consistency detection on the disparity map before converting the disparity map into a depth map.

[0066] Alternatively, as Figure 4 As shown, in some embodiments of the present application, the alarm module 104 includes:

[0067] The acquisition unit 1041 is specifically configured to acquire a first coordinate value set corresponding to a detection bounding box of the operator and a first depth value set within the bounding box, and a second coordinate value set corresponding to a detection bounding box of the energized device and a second depth value set within the bounding box;

[0068] The first calculation unit 1042 is specifically configured to use an average value of coordinates in the first coordinate value set as the coordinate value of the operator, an average value of depths in the first depth value set as the depth value of the operator, and an average value of coordinates in the second coordinate value set as the coordinate value of the energized device, and an average value of depths in the second depth value set as the depth value of the energized device;

[0069] The second calculation unit 1043 is specifically configured to calculate the distance between the operator and the live equipment based on the coordinate value and depth value of the operator and the coordinate value and depth value of the live equipment.

[0070] Optionally, in some embodiments of the present application, the stereo matching model is a CREStereo model, and the target detection model is a YOLOv8 model.

[0071] Optionally, in some embodiments of the present application, the target detection module 103 is also used to prune and lightweight the YOLOv8 model.

[0072] Optionally, in some embodiments of the present application, the acquisition module 101 is further configured to calibrate and correct the binocular camera before acquiring the left-eye working image captured by the left camera and the right-eye working image captured by the right camera.

[0073] It should be noted that, for the description of the same steps and contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0074] The embodiment of the present application provides a substation operation safety distance monitoring device, which uses a left-eye camera and a right-eye camera in a binocular camera to respectively collect high-resolution left-eye operation images and right-eye operation images, and then uses a stereo matching model to process the left-eye operation image and the right-eye operation image to obtain an accurate disparity map corresponding to the operator and the live equipment, and converts the disparity map into a depth map to understand the spatial distribution information. At the same time, the target detection model is used to process the left-eye operation image or the right-eye operation image to accurately identify the operator and the live equipment, and extract the three-dimensional point cloud information corresponding to the operator and the live equipment from the depth map, so that the distance between the operator and the live equipment can be quickly calculated based on the three-dimensional point cloud information. The required monitoring cost is low, and an alarm is issued in time when the distance is less than or equal to the preset threshold, thereby improving the overall safety and operational efficiency of the power system.

[0075] Based on the above embodiments, the present application provides an electronic device. Figure 5 The electronic device 200 may include a processor 201 and a memory 202. The memory 202 stores at least one instruction, at least one program, code set or instruction set, which is loaded and executed by the processor 201 to implement Figures 1 and 2 The steps of the method for monitoring the safe distance of substation operations in the corresponding embodiment.

[0076] As another aspect, the present invention provides a computer-readable storage medium for storing program code for executing the aforementioned Figures 1 and 2 Any implementation method of the substation safety distance monitoring method of the corresponding embodiment.

[0077] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0078] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. On the other hand, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0079] In addition, the functional modules in the various embodiments of the present application may be integrated into a single processing unit, or each module may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated units may be implemented in the form of hardware or in the form of software functional units. If the integrated units are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium.

[0080] Based on this understanding, the technical solution of the present application, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the substation operation safety distance monitoring method of each embodiment of the present application. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., various media that can store program code.

[0081] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for monitoring the safe distance of power transformation operation, characterized in that: The method for monitoring the safe distance of substation operation includes: Obtaining a left-eye operation image captured by a left camera and a right-eye operation image captured by a right camera in a binocular camera, wherein the binocular camera includes the left camera and the right camera arranged horizontally, and both the left-eye operation image and the right-eye operation image include an operator and live equipment; Inputting the left-eye working image and the right-eye working image into a pre-built stereo matching model to obtain a disparity map corresponding to the worker and the energized equipment, and converting the disparity map into a depth map; Inputting the left-eye operation image or the right-eye operation image into a pre-trained target detection model, identifying the operator and the live equipment, and extracting three-dimensional point cloud information corresponding to the operator and the live equipment respectively from the depth map; The distance between the operator and the energized equipment is calculated based on the three-dimensional point cloud information, and an alarm is issued when the distance is less than or equal to a preset threshold.

2. The method for monitoring the safe distance of substation operation according to claim 1, characterized in that: Before converting the disparity map into a depth map, the method for monitoring the safe distance of substation operations further includes performing hole filling, disparity smoothing, and left-right consistency detection on the disparity map.

3. The method for monitoring the safe distance of substation operation according to claim 1, characterized in that: Calculating the distance between the operator and the energized equipment based on the three-dimensional point cloud information includes: Obtaining a first coordinate value set corresponding to a detection bounding box of the operator and a first depth value set within the box, and a second coordinate value set corresponding to a detection bounding box of the energized device and a second depth value set within the box; Using an average value of coordinates in the first coordinate value set as the coordinate value of the operator, using an average value of depths in the first depth value set as the depth value of the operator, and using an average value of coordinates in the second coordinate value set as the coordinate value of the energized device, and using an average value of depths in the second depth value set as the depth value of the energized device; The distance between the worker and the live equipment is calculated based on the coordinate value of the worker and the depth value of the worker and the coordinate value of the live equipment and the depth value of the live equipment.

4. The method for monitoring the safe distance of substation operation according to any one of claims 1 to 3, characterized in that: The stereo matching model is a CREStereo model, and the target detection model is a YOLOv8 model.

5. The method for monitoring the safe distance of substation operation according to claim 4, characterized in that: The substation operation safety distance monitoring method also includes pruning and lightweight processing of the YOLOv8 model.

6. The method for monitoring the safe distance of substation operation according to claim 4, characterized in that: Before obtaining the left-eye operation image captured by the left camera and the right-eye operation image captured by the right camera in the binocular camera, the substation operation safety distance monitoring method further includes calibrating and correcting the binocular camera.

7. A safety distance monitoring device for substation operation, characterized in that: The power transformation operation safety distance monitoring device includes: an acquisition module, configured to acquire a left-eye operation image captured by a left camera and a right-eye operation image captured by a right camera in a binocular camera, wherein the binocular camera includes the left camera and the right camera arranged horizontally, and the left-eye operation image and the right-eye operation image both include an operator and live equipment; A stereo matching module is configured to input the left-eye working image and the right-eye working image into a pre-built stereo matching model to obtain a disparity map corresponding to the working person and the energized equipment, and convert the disparity map into a depth map; a target detection module, configured to input the left-eye operation image or the right-eye operation image into a pre-trained target detection model, identify the operator and the energized equipment, and extract three-dimensional point cloud information corresponding to the operator and the energized equipment respectively from the depth map; An alarm module is used to calculate the distance between the operator and the energized equipment based on the three-dimensional point cloud information, and to issue an alarm when the distance is less than or equal to a preset threshold.

8. The power transformation operation safety distance monitoring device according to claim 7, characterized in that: The alarm module comprises: an acquisition unit, specifically configured to acquire a first coordinate value set corresponding to a detection bounding box of the operator and a first depth value set within the box, and a second coordinate value set corresponding to a detection bounding box of the energized device and a second depth value set within the box; a first calculation unit, specifically configured to use an average coordinate value in the first coordinate value set as the coordinate value of the operator, an average depth value in the first depth value set as the depth value of the operator, and use an average coordinate value in the second coordinate value set as the coordinate value of the energized device, and use an average depth value in the second depth value set as the depth value of the energized device; The second calculation unit is specifically configured to calculate the distance between the operator and the live device based on the coordinate value of the operator and the depth value of the operator and the coordinate value of the live device and the depth value of the live device.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the instruction, the program, the code set or the instruction set is loaded and executed by the processor to implement the steps of the substation operation safety distance monitoring method described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the substation operation safety distance monitoring method according to any one of claims 1 to 6.

Images