CN116703822A - Transmission line insulation damage detection method, model training method, device and equipment - Google Patents

Abstract

The application relates to a power transmission line insulation breakage detection method, a model training method, a device, computer equipment and a storage medium. The method comprises the following steps: obtaining a visible light image and an infrared image of a power transmission line; registering the visible light image and the infrared image to obtain a registered visible light image and a registered infrared image; fusing the registration visible light image and the registration infrared image to obtain a visible infrared fused image; and inputting the visible infrared fusion image into an image segmentation model to generate a transmission line insulation breakage detection model. The method can automatically complete the positioning of the circuit and the detection of the insulation breakage.

Description

Transmission line insulation damage detection method, model training method, device and equipment

technical field

The present application relates to the technical field of transmission line insulation damage detection, in particular to a transmission line insulation damage detection method, model training method, device, computer equipment and storage medium.

Background technique

Transmission lines are a key link in power supply. If there are potential safety hazards or failures in transmission lines, it will directly affect the quality and safety of power consumption for users. Transmission lines are exposed to the natural environment for a long time and are affected by various natural factors such as wind, rain, lightning, etc., and are prone to corrosion, aging, cracks and other damages, resulting in insulation damage. Regular inspections can detect these problems in time and carry out timely maintenance to prolong the service life of transmission lines.

In the detection of line insulation defects, visible light images and infrared images have their own advantages and disadvantages. Single use of visible light or infrared images for insulation defect detection of transmission lines has a large scene limitation. In the visible light image, the positioning efficiency of the transmission line is high, but it is difficult to judge whether there is damage only through the texture, and the stability is poor. Although the infrared image can judge whether there is damage to the insulation surface through the radiation intensity of the transmission line, the texture of the infrared image is poor, and it cannot be directly used for accurate line positioning. Therefore, it is necessary to propose a method that can automatically locate the line and detect the insulation damage of the transmission line.

Contents of the invention

Based on this, it is necessary to address the above technical problems and provide a transmission line insulation damage detection method, model training method, device, computer equipment and storage medium that can automatically complete line location and transmission line insulation damage detection.

In a first aspect, the present application provides a training method for a transmission line insulation damage detection model. The method includes:

Obtain visible light images and infrared images of transmission lines;

Register the visible light image and the infrared image to obtain the registered visible light image and the registered infrared image;

Fusing the registered visible light image and the registered infrared image to obtain a visible infrared fusion image;

The visible infrared fusion image is input into the image segmentation model to generate a transmission line insulation damage detection model.

In one embodiment, registering the visible light image and the infrared image, and obtaining the registered visible light image and the registered infrared image includes:

Extract feature points of visible light images and infrared images;

Use feature descriptors to match feature points in visible light images and infrared images;

The visible light image and the infrared image are registered according to the result of feature point matching, and the registered visible light image and the registered infrared image are obtained.

In one embodiment, inputting the visible-infrared fusion image into the image segmentation model to generate a transmission line insulation damage detection model includes:

Input the visible infrared fusion image into the encoder to obtain the feature information of the visible infrared fusion image;

Input the feature information into the decoder to obtain the segmentation result of the visible infrared fusion image;

Use the cross-entropy loss function to obtain the difference value between the segmentation result and the real label based on the segmentation result;

The parameters of the image segmentation model are updated according to the difference value to generate a transmission line insulation damage detection model.

In one embodiment, inputting the visible-infrared fusion image into the encoder, and obtaining the feature information of the visible-infrared fusion image also includes:

Each encoder transmits feature information through a designated module.

In one embodiment, the feature information is input into the decoder, and the segmentation result of obtaining the visible-infrared fusion image also includes:

The feature information is transferred between the encoder and decoder through a skip connection module.

In the second aspect, the present application also provides a method for detecting insulation damage of a transmission line, which uses the detection model for insulation damage of a transmission line as provided in the first aspect. The method includes:

Get a live image set;

Call the transmission line insulation damage detection model, input the real-time image set into the transmission line insulation damage detection model, and obtain the location information of the insulation damage.

In a third aspect, the present application also provides a training device for a transmission line insulation damage detection model.

The unit includes:

An image acquisition module, configured to acquire visible light images and infrared images of power transmission lines;

The image registration module is used to register the visible light image and the infrared image, and obtain the registered visible light image and the registered infrared image;

The image fusion module is used to fuse the registered visible light image and the registered infrared image to obtain a visible infrared fusion image;

The model generation module is used for inputting the visible infrared fusion image into the image segmentation model to generate a transmission line insulation damage detection model.

In a fourth aspect, the present application provides a device for detecting insulation damage of a transmission line. The unit includes:

A real-time image acquisition module is used to obtain a real-time image set;

The damage information acquisition module is used to call the transmission line insulation damage detection model, input the real-time image set into the transmission line insulation damage detection model, and obtain the location information of the insulation damage.

In a fifth aspect, the present application also provides a computer device. The computer equipment includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned training method for the transmission line insulation damage detection model are realized; or, the above-mentioned transmission line insulation damage detection method is realized. A step of.

In a sixth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by the processor, the steps of the above-mentioned training method for the transmission line insulation damage detection model are realized; or, the steps of the above-mentioned transmission line insulation damage detection method are realized. .

The above transmission line insulation damage detection method, model training method, device, computer equipment and storage medium, by obtaining the visible light image and the infrared image of the transmission line, register the visible light image and the infrared image, and obtain the registered visible light image and the registered infrared image Image, the registered visible light image and the registered infrared image are fused to obtain a visible infrared fusion image, and the visible infrared fusion image is input into the image segmentation model to generate a transmission line insulation damage detection model. Make full use of and combine the advantages of visible light images and infrared images, and then obtain the real-time image set, call the transmission line insulation damage detection model, input the real-time image set into the transmission line insulation damage detection model, and obtain the location information of the insulation damage. The transmission line with abnormal temperature is segmented through the transmission line insulation damage detection model, so as to realize the automatic location and detection of insulation damage, which helps to improve the safety and reliability of the transmission line.

Description of drawings

Fig. 1 is a schematic flow chart of a training method of a transmission line insulation damage detection model in an embodiment;

Fig. 2 is a module schematic diagram of the training method of the transmission line insulation damage detection model in an embodiment;

FIG. 3 is a schematic flow diagram of a method for detecting insulation damage of a transmission line in another embodiment;

Fig. 4 is a device schematic diagram of a training device for a transmission line insulation damage detection model in an embodiment;

Figure 5 is an internal block diagram of a computer device in one embodiment.

Detailed ways

Transmission lines are a key link in power supply. If there are potential safety hazards or failures in transmission lines, it will directly affect the quality and safety of power consumption for users. Transmission line inspection refers to the regular or irregular inspection, detection, monitoring and maintenance of transmission lines in the process of power transmission, so as to ensure the safe and reliable operation of transmission lines. Through regular inspections, various hidden dangers of line failures can be discovered and eliminated in time to ensure the safety and reliability of power supply. Transmission lines are exposed to the natural environment for a long time and are affected by various natural factors such as wind, rain, lightning, etc., and are prone to corrosion, aging, cracks and other damages, resulting in insulation damage. Regular inspections can detect these problems in time and carry out timely maintenance to prolong the service life of transmission lines. At the same time, line inspection improves the reliability of the power grid. Regular inspection can detect and deal with line faults in time, reduce the probability of failure, and improve the reliability of the power grid. Timely transmission line defect detection can detect and deal with line faults in time, reducing power accidents and losses caused by line faults. Especially in the case of natural disasters and severe weather, regular inspections can reduce the risk of the line being damaged or damaged.

In the detection of line insulation defects, visible light images and infrared images have their own advantages and disadvantages. Visible light images are images obtained through the visible light frequency band. Its main advantage is high resolution, which can clearly display the shape, color and surface details of the target object. Visible light cameras are low cost and easy to operate, so they are widely used in industry and life. In the detection of line insulation defects, visible light images can be used to detect the size, shape, color and other information of defects, as well as dirt and damage. However, visible light images perform poorly at night or in low-light environments because visible light cannot penetrate smoke, haze, dust, etc. In addition, visible light has a narrow wavelength range, which limits the types of defects it can detect.

In contrast, an infrared image is an image obtained through infrared radiation, and its main advantage is that it has a thermal imaging function and can detect temperature differences on the surface of an object. Therefore, infrared images can be used to detect thermal damage, corrosion, aging, etc. of insulating materials, which are difficult to detect in visible light images. In addition, infrared images have strong penetrating power and can penetrate smog, smoke, etc., so they can also perform well in low-light environments. However, the resolution of infrared images is relatively low and cannot clearly show the morphology and surface details of target objects like visible light images. In addition, infrared cameras are expensive and require professional technicians to operate and maintain, so their use is relatively limited. Therefore, in the detection of line insulation defects, visible light images and infrared images can complement each other. Visible light images can be used to detect the size, shape, color and other information of defects, and infrared images can be used to detect thermal damage, aging, etc. At the same time, you can choose which image to use according to actual needs.

In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.

In one embodiment, as shown in FIG. 1 , a training method for a transmission line insulation damage detection model is provided. In this embodiment, the method is applied to a terminal for example. It can be understood that this method can also be applied to The server may also be applied to a system including a terminal and a server, and may be implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

Step 102, acquiring visible light images and infrared images of the transmission line.

Specifically, the visible light image and infrared image of the transmission line are obtained to prepare data for the insulation damage detection model of the transmission line.

Among them, drones or other equipment are used to collect visible light images and infrared images of power transmission lines. Optionally, devices such as digital cameras or infrared thermal imagers are usually used. Furthermore, the collected visible light images and infrared images need to be preliminarily processed, such as removing noise, adjusting image contrast and brightness, and so on. The data prepared for the transmission line insulation damage detection model should include the image of the transmission line and the corresponding label image, where the label image should include the mark of the defect area.

Step 104, registering the visible light image and the infrared image to obtain a registered visible light image and a registered infrared image.

Among them, the visible light image and the infrared image are preprocessed, including noise removal, smoothing, enhancement and contrast adjustment, etc., and then the preprocessed visible light image and the infrared image are registered to obtain the registered visible light image and the registered infrared image.

Step 106, fusing the aligned visible light image and the aligned infrared image to obtain a visible-infrared fused image.

Among them, the registered visible light image and the registered infrared image are fused, and techniques such as weighted average or transform domain fusion are usually used to obtain a visible infrared fused image.

Step 108, input the visible infrared fusion image into the image segmentation model to generate a transmission line insulation damage detection model.

Specifically, the image segmentation model includes TransUNet, which is a Transformer-based image segmentation model. Input the visible infrared fusion image into TransUNet for training to generate a transmission line insulation damage detection model, and then evaluate the transmission line insulation damage detection model. After the training and evaluation are completed, the model can be applied to the actual transmission line insulation damage defect segmentation task middle.

Specifically, after the model training is completed, the model needs to be evaluated. Commonly used evaluation indicators include precision, recall, F1 score and IoU (Intersection over Union) score, and the test data set can be used for model evaluation.

In the above-mentioned training method of the transmission line insulation damage detection model, the visible light image and the infrared image of the transmission line are obtained, and after the visible light image and the infrared image are registered, the registered visible light image and the registered infrared image are obtained, and the registered visible light image and the infrared image are registered. Register the infrared images for fusion to obtain an image with richer comprehensive information, which is the visible infrared fusion image. The comprehensive information includes information contained in visible light images and infrared images, including the shape, color, surface details and temperature differences of the target. The visible infrared fusion image is input into the image segmentation model to generate a transmission line insulation damage detection model. Information fusion of infrared images and visible light images by image registration and fusion methods can obtain a spatial distribution image that simultaneously reflects scene texture and thermal radiation. Combining the advantages of visible light images and infrared images, and pioneering the combination of visible light images and infrared images to generate a transmission line insulation damage detection model, so as to realize the automatic positioning and detection of transmission line insulation damage.

In one embodiment, registering the visible light image and the infrared image, and obtaining the registered visible light image and the registered infrared image includes:

Extract the feature points of the visible light image and the infrared image; use the feature descriptor to match the feature points in the visible light image and the infrared image; register the visible light image and the infrared image according to the matching results of the feature points, and obtain the registered visible light image and registration quasi-infrared images.

Specifically, after preprocessing the visible light image and the infrared image, feature points in the visible light image and the infrared image, such as corner points, edges, and spots, are extracted. Use feature descriptors to match feature points in infrared and visible light images, such as SIFT (Scale-invariant feature transform, scale-invariant feature transform) and SURF (Speeded Up Robust Features, accelerated robust features) and other algorithms.

Further, according to the feature matching result, the infrared image and the visible light image are registered, so that the infrared image and the visible light image are aligned in terms of geometric transformation, and the registered visible light image and the registered infrared image are obtained. The registration method can use a method based on feature points or an image-based method, such as algorithms such as ICP (Iterative Closest Point, iterative closest point) and Thin-plate-spline (thin plate spline).

Further, the registered visible light image and the registered infrared image are fused to obtain a visible-infrared fusion image. Commonly used fusion methods include pixel-level fusion and feature-level fusion. Among them, pixel-level fusion methods include simple average, weighted average, and Laplacian pyramid, etc., while feature-level fusion methods can use methods such as multi-resolution decomposition or wavelet transform. Data enhancement, data normalization and data cropping are performed on the registered and fused images. Among them, data enhancement can be carried out by means of rotation, flipping and scaling, so as to increase the diversity of data samples. Data normalization can scale the image pixel values to the range of 0-1 to facilitate network learning. Data cropping can crop images and labels to the same size respectively for network training.

In this embodiment, by extracting the feature points of the visible light image and the infrared image, using the feature descriptor to match the feature points in the visible light image and the infrared image, and registering the visible light image and the infrared image according to the matching result of the feature points, to obtain Register visible light images and register infrared images. Since the imaging methods and parameters of visible light images and infrared images are different, it is necessary to register the two images so that their pixels correspond in a physical sense. Further, preparations are made for fusing the registered visible light image and the registered infrared image to obtain a visible-infrared fusion image.

In one embodiment, inputting the visible-infrared fusion image into the image segmentation model to generate a transmission line insulation damage detection model includes:

Input the visible-infrared fusion image into the encoder to obtain the feature information of the visible-infrared fusion image; input the feature information into the decoder to obtain the segmentation result of the visible-infrared fusion image; use the cross-entropy loss function to obtain the distance between the segmentation result and the real label based on the segmentation result The difference value; according to the difference value, the parameters of the image segmentation model are updated to generate a transmission line insulation damage detection model.

Among them, TransUNet is selected as the image segmentation model, and the overall structure of TransUNet consists of two parts, namely encoder and decoder. The encoder adopts the structure of Transformer, which is used to extract the feature information of the visible infrared fusion image. The decoder is a U-Net structure, which is used to convert the feature information of the visible infrared fusion image extracted by the encoder into a segmentation result.

Specifically, the encoder consists of a series of Transformer Encoders (self-attention mechanism encoders), and each Encoder includes a multi-head self-attention mechanism (multi-head self-attention) and a feedforward network (feedforward network). The decoder adopts the U-Net structure and consists of a series of upsampling modules and convolution modules.

Further, the loss function used by TransUNet is the cross-entropy loss function, which is used to measure the difference between the segmentation results generated by the model and the real labels. Specifically, given an input image I, the segmentation result generated by the model is The real label is Y _ij , then the cross-entropy loss function can be defined as in formula (1):

Among them, N represents the number of pixels, and C represents the number of categories.

Further, the parameters of the image segmentation model are updated based on the difference between the segmentation result obtained by the cross-entropy loss function and the real label, and then the updated parameters are imported into the image segmentation model to generate a transmission line insulation damage detection model.

In this embodiment, TransUNet, as a Transformer-based image segmentation model, can process feature information of different scales at the same time, and has achieved excellent performance on multiple data sets. Input the visible infrared fusion image into the encoder to obtain the characteristic information of the visible infrared fusion image. The encoder consists of a series of self-attention mechanism encoders, which can effectively extract the features of the input image. The feature information is input into the decoder to obtain the segmentation result of the visible infrared fusion image. Among them, the encoder consists of a series of upsampling modules and convolution modules, which can restore the feature information extracted by the encoder to the size of the original image and generate segmentation results. Use the cross-entropy loss function to obtain the difference between the segmentation result and the real label based on the segmentation result. This loss function can help the model learn more accurate segmentation results during the training process. The parameters of the image segmentation model are updated according to the difference value to generate a transmission line insulation damage detection model, which is the key to automatic location and detection of insulation damage.

In one embodiment, inputting the visible-infrared fusion image into the encoder, and obtaining the feature information of the visible-infrared fusion image also includes:

Each encoder transmits feature information through a designated module.

Among them, the specified module is Transformer Inter-Block.

Specifically, the encoder is composed of a series of Transformer Encoders (self-attention mechanism encoders). Between each Encoder, TransUNet also adds a new module - Transformer Inter-Block, which is used for multiple scales. transfer feature information. Among them, multiple scales refer to signal sampling at different scales, and different features can be observed at different scales.

In this embodiment, each encoder transmits feature information through a designated module, so as to ensure that the model can process feature information of different scales.

In one embodiment, the feature information is input into the decoder, and the segmentation result of obtaining the visible-infrared fusion image also includes:

The feature information is transferred between the encoder and decoder through a skip connection module.

Specifically, a new module, the Skip Connection TransformerBlock, is added to the decoder, which is the skip connection module, which is used to transfer feature information between the encoder and the decoder.

In this embodiment, the feature information is transferred between the encoder and the decoder through the skip connection module, so as to better preserve the detail information.

In one embodiment, a method for detecting insulation damage of a transmission line is provided. The method for detecting insulation damage of a transmission line uses a detection model for a transmission line insulation damage. The method includes:

Obtain a real-time image set; call the transmission line insulation damage detection model, input the real-time image set into the transmission line insulation damage detection model, and obtain the location information of the insulation damage.

Specifically, after the training and evaluation of the transmission line insulation damage detection model is completed, the model can be applied to the actual transmission line insulation damage defect segmentation task.

In this embodiment, by acquiring the real-time image set, calling the transmission line insulation damage detection model, inputting the real-time image set into the transmission line insulation damage detection model, and obtaining the location information of the insulation damage, the transmission line image can be segmented to effectively detect and Locating insulation damage defects helps to improve the safety and reliability of transmission lines.

In one embodiment, as shown in FIG. 2 , a block diagram of a training method for a transmission line insulation damage detection model is provided. First, drones or other devices are used to collect visible light images and infrared images of the transmission line, and the mask marking of the damaged insulation lines contained in the visible light images and infrared images is performed. Then the visible light image and the infrared image are registered to obtain the registered visible light image and the registered infrared image. The registered visible light image and the registered infrared image are fused to obtain a visible infrared fused image. The TransUNet model is trained based on visible-infrared fusion images to generate a transmission line insulation damage detection model. Evaluate whether the insulation damage detection model of the transmission line meets the accuracy requirements. If it does not meet the accuracy requirements, continue to train the TransUNet model; if it meets the accuracy requirements, it will be put into transmission line inspection.

In another embodiment, as shown in FIG. 3 , a method for detecting insulation damage of a transmission line is provided.

Step 302, acquiring visible light images and infrared images of the transmission line.

Step 304, extracting feature points of the visible light image and the infrared image.

Step 306, using the feature descriptor to match the feature points in the visible light image and the infrared image.

Step 308, registering the visible light image and the infrared image according to the matching result of the feature points, to obtain the registered visible light image and the registered infrared image.

Step 310, input the visible-infrared fusion image into the encoder to obtain feature information of the visible-infrared fusion image.

In step 312, each encoder transmits feature information through a designated module.

Step 314, input the feature information into the decoder, and obtain the segmentation result of the visible-infrared fusion image.

Step 316, transfer feature information between the encoder and the decoder through the skip connection module.

Step 318, using a cross-entropy loss function to obtain a difference value between the segmentation result and the real label based on the segmentation result.

Step 320, updating the parameters of the image segmentation model according to the difference value to generate a transmission line insulation damage detection model.

Step 322, acquiring a real-time image set.

Step 324, call the insulation damage detection model of the transmission line, input the real-time image set into the insulation damage detection model of the transmission line, and obtain the position information of the insulation damage.

In this embodiment, the infrared image and the visible light image are registered and fused, and the obtained visible-infrared fused image not only reflects the texture difference between the transmission line and the background, but also reflects the heat increase at the position where the insulation of the transmission line is damaged. Then the TransUnet model is trained based on the visible-infrared fusion image to generate a transmission line insulation damage detection model. The model can locate the transmission line from a complex background and at the same time complete the segmentation of the defective part, thus helping the inspectors to locate it efficiently.

It should be understood that although the steps in the flow charts involved in the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flow charts involved in the above-mentioned embodiments may include multiple steps or stages, and these steps or stages are not necessarily executed at the same time, but may be performed at different times For execution, the execution order of these steps or stages is not necessarily performed sequentially, but may be executed in turn or alternately with other steps or at least a part of steps or stages in other steps.

Based on the same inventive concept, an embodiment of the present application further provides a training device for a transmission line insulation damage detection model for implementing the above-mentioned training method for a transmission line insulation damage detection model. The solution to the problem provided by the device is similar to the implementation described in the above method, so the specific limitations in the embodiment of the training device for one or more transmission line insulation damage detection models provided below can be referred to above for The limitation of the training method of the transmission line insulation damage detection model will not be repeated here.

In one embodiment, as shown in FIG. 4 , a training device for a transmission line insulation damage detection model is provided, including: an image acquisition module 402, an image registration module 404, an image fusion module 406, and a model generation module 408

The image acquisition module 402 is configured to acquire visible light images and infrared images of the transmission line.

The image registration module 404 is configured to register the visible light image and the infrared image, and obtain the registered visible light image and the registered infrared image.

The image fusion module 406 is configured to fuse the registered visible light image and the registered infrared image to obtain a visible infrared fusion image.

The model generation module 408 is configured to input the visible infrared fusion image into the image segmentation model to generate a transmission line insulation damage detection model.

In one embodiment, the device also includes:

The feature point extraction module is used to extract feature points of visible light images and infrared images.

The feature point matching module is used to match the feature points in the visible light image and the infrared image by using the feature descriptor.

The registered image acquisition module is configured to register the visible light image and the infrared image according to the result of feature point matching, and obtain the registered visible light image and the registered infrared image.

In one embodiment, the device also includes:

The encoder input module is configured to input the visible-infrared fusion image into the encoder to obtain feature information of the visible-infrared fusion image.

The decoder input module is used to input feature information into the decoder to obtain the segmentation result of the visible infrared fusion image.

The difference value acquisition module is used to obtain the difference value between the segmentation result and the real label based on the segmentation result using a cross-entropy loss function.

The parameter update module is used to update the parameters of the image segmentation model according to the difference value, and generate a transmission line insulation damage detection model.

In one embodiment, the device also includes:

The feature information transfer module is used to transfer feature information between each encoder through a specified module; it is also used to transfer feature information between an encoder and a decoder through a skip connection module.

Each module in the training device for the above-mentioned transmission line insulation damage detection model can be fully or partially realized by software, hardware and combinations thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.

In one embodiment, a transmission line insulation damage detection device is provided, including: a real-time image acquisition module and a damage information acquisition module, wherein:

A real-time image acquisition module is used to obtain a real-time image set;

The damage information acquisition module is used to call the transmission line insulation damage detection model, input the real-time image set into the transmission line insulation damage detection model, and obtain the location information of the insulation damage.

Each module in the above transmission line insulation damage detection device can be fully or partially realized by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules.

In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure may be as shown in FIG. 5 . The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit and an input device. Wherein, the processor, the memory and the input/output interface are connected through the system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein, the processor of the computer device is used to provide calculation 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 computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input/output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, mobile cellular network, NFC (Near Field Communication) or other technologies. When the computer program is executed by a processor, a training method for a transmission line insulation damage detection model and a power line insulation damage detection method are realized. The display unit of the computer equipment is used to form a visually visible picture, which may be a display screen, a projection device or a virtual reality imaging device. The display screen may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a button, a trackball or a touch pad set on the casing of the computer device, or a External keyboard, touchpad or mouse etc.

Those skilled in the art can understand that the structure shown in Figure 5 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation to the computer equipment on which the solution of the application is applied. The specific computer equipment may include There may be more or fewer components than shown in the figures, or certain components may be combined, or have different component arrangements.

In one embodiment, there is also provided a computer device, including a memory and a processor, where a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the foregoing method embodiments are implemented.

It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all It is information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above-mentioned embodiments can be completed by instructing related hardware through computer programs, and the computer programs can be stored in a non-volatile computer-readable memory In the medium, when the computer program is executed, it may include the processes of the embodiments of the above-mentioned methods. Wherein, any reference to storage, database or other media used in the various embodiments provided in the present application may include at least one of non-volatile and volatile storage. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive variable memory (ReRAM), magnetic variable memory (Magnetoresistive Random Access Memory, MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (Phase Change Memory, PCM), graphene memory, etc. The volatile memory may include random access memory (Random Access Memory, RAM) or external cache memory. As an illustration and not a limitation, RAM can be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (Dynamic Random Access Memory, DRAM). The databases involved in the various embodiments provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a blockchain-based distributed database, etc., but is not limited thereto. The processors involved in the various embodiments provided by this application can be general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, data processing logic devices based on quantum computing, etc., and are not limited to this.

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

The above-mentioned embodiments only express several implementation modes of the present application, and the description thereof is relatively specific and detailed, but should not be construed as limiting the patent scope of the present application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be determined by the appended claims.

Claims (10)

1. A training method of a transmission line insulation damage detection model, characterized in that, the method comprises: Obtain visible light images and infrared images of transmission lines; Registering the visible light image and the infrared image to obtain a registered visible light image and a registered infrared image; Fusing the registered visible light image and the registered infrared image to obtain a visible infrared fusion image; The visible-infrared fusion image is input into an image segmentation model to generate a transmission line insulation damage detection model. 2. The method according to claim 1, wherein said registering said visible light image and infrared image, and obtaining a registered visible light image and a registered infrared image comprises: Extract feature points of visible light images and infrared images; matching feature points in the visible light image and the infrared image using a feature descriptor; The visible light image and the infrared image are registered according to the matching result of the feature points to obtain a registered visible light image and a registered infrared image. 3. The method according to claim 1, wherein said inputting said visible infrared fusion image into an image segmentation model to generate a transmission line insulation damage detection model comprises: Inputting the visible-infrared fusion image into an encoder to obtain feature information of the visible-infrared fusion image; The feature information is input into a decoder to obtain the segmentation result of the visible infrared fusion image; Using a cross-entropy loss function to obtain a difference value between the segmentation result and the real label based on the segmentation result; The parameters of the image segmentation model are updated according to the difference value to generate a transmission line insulation damage detection model. 4. The method according to claim 3, wherein the inputting the visible-infrared fusion image into an encoder and obtaining the feature information of the visible-infrared fusion image further comprises: Each encoder transmits feature information through a designated module. 5. The method according to claim 3, wherein said inputting said characteristic information into a decoder and obtaining the segmentation result of a visible-infrared fusion image further comprises: Feature information is transferred between the encoder and decoder through a skip connection module. 6. A transmission line insulation damage detection method, characterized in that the transmission line insulation damage detection method uses the transmission line insulation damage detection model as claimed in claim 1, and the method comprises: Get a live image set; Invoking a transmission line insulation damage detection model, inputting the real-time image set into the transmission line insulation damage detection model, and obtaining insulation damage location information. 7. A training device for a transmission line insulation damage detection model, characterized in that the device comprises: An image acquisition module, configured to acquire visible light images and infrared images of power transmission lines; An image registration module, configured to register the visible light image and the infrared image to obtain a registered visible light image and a registered infrared image; An image fusion module, configured to fuse the registered visible light image and the registered infrared image to obtain a visible infrared fusion image; The model generation module is used to input the visible infrared fusion image into the image segmentation model to generate a transmission line insulation damage detection model. 8. A transmission line insulation damage detection device, characterized in that the device comprises: A real-time image acquisition module is used to obtain a real-time image set; The damage information acquisition module is configured to invoke a power transmission line insulation damage detection model, input the real-time image set into the power transmission line insulation damage detection model, and obtain location information of insulation damage. 9. A computer device, comprising a memory and a processor, the memory stores a computer program, wherein the processor implements the method according to any one of claims 1 to 6 when executing the computer program step. 10. A computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are realized.