Method and device for determining thermal defect of equipment, equipment, medium and program product

Abstract

The invention relates to an equipment thermal defect determination method and device, equipment, a medium and a program product. The method comprises the following steps: acquiring a current visible light image and a current infrared image which are respectively shot by a visible light camera and an infrared light camera at the same position and the same shooting angle in the process of polling the power distribution network by polling equipment; inputting the current visible light image into a power equipment identification model, and outputting a power equipment layer which identifies the outline of the power equipment; superposing the electrical equipment layer and the current infrared image to generate a superposed image; and inputting the superposed image into the defect identification model, and outputting the position coordinates, the abnormality type and the thermal defect degree of the abnormal hot spot corresponding to the power equipment in the current infrared image. According to the method provided by the invention, the thermal defect of the power equipment can be accurately detected.

Description

Technical Field

[0001] The present application relates to the technical field of power grids, and particularly to a method, device, equipment, medium and program product for determining thermal defects of equipment. Background Art

[0002] With the continuous development of unmanned aerial vehicle (UAV) technology, UAVs have been widely used in the inspection process of power lines. UAVs can obtain data of power equipment in power lines through devices such as visible light cameras, infrared light cameras, and lidar carried on them. Among them, the infrared light camera uses the infrared radiation emitted by the object itself for imaging, which can reveal the thermal state of the object. In UAV inspections, infrared light cameras are widely used to detect thermal defects of power equipment, such as overheated joints and aging insulators. These thermal defects may lead to equipment failures or even accidents, so it is of great significance to detect and handle them in a timely manner. However, in the prior art, it is impossible to clearly distinguish power equipment from other objects in infrared images, resulting in inaccurate detection results of power equipment. Summary of the Invention

[0003] Based on this, in view of the above technical problems, it is necessary to provide a method, device, equipment, medium and program product for determining thermal defects of equipment that can improve the accuracy of the detection results of power equipment.

[0004] In a first aspect, the present application provides a method for determining thermal defects of equipment, the method comprising:

[0005] During the process of a patrol device patrolling a distribution network, obtaining a current visible light image and a current infrared image respectively taken by a visible light camera and an infrared light camera at the same position and the same shooting angle; wherein, the sizes of the current visible light image and the current infrared image are the same;

[0006] Inputting the current visible light image into a power equipment recognition model, and outputting a power equipment layer that marks the contour of the power equipment;

[0007] Overlaying the power equipment layer and the current infrared image to generate an overlay image;

[0008] Inputting the overlay image into a defect recognition model, and outputting the position coordinates, abnormal type and thermal defect degree of the abnormal hot spots corresponding to the power equipment in the current infrared image.

[0009] In one embodiment, the pixel values of the pixel points located inside the contour of the power equipment in the power equipment layer are the corresponding white pixel values, and the pixel values of the pixel points located outside the contour of the power equipment are the corresponding black pixel values.

[0010] In one embodiment, overlaying the power equipment layer and the current infrared image includes:

[0011] Replace the pixel values of the pixel points within the outline of the power equipment in the power equipment layer with the pixel values of the pixel points at the same position in the current infrared image, and replace the pixel values of the pixel points outside the outline of the power equipment in the power equipment layer with preset pixel values.

[0012] In one embodiment, the training process of the power equipment recognition model includes:

[0013] Obtain multiple historical visible light images captured by a visible light camera; wherein, the outlines of power equipment are marked in the historical visible light images;

[0014] Input the multiple historical visible light images into a convolutional neural network model, output a loss value, and update the model parameters of the convolutional neural network model based on the loss value;

[0015] When the loss value is within a preset loss value range or the number of training rounds reaches a preset number of training rounds, determine the convolutional neural network model with updated parameters as the power equipment recognition model.

[0016] In one embodiment, the abnormal type and the degree of thermal defect are determined by a defect recognition model based on the heat difference value between the abnormal hot spot and the remaining pixel points within a preset range where the abnormal hot spot is located.

[0017] In one embodiment, before superimposing the power equipment layer and the current infrared image, it further includes:

[0018] Perform noise reduction processing on the current infrared image and perform temperature calibration on the pixel points in the current infrared image.

[0019] In a second aspect, the present application further provides a device thermal defect determination device, and the device includes:

[0020] An acquisition module, configured to acquire a current visible light image and a current infrared image respectively captured by a visible light camera and an infrared light camera at the same position and the same shooting angle during the process of a patrol device patrolling a distribution network; wherein, the current visible light image and the current infrared image have the same size;

[0021] A first input module, configured to input the current visible light image into the power equipment recognition model and output a power equipment layer marking the outline of the power equipment;

[0022] A superimposing module, configured to superimpose the power equipment layer and the current infrared image to generate a superimposed image;

[0023] A second input module, configured to input the superimposed image into a defect recognition model, and output the position coordinates, the abnormal type, and the thermal defect degree of the corresponding abnormal hot spot of the power equipment in the current infrared image.

[0024] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the method in any one of the above embodiments are implemented.

[0025] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method in any one of the above embodiments are implemented.

[0026] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the method in any one of the above embodiments are implemented.

[0027] In the above method, device, equipment, medium, and program product for determining the thermal defect of the device, during the process of the inspection device inspecting the distribution network, the current visible light image and the current infrared image respectively taken by the visible light camera and the infrared light camera at the same position and the same shooting angle are obtained; wherein, the sizes of the current visible light image and the current infrared image are the same; the current visible light image is input into a power equipment recognition model, and a power equipment layer identifying the outline of the power equipment is output; the power equipment layer and the current infrared image are superimposed to generate a superimposed image; the superimposed image is input into a defect recognition model, and the position coordinates, the abnormal type, and the thermal defect degree of the corresponding abnormal hot spot of the power equipment in the current infrared image are output. The method provided by the present application can accurately identify the outline of the power equipment from the current infrared image, so as to accurately detect the thermal defect of the power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings required to be used in the description of the embodiments of the present application or the related art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0029] Figure 1 It is a schematic flowchart of a method for determining the thermal defect of a device in an embodiment;

[0030] Figure 2 It is a schematic flowchart of a method for training a power equipment recognition model in an embodiment;

[0031] Figure 3 It is a structural block diagram of a device thermal defect determination device in an embodiment;

[0032] Figure 4 It is an internal structure diagram of a computer device in an embodiment. Specific Embodiments

[0033] In order to make the objectives, technical solutions, and advantages of this application clearer, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely used to explain this application and are not used to limit this application.

[0034] In one embodiment, as Figure 1 shown, a method for determining device thermal defects is provided. In this embodiment, it is exemplified that this method is applied to a terminal. It can be understood that this method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0035] S102. During the process of the inspection device inspecting the distribution network, obtain the current visible light image and the current infrared image respectively taken by the visible light camera and the infrared light camera at the same position and the same shooting angle; wherein, the sizes of the current visible light image and the current infrared image are the same.

[0036] Optionally, the inspection device can be, but is not limited to, a drone.

[0037] Optionally, after obtaining the visible light image, the visible light image can be preprocessed to improve the accuracy of feature extraction. For example, the content of the preprocessing can include image denoising and grayscale processing.

[0038] Optionally, the visible light image and the infrared image are spatially aligned, that is, the coordinate systems of the two images are the same.

[0039] S104. Input the current visible light image into the power equipment recognition model, and output a power equipment layer that marks the outline of the power equipment.

[0040] Optionally, the type of the power equipment recognition model can be, but is not limited to, a convolutional neural network model.

[0041] S106. Superimpose the power equipment layer and the current infrared image to generate a superimposed image.

[0042] Optionally, image processing can be performed on the superimposed image to ensure that only the pixel points in the current infrared image within the outline of the power equipment are used for defect recognition.

[0043] S108. Input the superimposed image into the defect recognition model, and output the position coordinates, abnormal type, and thermal defect degree of the corresponding abnormal hot spots of the power equipment in the current infrared image.

[0044] Optionally, the abnormal types of the abnormal hot spots may include, but are not limited to, local temperature anomalies, temperature drifts, temperature non-uniformities, overexposure, and underexposure; the thermal defect degree refers to the degree of difference between the temperature value of the abnormal hot spot and the temperature value of the surrounding normal area.

[0045] In the above method for determining the thermal defect of the equipment, during the inspection of the power distribution network by the inspection equipment, the current visible light image and the current infrared image taken by the visible light camera and the infrared light camera respectively at the same position and the same shooting angle are obtained; wherein, the sizes of the current visible light image and the current infrared image are the same; the current visible light image is input into the power equipment recognition model, and the power equipment layer identifying the outline of the power equipment is output; the power equipment layer and the current infrared image are superimposed to generate a superimposed image; the superimposed image is input into the defect recognition model, and the position coordinates, abnormal type, and thermal defect degree of the corresponding abnormal hot spots of the power equipment in the current infrared image are output. The method provided by this application can accurately identify the outline of the power equipment from the current infrared image, so as to accurately detect the thermal defect of the power equipment.

[0046] In some embodiments, the pixel values of the pixel points located inside the outline of the power equipment in the power equipment layer are the corresponding white pixel values, and the pixel values of the pixel points located outside the outline of the power equipment are the corresponding black pixel values.

[0047] Optionally, the corresponding white pixel value is 1, and the corresponding black pixel value is 0.

[0048] In this embodiment, the pixel values of the pixel points located inside the outline of the power equipment in the power equipment layer are the corresponding white pixel values, and the pixel values of the pixel points located outside the outline of the power equipment are the corresponding black pixel values, which enables the subsequent more accurate identification of the power equipment in the infrared image.

[0049] In some embodiments, superimposing the power equipment layer and the current infrared image includes: replacing the pixel values of the pixel points located inside the outline of the power equipment in the power equipment layer with the pixel values of the pixel points at the same position in the current infrared image, and replacing the pixel values of the pixel points located outside the outline of the power equipment in the power equipment layer with preset pixel values.

[0050] Optionally, the pixel values of the pixel points in the infrared image may include, but are not limited to, representing the temperature, radiation intensity, and emissivity of the pixel points.

[0051] In this embodiment, the pixel values of the pixel points within the contour of the power equipment in the power equipment layer are replaced with the pixel values of the pixel points at the same positions in the current infrared image, and the pixel values of the pixel points outside the contour of the power equipment in the power equipment layer are replaced with preset pixel values, so that the power equipment in the infrared image can be recognized more accurately subsequently.

[0052] In some embodiments, as Figure 2 shown, the training process of the power equipment recognition model includes:

[0053] S202. Obtain multiple historical visible light images captured by a visible light camera; among them, the contours of the power equipment are marked in the historical visible light images.

[0054] S204. Input the multiple historical visible light images into the convolutional neural network model, output the loss value, and update the model parameters of the convolutional neural network model based on the loss value.

[0055] S206. When the loss value is within the preset loss value range or the number of training rounds reaches the preset number of training rounds, determine the convolutional neural network model with updated parameters as the power equipment recognition model.

[0056] Optionally, the model parameters of the convolutional neural network model may include, but are not limited to, bias terms and weight matrices.

[0057] In this embodiment, when the loss value is within the preset loss value range or the number of training rounds reaches the preset number of training rounds, determine the convolutional neural network model with updated parameters as the power equipment recognition model, so that the accuracy of the trained power equipment recognition model is higher.

[0058] In some embodiments, the abnormal type and the degree of thermal defect are determined by the defect recognition model based on the heat difference value between the abnormal hot spot and the remaining pixel points within the preset range where the abnormal hot spot is located.

[0059] In some embodiments, before superimposing the power equipment layer and the current infrared image, it further includes: performing noise reduction processing on the current infrared image and performing temperature calibration on the pixel points in the current infrared image.

[0060] Optionally, performing temperature calibration on the pixel points in the infrared image means converting the original data obtained by the infrared light camera into accurate temperature values through a series of processing steps.

[0061] In this embodiment, by performing noise reduction processing and temperature calibration on the current infrared image, the processed infrared image is more accurate, so that the contour of the power equipment recognized from the infrared image is more accurate.

[0062] In one embodiment, another method for determining thermal defects of a device is provided, and the method includes the following:

[0063] Step 1: Data acquisition

[0064] Use a visible light camera and an infrared light camera carried by a drone to simultaneously take visible light photos and infrared light photos of the power device from the same angle.

[0065] Step 2: Preprocessing of visible light images

[0066] Preprocess the acquired visible light image data, including operations such as image denoising and grayscale processing, to improve the accuracy and reliability of feature extraction.

[0067] Step 3: Training of the power device recognition model

[0068] Collect a large number of visible light images of power devices, mark the range and key components of the devices, and use a deep learning algorithm (such as the convolutional neural network CNN) to train a power device recognition model.

[0069] Step 4: Determination of device range

[0070] Apply the trained model to the new visible light photos, automatically identify and delimit the range of the power device, and generate a power device layer corresponding to the visible light photos.

[0071] Step 5: Preprocessing of infrared light images

[0072] Preprocess the acquired infrared light image data, including operations such as image denoising and temperature calibration, to ensure the image quality.

[0073] Step 6: Layer processing

[0074] Overlay the power device layer generated from the visible light photo with the infrared light photo, and through image processing technology, ensure that only the infrared image data within the device layer is used for defect recognition.

[0075] Step 7: Defect recognition

[0076] Use the defect recognition model to analyze the infrared image within the device layer. The model will identify abnormal hot spots within the layer and mark them as potential defects. Output information such as the location, type, and severity of the defects.

[0077] The above steps are all realized by computer control without manual intervention and can achieve full automation. Through this embodiment, an efficient and accurate method for infrared imaging defect recognition of power devices with visible light AI as an auxiliary means can be realized.

[0078] In the layer processing step, it is necessary to overlay the power equipment layer identified in the visible light image with the infrared light image, so as to analyze only the infrared data within the equipment area in the subsequent steps. The calculation steps of this image processing algorithm are as follows:

[0079] (1) Ensure that the visible light image and the infrared light image are spatially aligned, that is, the coordinate systems of the two images are the same.

[0080] (2) Use the power equipment recognition model to identify the equipment area in the visible light image and create a binary mask layer (Mask), where the equipment area is white (pixel value is 1) and the non-equipment area is black (pixel value is 0).

[0081] (3) Overlay the infrared light image with the mask layer, retain the infrared image data corresponding to the white area in the mask layer, and set the data in the black area to the background value or transparent.

[0082] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps or stages in other steps.

[0083] Based on the same inventive concept, the embodiments of the present application also provide a device thermal defect determination device for implementing the device thermal defect determination method involved above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the device thermal defect determination device provided below can refer to the limitations on the device thermal defect determination method in the above text, and will not be repeated here.

[0084] In an exemplary embodiment, as Figure 3 shown, a device thermal defect determination device 300 is provided, including: an acquisition module 301, a first input module 302, an overlay module 303, and a second input module 304, where:

[0085] An acquisition module 301, configured to obtain a current visible light image and a current infrared image respectively captured by a visible light camera and an infrared light camera at the same position and the same shooting angle during the process of the inspection device inspecting the distribution network; wherein, the current visible light image and the current infrared image have the same size.

[0086] A first input module 302, configured to input the current visible light image into a power equipment recognition model, and output a power equipment layer identifying the outline of the power equipment.

[0087] An overlay module 303, configured to overlay the power equipment layer and the current infrared image to generate an overlay image.

[0088] A second input module 304, configured to input the overlay image into a defect recognition model, and output the position coordinates, the abnormal type, and the thermal defect degree of the abnormal hot spots corresponding to the power equipment in the current infrared image.

[0089] In some embodiments, the first input module 302 is further configured that the pixel values of the pixel points located within the outline of the power equipment in the power equipment layer are corresponding white pixel values, and the pixel values of the pixel points located outside the outline of the power equipment are corresponding black pixel values.

[0090] In some embodiments, the overlay module 303 is further configured to replace the pixel values of the pixel points located within the outline of the power equipment in the power equipment layer with the pixel values of the pixel points at the same position in the current infrared image, and replace the pixel values of the pixel points located outside the outline of the power equipment in the power equipment layer with preset pixel values.

[0091] In some embodiments, the device thermal defect determination device 300 is specifically configured to obtain multiple historical visible light images captured by a visible light camera; wherein, the outlines of power equipment are marked in the historical visible light images; input the multiple historical visible light images into a convolutional neural network model, output a loss value, and update the model parameters of the convolutional neural network model based on the loss value; when the loss value is within a preset loss value range or the number of training rounds reaches a preset number of training rounds, determine the convolutional neural network model with updated parameters as the power equipment recognition model.

[0092] In some embodiments, the second input module 304 is further configured that the abnormal type and the thermal defect degree are determined by the defect recognition model based on the heat difference value between the abnormal hot spot and the remaining pixel points within a preset range where the abnormal hot spot is located.

[0093] In some embodiments, the device thermal defect determination device 300 is further configured to perform noise reduction processing on the current infrared image and perform temperature calibration on the pixel points in the current infrared image.

[0094] Each module in the above device thermal defect determination device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form to facilitate the processor to call and execute the operations corresponding to each of the above modules.

[0095] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 4 shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a device thermal defect determination method.

[0096] Those skilled in the art can understand that Figure 4 the structure shown in

[0097] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, the following steps are implemented: During the process of the inspection device inspecting the distribution network, obtain the current visible light image and the current infrared image captured by the visible light camera and the infrared light camera respectively at the same position and the same shooting angle; wherein, the sizes of the current visible light image and the current infrared image are the same; input the current visible light image into the power equipment recognition model, and output a power equipment layer identifying the outline of the power equipment; superimpose the power equipment layer and the current infrared image to generate a superimposed image; input the superimposed image into the defect recognition model, and output the position coordinates, the abnormal type, and the thermal defect degree of the abnormal hot spots corresponding to the power equipment in the current infrared image.

[0098] In an embodiment, the pixel values of the pixel points located within the outline of the power equipment in the power equipment layer implemented when the processor executes the computer program are the corresponding white pixel values, and the pixel values of the pixel points located outside the outline of the power equipment are the corresponding black pixel values.

[0099] In an embodiment, the superimposing of the power equipment layer and the current infrared image implemented when the processor executes the computer program includes: replacing the pixel values of the pixel points located within the outline of the power equipment in the power equipment layer with the pixel values of the pixel points at the same position in the current infrared image, and replacing the pixel values of the pixel points located outside the outline of the power equipment in the power equipment layer with preset pixel values.

[0100] In an embodiment, the training process of the power equipment recognition model implemented when the processor executes the computer program includes: obtaining multiple historical visible light images captured by the visible light camera; wherein, the outlines of the power equipment are marked in the historical visible light images; inputting the multiple historical visible light images into the convolutional neural network model, outputting a loss value, and updating the model parameters of the convolutional neural network model based on the loss value; when the loss value is within the preset loss value range or the number of training rounds reaches the preset number of training rounds, determining the convolutional neural network model with updated parameters as the power equipment recognition model.

[0101] In an embodiment, the abnormal type and the thermal defect degree are determined by the defect recognition model based on the heat difference value between the abnormal hot spot and the remaining pixel points within the preset range where the abnormal hot spot is located.

[0102] In an embodiment, before the processor executes the computer program to superimpose the power equipment layer and the current infrared image, it further includes: performing noise reduction processing on the current infrared image and performing temperature calibration on the pixel points in the current infrared image.

[0103] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: during the process of a patrol device patrolling a distribution network, obtain a current visible light image and a current infrared image respectively captured by a visible light camera and an infrared light camera at the same position and the same shooting angle; wherein, the current visible light image and the current infrared image have the same size; input the current visible light image into a power equipment recognition model, and output a power equipment layer identifying the contour of the power equipment; superimpose the power equipment layer and the current infrared image to generate a superimposed image; input the superimposed image into a defect recognition model, and output the position coordinates, the abnormal type, and the degree of thermal defect of the corresponding abnormal hot spot of the power equipment in the current infrared image.

[0104] In one embodiment, for the pixel points within the contour of the power equipment in the power equipment layer implemented when the computer program is executed by the processor, the pixel values are the corresponding white pixel values, and for the pixel points outside the contour of the power equipment, the pixel values are the corresponding black pixel values.

[0105] In one embodiment, when the computer program is executed by the processor, the superimposing of the power equipment layer and the current infrared image includes: replacing the pixel values of the pixel points within the contour of the power equipment in the power equipment layer with the pixel values of the pixel points at the same position in the current infrared image, and replacing the pixel values of the pixel points outside the contour of the power equipment in the power equipment layer with preset pixel values.

[0106] In one embodiment, the training process of the power equipment recognition model implemented when the computer program is executed by the processor includes: obtaining multiple historical visible light images captured by a visible light camera; wherein, the contours of the power equipment are marked in the historical visible light images; inputting the multiple historical visible light images into a convolutional neural network model, outputting a loss value, and updating the model parameters of the convolutional neural network model based on the loss value; when the loss value is within a preset loss value range or the number of training rounds reaches a preset number of training rounds, determining the convolutional neural network model with updated parameters as the power equipment recognition model.

[0107] In one embodiment, the abnormal type and the degree of thermal defect are determined by the defect recognition model based on the heat difference value between the abnormal hot spot and the remaining pixel points within a preset range where the abnormal hot spot is located.

[0108] In one embodiment, before superimposing the power equipment layer and the current infrared image implemented when the computer program is executed by the processor, it further includes: performing noise reduction processing on the current infrared image and performing temperature calibration on the pixel points in the current infrared image.

[0109] In one embodiment, a computer program product is provided, including a computer program which, when executed by a processor, implements the following steps: during the inspection of a distribution network by an inspection device, obtain a current visible light image and a current infrared image respectively captured by a visible light camera and an infrared light camera at the same position and the same shooting angle; wherein, the current visible light image and the current infrared image have the same size; input the current visible light image into a power equipment recognition model, and output a power equipment layer identifying the contour of the power equipment; superimpose the power equipment layer and the current infrared image to generate a superimposed image; input the superimposed image into a defect recognition model, and output the position coordinates, the abnormal type and the thermal defect degree of the corresponding abnormal hot spot of the power equipment in the current infrared image.

[0110] In one embodiment, for the pixel points within the contour of the power equipment in the power equipment layer implemented when the computer program is executed by the processor, the pixel values are the corresponding white pixel values, and for the pixel points outside the contour of the power equipment, the pixel values are the corresponding black pixel values.

[0111] In one embodiment, when the computer program is executed by the processor, the superimposing of the power equipment layer and the current infrared image includes: replacing the pixel values of the pixel points within the contour of the power equipment in the power equipment layer with the pixel values of the pixel points at the same position in the current infrared image, and replacing the pixel values of the pixel points outside the contour of the power equipment in the power equipment layer with preset pixel values.

[0112] In one embodiment, the training process of the power equipment recognition model implemented when the computer program is executed by the processor includes: obtaining multiple historical visible light images captured by a visible light camera; wherein, the contours of the power equipment are marked in the historical visible light images; inputting the multiple historical visible light images into a convolutional neural network model, outputting a loss value, and updating the model parameters of the convolutional neural network model based on the loss value; when the loss value is within a preset loss value range or the number of training rounds reaches a preset number of training rounds, determining the convolutional neural network model with updated parameters as the power equipment recognition model.

[0113] In one embodiment, the abnormal type and the thermal defect degree are determined by the defect recognition model based on the heat difference value between the abnormal hot spot and the remaining pixel points within a preset range where the abnormal hot spot is located.

[0114] In one embodiment, before superimposing the power equipment layer and the current infrared image implemented when the computer program is executed by the processor, it further includes: performing noise reduction processing on the current infrared image and performing temperature calibration on the pixel points in the current infrared image.

[0115] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0116] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0117] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, 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, it should be considered as the scope recorded in this application.

[0118] The above-described embodiments merely represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several variations and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application shall be subject to the appended claims.

Claims

1. A method for determining thermal defects of equipment, characterized in that: The method comprises: In the process of the inspection device inspecting the distribution network, a current visible light image and a current infrared image taken by a visible light camera and an infrared light camera at the same position and the same shooting angle are obtained; wherein the current visible light image and the current infrared image have the same size; Inputting the current visible light image into an electric power equipment recognition model, and outputting an electric power equipment layer that identifies the outline of the electric power equipment; Superimposing the electric power equipment layer and the current infrared image to generate a superimposed image; The superimposed image is input into a defect recognition model, and the position coordinates, abnormality type and thermal defect degree of the corresponding abnormal hot spot of the power equipment in the current infrared image are output.

2. The method according to claim 1, characterized in that: The pixel values ​​of the pixels located within the outline of the power equipment in the power equipment layer are pixel values ​​corresponding to white, and the pixel values ​​of the pixels located outside the outline of the power equipment are pixel values ​​corresponding to black.

3. The method according to claim 2, characterized in that The step of superimposing the electric power equipment layer and the current infrared image includes: The pixel values ​​of the pixels in the power equipment layer located within the outline of the power equipment are replaced with the pixel values ​​of the pixels at the same positions in the current infrared image, and the pixel values ​​of the pixels in the power equipment layer located outside the outline of the power equipment are replaced with preset pixel values.

4. The method according to claim 1, characterized in that The training process of the electric power equipment identification model includes: Acquire a plurality of historical visible light images taken by the visible light camera; wherein the outline of the power equipment is marked in the historical visible light images; Inputting the plurality of historical visible light images into a convolutional neural network model, outputting a loss value, and updating model parameters of the convolutional neural network model based on the loss value; When the loss value is within a preset loss value range or the number of training rounds reaches a preset number of training rounds, the convolutional neural network model with updated parameters is determined as the power equipment identification model.

5. The method according to claim 1, characterized in that The abnormality type and the degree of the thermal defect are determined by the defect recognition model based on a heat difference value between the abnormal hot spot and remaining pixels within a preset range where the abnormal hot spot is located.

6. The method according to claim 1, characterized in that Before superimposing the electric power equipment layer and the current infrared image, the method further includes: The current infrared image is subjected to noise reduction processing, and the pixel points in the current infrared image are subjected to temperature calibration.

7. A device for determining thermal defects of equipment, characterized in that: The device comprises: An acquisition module is used to acquire a current visible light image and a current infrared image respectively taken by a visible light camera and an infrared light camera at the same position and the same shooting angle during the inspection of the distribution network by the inspection equipment; wherein the current visible light image and the current infrared image have the same size; A first input module, used for inputting the current visible light image into the electric power equipment recognition model, and outputting an electric power equipment layer that identifies the outline of the electric power equipment; A superposition module, used for superimposing the electric power equipment layer and the current infrared image to generate a superimposed image; The second input module is used to input the superimposed image into the defect recognition model, and output the position coordinates, abnormality type and thermal defect degree of the corresponding abnormal hot spot of the power equipment in the current infrared image.

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

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

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

Images