Image enhancement method and device, computer equipment and readable storage medium

Abstract

The invention relates to an image enhancement method and device, computer equipment and a readable storage medium. The method comprises the steps of obtaining an original power transmission inspection image of a target power transmission line; extracting image features of the original power transmission inspection image by using a preset image enhancement model, and performing fuzzy quantization on the extracted image features by using a membership function to obtain fuzzy feature information corresponding to the original power transmission inspection image; performing integration processing on the fuzzy feature information corresponding to the original power transmission inspection image to obtain multi-layer advanced feature information corresponding to the original power transmission inspection image; the advanced feature information represents key attributes of the target power transmission line; and performing image enhancement processing on the original power transmission inspection image based on the multi-layer advanced feature information to obtain an image enhancement result of the original power transmission inspection image. By adopting the method, the quality of the enhanced image can be improved.

Description

Technical Field

[0001] This application relates to the field of power transmission technology, and in particular to an image enhancement method, apparatus, computer device, and readable storage medium. Background Technology

[0002] As the scale of power systems continues to expand, the operating environment of target transmission lines is becoming increasingly complex, making their inspection work crucial for ensuring the safe operation of the power grid. During the inspection process, transmission line inspection images can be collected, and then the operational safety of the target transmission line can be determined by identifying these images.

[0003] Power transmission line inspection images are often acquired manually or using rudimentary automated imaging equipment. This equipment is susceptible to external factors such as weather, lighting, and angle, frequently resulting in blurry images, insufficient contrast, and severe noise interference. Related technologies primarily employ image enhancement techniques based on convolutional neural networks or generative adversarial networks to process power transmission line inspection images.

[0004] However, the methods used in these techniques often result in poor quality of the enhanced images. Summary of the Invention

[0005] Therefore, it is necessary to provide an image enhancement method, apparatus, computer device, and readable storage medium to address the aforementioned technical problems, which can improve the quality of enhanced images.

[0006] In a first aspect, this application provides an image enhancement method, comprising:

[0007] Acquire raw transmission line inspection images of the target transmission line;

[0008] The image features of the original power transmission inspection image are extracted using a preset image enhancement model, and the extracted image features are fuzzy quantized using a membership function to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0009] The fuzzy feature information corresponding to the original power transmission inspection image is integrated and processed to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image; the high-level feature information characterizes the key attributes of the target power transmission line.

[0010] Image enhancement processing is performed on the original power transmission inspection image based on multi-layer high-level feature information to obtain the image enhancement result of the original power transmission inspection image.

[0011] In one embodiment, the extracted image features are fuzzy quantized using a membership function to obtain fuzzy feature information corresponding to the original power transmission inspection image, including:

[0012] Membership functions are used to map the extracted image features to determine the fuzzy membership value of each pixel in the original power transmission inspection image.

[0013] Based on the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy feature information corresponding to the original power transmission inspection image is obtained.

[0014] In one embodiment, the image features include at least color features, texture features, brightness features, and edge features;

[0015] Membership functions are used to map the extracted image features to determine the fuzzy membership value of each pixel in the original power transmission inspection image, including:

[0016] The RGB values ​​of each pixel in the original power transmission inspection image are blurred using membership functions to determine the color blur membership value of each pixel. Different texture regions in the original power transmission inspection image are blurred using membership functions to determine the texture blur membership value of each pixel. The brightness values ​​of each pixel in the original power transmission inspection image are blurred using membership functions to determine the brightness blur membership value of each pixel. Finally, the edge values ​​of edge pixels in the original power transmission inspection image are blurred using membership functions to determine the edge blur membership value of the edge pixels.

[0017] The fuzzy membership values ​​of each color, each texture, each brightness, and each edge are used as the fuzzy membership values ​​of each pixel in the original power transmission inspection image.

[0018] In one embodiment, based on the fuzzy membership value of each pixel in the original power transmission inspection image, fuzzy feature information corresponding to the original power transmission inspection image is obtained, including:

[0019] Based on the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image are weighted to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0020] In one embodiment, the method for obtaining the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image includes:

[0021] Calculate the activation degree of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image; the activation degree is used to describe the joint strength between the pixel and other feature types.

[0022] Given that the activation degree of each feature type meets the preset conditions, the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image is assigned based on the minimum activation degree among all activation degrees.

[0023] In one embodiment, the fuzzy feature information corresponding to the original power transmission inspection image is integrated to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image, including:

[0024] Redundant fuzzy feature information is filtered out from the fuzzy feature information corresponding to the original power transmission line inspection image to obtain the target fuzzy feature information related to the target power transmission line;

[0025] By integrating the target fuzzy feature information related to the target transmission line, multi-layer high-level feature information corresponding to the original transmission line inspection image is obtained.

[0026] In one embodiment, image enhancement processing is performed on the original power transmission inspection image based on multi-layer high-level feature information to obtain the image enhancement result of the original power transmission inspection image, including:

[0027] Image enhancement processing is performed on the original power transmission inspection image based on multi-layer high-level feature information to generate an enhanced original power transmission inspection image.

[0028] The enhanced original power transmission inspection image is subjected to detail enhancement and / or brightness adjustment to obtain the image enhancement result of the original power transmission inspection image.

[0029] Secondly, this application also provides an image enhancement apparatus, comprising:

[0030] The acquisition module is used to acquire the original transmission line inspection images of the target transmission line;

[0031] The quantization module is used to extract image features from the original power transmission inspection image using a preset image enhancement model, and to perform fuzzy quantization on the extracted image features using a membership function to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0032] The processing module is used to integrate and process the fuzzy feature information corresponding to the original power transmission inspection images to obtain multi-layer high-level feature information corresponding to the original power transmission inspection images; the high-level feature information characterizes the key attributes of the target power transmission line.

[0033] The enhancement module is used to perform image enhancement processing on the original power transmission inspection image based on multi-layer high-level feature information to obtain the image enhancement result of the original power transmission inspection image.

[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the content of any embodiment of the image enhancement method in the first aspect described above.

[0035] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the content of any embodiment of the image enhancement method in the first aspect described above.

[0036] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the content of any embodiment of the image enhancement method in the first aspect described above.

[0037] The aforementioned image enhancement method, apparatus, computer equipment, and readable storage medium acquire original transmission line inspection images of the target transmission line; extract image features from the original transmission line inspection images using a preset image enhancement model, and perform fuzzy quantization on the extracted image features using a membership function to obtain fuzzy feature information corresponding to the original transmission line inspection images; integrate and process the fuzzy feature information corresponding to the original transmission line inspection images to obtain multi-layer high-level feature information corresponding to the original transmission line inspection images; the high-level feature information characterizes the key attributes of the target transmission line; and perform image enhancement processing on the original transmission line inspection images based on the multi-layer high-level feature information to obtain the image enhancement result of the original transmission line inspection images. This method extracts blurred features from the original power transmission inspection image through membership functions and, combined with an integrated processing approach, extracts deeper, high-level feature information. Through the enhancement process of multi-layer high-level feature information, it can significantly improve the clarity and detail fidelity of the original power transmission inspection image, effectively cope with working conditions such as changes in lighting, noise interference, and complex backgrounds, and make the enhanced power transmission inspection image closer to the real scene in terms of structure, texture, and hierarchy, thereby improving the quality of the enhanced power transmission inspection image. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a diagram illustrating the application environment of an image enhancement method in one embodiment;

[0040] Figure 2 This is a flowchart illustrating an image enhancement method in one embodiment;

[0041] Figure 3 This is a flowchart illustrating an image enhancement method in one embodiment;

[0042] Figure 4 This is a flowchart illustrating an image enhancement method in one embodiment;

[0043] Figure 5 This is a flowchart illustrating an image enhancement method in one embodiment;

[0044] Figure 6 This is a flowchart illustrating an image enhancement method in one embodiment;

[0045] Figure 7 This is a flowchart illustrating an image enhancement method in one embodiment;

[0046] Figure 8 This is a flowchart illustrating an image enhancement method in one embodiment;

[0047] Figure 9 This is a structural block diagram of an image enhancement device in one embodiment. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0049] Before providing a detailed description of the technical solution of this application, a brief explanation of the background technology of this application will be given first.

[0050] As the scale of power systems continues to expand, the operating environment of target transmission lines is becoming increasingly complex, making their inspection work crucial for ensuring the safe operation of the power grid. During the inspection process, transmission line inspection images can be collected, and then the operational safety of the target transmission line can be determined by identifying these images.

[0051] The acquisition of images for power transmission inspections often relies on manual shooting or low-level automatic imaging equipment. Imaging equipment is easily affected by external factors such as weather, lighting, and angle, which often leads to problems such as blurry images, insufficient contrast, and severe noise interference.

[0052] In related technologies, image enhancement processing of power transmission inspection images is mainly performed using convolutional neural networks or generative adversarial networks. However, these models often suffer from insufficient feature extraction and weak generalization ability when faced with uncertainties in fuzzy features, nonlinear distribution characteristics, and complex noise scenes. Furthermore, these models lack semantic understanding of fuzzy features, making it difficult to accurately represent phenomena such as image blurriness and unclear edges, resulting in enhanced images that still fall short in terms of clarity and realism.

[0053] To address the aforementioned problems, this application provides an image enhancement method, apparatus, computer device, and readable storage medium. This method can more accurately perform semantic understanding of blurred features and can specifically address issues such as image blurriness and unclear edges, resulting in better quality enhanced power transmission inspection images. The specific details of the image enhancement method will be described below.

[0054] The image enhancement method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown is as follows. For example, the computer device can be a server, personal computer, laptop, smartphone, tablet, mobile phone, etc. The computer device may include a processor, memory, and network interface connected via a system bus or wirelessly. The processor provides computing and control capabilities. The memory may include non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores data during the image enhancement process. The network interface communicates with external terminals via a network connection, and the computer program is executed by the processor to implement an image enhancement method. The computer device can be implemented using a standalone computer device or a cluster of multiple computer devices. It should be noted that the memory of the computer device is not limited to the above-mentioned memory and may also include high-speed random access memory, volatile solid-state memory, etc. Furthermore, the architecture of the computer device is not limited to the above-described cases; some components may be added or omitted.

[0055] In one exemplary embodiment, such as Figure 2 As shown, an image enhancement method is provided, which is applied to... Figure 1 Taking a computer device as an example, the explanation includes the following steps S101 to S104. Wherein:

[0056] S101, acquire the original transmission line inspection image of the target transmission line.

[0057] In this embodiment of the application, the original power transmission inspection images of multiple power transmission lines are stored in an image database. The computer device can search for power transmission information that is the same as the identification information of the target power transmission line from the image database, and use the original power transmission inspection image of the target power transmission line as the original power transmission inspection image of the target power transmission line.

[0058] Alternatively, the computer equipment can send flight commands to a drone equipped with image acquisition devices, and during the drone's flight, control the image acquisition devices to acquire images of the target power transmission line, thereby obtaining the original power transmission inspection images of the target power transmission line. This application embodiment does not limit the method of obtaining the original power transmission inspection images of the target power transmission line.

[0059] S102, the image features of the original power transmission inspection image are extracted using a preset image enhancement model, and the extracted image features are fuzzy quantized using a membership function to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0060] Before using an image enhancement model to enhance original power transmission inspection images, the model needs to be trained first. The training objective of the image enhancement model is to minimize the error between the enhanced image and the gold standard image, enabling the model to gradually learn the mapping relationship between the features of the original power transmission inspection image and the enhanced image. During training, each iteration calculates the output result through forward propagation and uses a loss function to evaluate the deviation between the prediction of the image enhancement model and the true value. Subsequently, the gradient information is calculated through backpropagation to guide the update direction of the weights and bias parameters of the image enhancement model. However, in training with complex image data, a fixed learning rate can easily lead to slow convergence or oscillations.

[0061] Therefore, the image enhancement model can be trained using gradient descent and its parameters optimized using an adaptive learning rate. The formula for updating the learning rate can be expressed as:

[0062]

[0063] in, β is the current learning rate; t is the decay factor; and t is the current training round.

[0064] During training, the learning rate is dynamically adjusted based on the number of iterations. A larger learning rate is used in the early stages to quickly reduce the loss value, while the learning rate is gradually decreased in later stages to avoid over-updating and causing oscillations. Through the learning rate update formula, the learning rate gradually decreases with each training epoch, enabling the image enhancement model to have more refined weight adjustment capabilities in the later stages of training, thus achieving smooth convergence. Generally, the initial learning rate is usually set at 10. -3 Up to 10 -4The decay factor β ranges from 0.01 to 0.1. After each training batch, the learning rate is automatically assessed based on the rate of decrease of the loss function to determine whether adjustment is needed. When the rate of decrease of the loss function slows down, the learning rate is automatically reduced to ensure that the model is fine-tuned within the optimal neighborhood.

[0065] The backpropagation algorithm updates the weights and bias parameters, and the updated weights are used to optimize the enhancement effect of the original power transmission inspection image. This backpropagation algorithm adjusts the image based on a loss function between the enhanced image and the gold standard image, which is:

[0066]

[0067] in, As the gold standard for images, Image enhancement results.

[0068] The loss function can be the mean squared error loss function, which quantifies the pixel-level differences between two images. The loss value L tends to decrease as the image enhancement result gets closer to the gold standard. During backpropagation, the computer can calculate the error term of each neuron layer by layer based on the gradient information of the loss function. The gradient descent algorithm is used to adjust the network weights and bias parameters along the negative gradient direction of the loss function, reducing the loss function value in the next iteration. After each iteration, the network updates the weight matrix based on the gradient information, allowing the image enhancement model to continuously approximate the true distribution on different training samples. When the change in the loss function is below a set threshold, the training of the image enhancement model reaches convergence, at which point the generated enhanced image meets the expected standards in terms of sharpness, texture, and structure.

[0069] In this embodiment, after obtaining the original power transmission inspection image, the image can be input into a trained image enhancement model. The image enhancement model can extract low-dimensional basic features and high-dimensional semantic features from the original power transmission inspection image in a hierarchical manner. For example, low-dimensional basic features may include the line outline edge, tower structure texture, gray-level gradient of insulator geometry, and pixel distribution density. High-dimensional semantic features may include the spatial correlation of equipment components, local contrast of defect areas, etc.

[0070] Building upon this, the extracted image features can be mapped into discrete fuzzy membership vectors using membership functions, generating fuzzy feature information that includes feature categories, fuzzy subset affiliations, and corresponding membership degrees. This fuzzy feature information preserves the essential differences in the original power transmission inspection images while reducing the impact of background interference and environmental noise on feature recognition through fuzzy quantization.

[0071] S103, integrate and process the fuzzy feature information corresponding to the original power transmission inspection image to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image; the high-level feature information characterizes the key attributes of the target power transmission line.

[0072] In this embodiment, after obtaining the fuzzy feature information corresponding to the original power transmission inspection image, different types of fuzzy features can be dimensionally aligned through horizontal feature stitching, and weights can be assigned to each type of fuzzy feature information through an adaptive weight allocation algorithm. Based on the weights assigned to each type, a feature extraction network is constructed using vertical multi-level feature aggregation, residual connections, and dilated convolutions, sequentially generating low-level high-level features, mid-level high-level features, and high-level high-level features. Low-level high-level features focus on the local key attributes of the target power transmission line, such as the diameter of the conductors, the number and arrangement of insulators, and the material texture of the towers. Mid-level high-level features emphasize the key attributes related to components, such as the connection and tightness of the conductors and insulators, the verticality of the tower crossarms to the tower body, and the installation accuracy of the hardware. High-level high-level features represent the core key attributes of the overall line operation, such as the icing / corrosion risk level of the line section, whether the safe distance from surrounding obstacles meets the standards, and the overall structural integrity assessment results of the equipment.

[0073] S104. Based on multi-layer high-level feature information, image enhancement processing is performed on the original power transmission inspection image to obtain the image enhancement result of the original power transmission inspection image.

[0074] In this embodiment, after obtaining multi-layer high-level feature information of the original power transmission inspection image, the computer device can perform enhancement processing on the corresponding region of the original power transmission inspection image according to the multi-layer high-level feature information to obtain the image enhancement result of the original power transmission inspection image. Alternatively, the computer device can also generate a feature image according to the multi-layer high-level feature information and fuse the original power transmission inspection image with the feature image to obtain the image enhancement result of the original power transmission inspection image.

[0075] It should also be noted that the image enhancement result of the original power transmission line inspection image is the enhanced power transmission line inspection image. This enhanced image includes multiple resolution levels, meaning it is obtained by scaling the resolution of the original power transmission line inspection image to suit different analytical needs. The resolution scaling process can be represented as:

[0076]

[0077] Among them, R n (x, y) is the image at the nth resolution level; I(x, y) is the original image; n is the resolution level.

[0078] The image enhancement method described above involves: acquiring the original transmission line inspection image of the target transmission line; extracting image features from the original transmission line inspection image using a preset image enhancement model; performing fuzzy quantization on the extracted image features using a membership function to obtain fuzzy feature information corresponding to the original transmission line inspection image; integrating the fuzzy feature information corresponding to the original transmission line inspection image to obtain multi-layer high-level feature information corresponding to the original transmission line inspection image; the high-level feature information characterizes the key attributes of the target transmission line; and performing image enhancement processing on the original transmission line inspection image based on the multi-layer high-level feature information to obtain the image enhancement result of the original transmission line inspection image. This method, through membership functions, can extract fuzzy features from the original transmission line inspection image, and combined with the integration processing method, can extract deeper-level high-level feature information. Through the enhancement process of multi-layer high-level feature information, the clarity and detail fidelity of the original transmission line inspection image can be significantly improved, effectively coping with working conditions such as changes in illumination, noise interference, and complex backgrounds. This makes the enhanced transmission line inspection image closer to the real scene in terms of structure, texture, and layering, thus improving the quality of the enhanced transmission line inspection image.

[0079] The following example details the process of obtaining fuzzy feature information corresponding to the original power transmission inspection image by performing fuzzy quantization on the extracted image features using membership functions. Figure 3 As shown, the details include:

[0080] S201, the membership function is used to perform membership degree mapping on the extracted image features to determine the fuzzy membership degree value of each pixel in the original power transmission inspection image.

[0081] In this embodiment, before performing membership mapping on the image features, the extracted image features are first normalized to map them to the [0,1] interval. Then, the membership value of each pixel's image features in the fuzzy set is calculated using the membership function to obtain the fuzzy membership value of each pixel in the original power transmission inspection image.

[0082] S202, based on the fuzzy membership value of each pixel in the original power transmission inspection image, obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0083] In this embodiment, after obtaining the fuzzy membership value of each pixel in the original power transmission line inspection image, an adaptive threshold segmentation algorithm can be used to divide the original power transmission line inspection image into a target power transmission line region and a background region. For the target power transmission line region, core statistical features such as variance, mean, maximum, and minimum of the pixel fuzzy membership value are calculated. Simultaneously, for the background region, core statistical features such as variance, mean, maximum, and minimum of the pixel fuzzy membership value are calculated. Finally, dimensionality reduction processing is performed on the statistical features of each region to obtain the fuzzy feature information corresponding to the original power transmission line inspection image.

[0084] In the image enhancement method described above, membership functions are used to map the extracted image features to determine the fuzzy membership value of each pixel in the original power transmission inspection image. Based on the fuzzy membership value of each pixel in the original power transmission inspection image, the corresponding fuzzy feature information of the original power transmission inspection image is obtained. This method accurately determines the fuzzy membership value of each pixel in the original power transmission inspection image by mapping the extracted image features to pixel-level membership functions, thus accurately obtaining the fuzziness differences between different pixels in the original power transmission inspection image.

[0085] Assuming that image features include at least color features, texture features, brightness features, and edge features, then in one embodiment, such as Figure 4 As shown, the detailed content of determining the fuzzy membership value of each pixel in the original power transmission inspection image by using membership functions to map the extracted image features includes:

[0086] S301, the RGB values ​​of each pixel in the original power transmission inspection image are blurred using a membership function to determine the color blur membership value of each pixel in the original power transmission inspection image; different texture regions in the original power transmission inspection image are blurred using a membership function to determine the texture blur membership value of each pixel in the original power transmission inspection image; the brightness value of each pixel in the original power transmission inspection image is blurred using a membership function to determine the brightness blur membership value of each pixel in the original power transmission inspection image; and the edge values ​​of edge pixels in the original power transmission inspection image are blurred using a membership function to determine the edge blur membership value of edge pixels in the original power transmission inspection image.

[0087] In this embodiment, the image features include at least color features, texture features, brightness features, and edge features. For color features, the RGB value of each pixel can be blurred to obtain its membership degree, representing the degree of ambiguity in which the color belongs to a certain color set. For texture features, the texture of each pixel can be blurred to obtain its membership degree, representing the membership degree of the image in different texture regions. It should be noted that texture features can be extracted using a gray-level co-occurrence matrix. Brightness features and edge features can also be blurred through local gray-level changes.

[0088] S302, take the fuzzy membership values ​​of each color, each texture, each brightness, and each edge as the fuzzy membership value of each pixel in the original power transmission inspection image.

[0089] In this embodiment of the application, after obtaining the fuzzy membership values ​​of different features, the fuzzy membership values ​​of different features can be fused, and the fusion result can be used as the fuzzy membership value of each pixel in the original power transmission inspection image.

[0090] In the aforementioned image enhancement method, the RGB values ​​of each pixel in the original power transmission inspection image are blurred using membership functions to determine the color blur membership value of each pixel in the original power transmission inspection image; different texture regions in the original power transmission inspection image are blurred using membership functions to determine the texture blur membership value of each pixel in the original power transmission inspection image; the brightness values ​​of each pixel in the original power transmission inspection image are blurred using membership functions to determine the brightness blur membership value of each pixel in the original power transmission inspection image; and the edge values ​​of edge pixels in the original power transmission inspection image are blurred using membership functions to determine the edge blur membership value of edge pixels in the original power transmission inspection image; and each color blur membership value, each texture blur membership value, each brightness blur membership value, and each edge blur membership value are used as the blur membership value of each pixel in the original power transmission inspection image. This method performs multi-dimensional fuzzification processing on the RGB value, texture features, brightness value, and edge value of each pixel in the original power transmission inspection image using targeted membership functions. It accurately obtains the corresponding fuzzy membership values ​​for each dimension and integrates them into a pixel-level comprehensive fuzzy membership value, which significantly improves the comprehensiveness and accuracy of the feature representation of the power transmission inspection image.

[0091] In one embodiment, the specific content of obtaining the fuzzy feature information corresponding to the original power transmission inspection image based on the fuzzy membership value of each pixel in the original power transmission inspection image includes:

[0092] Based on the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image are weighted to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0093] In this embodiment, for any feature type, the computer device can calculate the weight of the feature type and multiply it by the fuzzy membership value of the corresponding feature type to obtain the membership feature information of the feature type. Then, the membership feature information of all feature types is added together to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0094] In the aforementioned image enhancement method, based on the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image are weighted to obtain the fuzzy feature information corresponding to the original power transmission inspection image. This method generates fuzzy feature information by assigning differentiated weights to each feature type and performing weighted processing based on the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image. This ensures that the obtained fuzzy feature information retains the differences between each feature type, making the fuzzy feature information more comprehensive and accurate.

[0095] In one embodiment, such as Figure 5 As shown, the methods for obtaining the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image include:

[0096] S401, calculate the activation degree of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image; the activation degree is used to describe the joint strength between the pixel and other feature types.

[0097] In this embodiment, for any feature type among the fuzzy membership values ​​of each pixel, the computer device can calculate the activation degree of that feature type according to the activation degree calculation formula. This activation degree can also describe the intersection of two features at the same point in time.

[0098] S402, when the activation degree of each feature type meets the preset conditions, assign the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image based on the minimum activation degree among each activation degree.

[0099] In this embodiment, when the activation degree of each feature type meets the preset condition, the result is activation. At this time, the minimum activation degree among all activation degrees can be taken as the activation degree of that pixel. Based on the mapping relationship between activation degree and weight, the activation degree is mapped to obtain the weight of each feature type in the fuzzy membership value of each pixel.

[0100] In the image enhancement method described above, the activation degree of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image is calculated. The activation degree describes the joint strength between each feature type and other feature types. When the activation degree of each feature type satisfies a preset condition, the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image is assigned based on the minimum activation degree among all activation degrees. This method accurately quantifies the joint strength between feature types by calculating the activation degree of the fuzzy membership value of each pixel in the original power transmission inspection image. Furthermore, by assigning weights to corresponding feature types based on the minimum activation degree among all activation degrees using a preset activation degree threshold condition, the synergy between features can be strengthened, enabling fuzzy feature information to more accurately focus on the core features of the power transmission equipment.

[0101] The above embodiments are all introductions to the fuzzy quantization process. The following embodiment will explain in detail the process of integrating and processing the fuzzy feature information corresponding to the original power transmission inspection image to obtain the multi-layer high-level feature information corresponding to the original power transmission inspection image. Figure 6 As shown, the details include:

[0102] S501, redundant fuzzy feature information is filtered out from the fuzzy feature information corresponding to the original power transmission inspection image to obtain target fuzzy feature information related to the target power transmission line.

[0103] In this embodiment, after obtaining the fuzzy feature information corresponding to the original power transmission inspection image, the computer device can learn the fuzzy feature information through convolutional layers. Specifically, the convolutional kernel slides across the fuzzy feature information to scan the fuzzy feature information corresponding to different image regions, extracting local feature information related to the target power transmission line. For example, local feature information can include edges, corners, texture directions, and structural morphology. The weights of the convolutional layers can be automatically adjusted during training, enabling the image enhancement model to autonomously learn the most critical feature patterns for generating enhanced images.

[0104] S502, integrate the target fuzzy feature information related to the target transmission line to obtain multi-layer high-level feature information corresponding to the original transmission line inspection image.

[0105] Among them, the multi-level high-level features contain a complete hierarchy from low-level edge features to high-level semantic information.

[0106] In this embodiment, the target fuzzy feature information related to the target transmission line extracted by the convolutional layer is input into the pooling layer for dimensionality reduction. Max pooling or average pooling compresses the feature size while retaining the main feature information. This also reduces computational complexity and the number of parameters. In other words, pooling not only improves the computational efficiency of the image enhancement model but also enhances its robustness to image rotation, translation, and lighting changes.

[0107] For example, in max pooling, the maximum activation value within each sub-region is taken as the output to ensure that key features are not lost during feature propagation. The high-dimensional features compressed by the pooling layer are input to the fully connected layer for feature fusion and decision mapping. The fully connected layer models the relationships between different features using a weight matrix, achieving a comprehensive representation of global features—that is, multi-layered high-level feature information. In other words, the fully connected layer of the image enhancement model outputs not only a single classification result but also a high-dimensional feature vector, providing accurate deep feature representation for subsequent image reconstruction and enhancement.

[0108] In the aforementioned image enhancement method, redundant fuzzy features are filtered out from the fuzzy feature information corresponding to the original power transmission line inspection image to obtain target fuzzy feature information related to the target power transmission line. This target fuzzy feature information related to the target power transmission line is then integrated to obtain multi-layer high-level feature information corresponding to the original power transmission line inspection image. This method, by selectively filtering out redundant fuzzy features unrelated to the target power transmission line from the fuzzy feature information of the original power transmission line inspection image and then integrating features, can enhance the semantic relevance and recognizability of the features, resulting in higher accuracy of the multi-layer high-level feature information.

[0109] In one embodiment, such as Figure 7 As shown, the detailed content of the image enhancement result of the original power transmission inspection image obtained by performing image enhancement processing on the original power transmission inspection image based on multi-layer high-level feature information includes:

[0110] S601 performs image enhancement processing on the original power transmission inspection image based on multi-layer high-level feature information to generate an enhanced original power transmission inspection image.

[0111] In this embodiment, after obtaining multi-layer high-level feature information, the computer device can enhance the corresponding region of the original power transmission inspection image according to the multi-layer high-level feature information to obtain an enhanced original power transmission inspection image. Alternatively, the computer device can also generate a feature image according to the multi-layer high-level feature information and fuse the original power transmission inspection image with the feature image to obtain an enhanced original power transmission inspection image.

[0112] S602, perform detail enhancement and / or brightness adjustment on the enhanced original power transmission inspection image to obtain the image enhancement result of the original power transmission inspection image.

[0113] After image enhancement using multiple layers of advanced feature information, the resulting enhanced image already possesses high clarity and structural fidelity. However, in actual power transmission inspection scenarios, due to factors such as shooting equipment, lighting conditions, and weather conditions, the brightness and contrast of the image may deviate. Therefore, in order to improve the visualization effect and analysis accuracy of the image under different monitoring environments, detail enhancement and brightness adjustment are still required.

[0114] In this embodiment, for detail enhancement, the computer device can further process the enhanced original power transmission inspection image using a sharpening algorithm to improve the image clarity. The sharpening algorithm aims to enhance the detail areas of the image, making the contour edges clearer and the structural contrast more obvious. Its core idea is to detect the high-frequency components of the power transmission inspection image and linearly superimpose them with the original power transmission inspection image to enhance visual details. The convolution kernel k(i,j) generally uses a 3×3 or 5×5 template, with negative weights at the center and positive weights at the periphery. During the operation, the convolution kernel extracts the high-frequency information of the power transmission inspection image by performing a convolution operation with the input image. This high-frequency information is then weighted and superimposed back onto the original power transmission inspection image to generate a sharpened image with higher clarity. In different types of power transmission inspection images (such as high-voltage towers, insulators, and hardware), the image texture and lighting vary significantly. Therefore, the sharpening coefficient α can be dynamically adjusted according to the local contrast and noise level of the image, typically ranging from 0.5 to 2.0. When image noise is high, the sharpening intensity is automatically reduced to avoid artifact amplification. Furthermore, to prevent over-enhancement leading to edge breakage, a threshold suppression mechanism is introduced; sharpening is not performed when the local gradient change falls below a set threshold, thus ensuring a natural image transition. After sharpening, the output image shows improvements in detail, edge sharpness, and structural visibility. This is particularly evident in high-altitude power transmission equipment inspection scenarios, where features such as the edges, cracks, and stains on metal components are more clearly visible, aiding subsequent intelligent recognition algorithms in quickly locating fault areas.

[0115] The detail enhancement process of the sharpening algorithm can be represented as:

[0116]

[0117] Where G(x,y) is the image after detail enhancement; I(x,y) is the image before detail enhancement; α is the sharpening coefficient; and k(i,j) is the convolution kernel.

[0118] For brightness adjustment, computer equipment can use a normalized linear contrast adjustment algorithm to adjust the enhanced original power transmission inspection image or the original power transmission inspection image with enhanced details. By adjusting the dynamic range of pixel values ​​in the power transmission inspection image, dark areas become clearer and bright areas become more transparent. Specifically, through linear stretching, the pixel value distribution is expanded to the [0, 1] or [0, 255] range, thereby achieving global contrast optimization. Brightness adjustment can preserve image details while enhancing the tonal differences between bright and dark areas, making the overall visual effect of the image more vivid.

[0119] The brightness adjustment process of the contrast algorithm can be represented as:

[0120]

[0121] Where C(x) is the adjusted contrast value, I(x) is the pixel value of the input image, and min(I) and max(I) are the minimum and maximum values ​​of the total image, respectively.

[0122] In the aforementioned image enhancement method, image enhancement processing is performed on the original power transmission inspection image based on multi-layer high-level feature information to generate an enhanced original power transmission inspection image. The enhanced original power transmission inspection image is then subjected to detail enhancement and / or brightness adjustment to obtain the image enhancement result. This method enhances the original power transmission inspection image from a global perspective through multi-layer high-level feature information, and further enhances the image quality from a local perspective through detail enhancement and / or brightness adjustment, resulting in a better image enhancement result for the original power transmission inspection image.

[0123] Based on the above embodiments, after obtaining the image enhancement result of the original power transmission inspection image, i.e., the enhanced power transmission inspection image, the computer device can send it to the intelligent analysis system through a secure transmission channel (such as a local area network, a fifth-generation (5G) edge node, or an encrypted application programming interface (API)). During transmission, encryption protocols (such as Transport Layer Security (TLS) / Secure Sockets Layer (SSL)) and hash verification mechanisms are used to ensure the integrity and security of the image data. Before entering the intelligent analysis model, the system performs size normalization, noise filtering, and grayscale standardization on the input image to ensure the consistency and accuracy of subsequent model inputs. The enhanced power transmission inspection image is input into a preset fault diagnosis model, which performs classification and regression analysis on the enhanced power transmission inspection image to generate fault diagnosis results for the target power transmission line. These fault diagnosis results include the fault type, fault location, and fault severity. The fault diagnosis results can be presented in the form of a report.

[0124] It should be noted that the fault diagnosis model uses deep learning algorithms to perform multi-dimensional analysis of the enhanced images. Taking the classification branch as an example, the model identifies labels for abnormal regions in the image and outputs the corresponding fault types, such as "insulator crack," "fitting corrosion," and "foreign object attachment." Taking the regression branch as an example, the model calculates the spatial distribution parameters and feature weights of the abnormal regions, outputting the fault's location coordinates (x, y) and severity score S, where a higher score indicates a greater risk. Through joint optimization of the dual-branch structure, the model achieves both high accuracy and strong interpretability in fault detection.

[0125] In a detailed embodiment, such as Figure 8 As shown, the above image enhancement method includes:

[0126] S701, acquire the original transmission line inspection image of the target transmission line;

[0127] S702, the image features of the original power transmission inspection image are extracted using a preset image enhancement model; the RGB values ​​of each pixel in the original power transmission inspection image are blurred using a membership function to determine the color blur membership value of each pixel in the original power transmission inspection image; different texture regions in the original power transmission inspection image are blurred using a membership function to determine the texture blur membership value of each pixel in the original power transmission inspection image; the brightness value of each pixel in the original power transmission inspection image is blurred using a membership function to determine the brightness blur membership value of each pixel in the original power transmission inspection image; and the edge values ​​of edge pixels in the original power transmission inspection image are blurred using a membership function to determine the edge blur membership value of edge pixels in the original power transmission inspection image.

[0128] S703 uses the fuzzy membership values ​​of each color, each texture, each brightness, and each edge as the fuzzy membership value of each pixel in the original power transmission inspection image.

[0129] S704, based on the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image are weighted to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0130] S705, redundant fuzzy feature information is filtered out from the fuzzy feature information corresponding to the original power transmission inspection image to obtain the target fuzzy feature information related to the target power transmission line;

[0131] S706, integrate the target fuzzy feature information related to the target transmission line to obtain multi-layer high-level feature information corresponding to the original transmission line inspection image;

[0132] S707 performs image enhancement processing on the original power transmission inspection image based on multi-layer high-level feature information to generate an enhanced original power transmission inspection image.

[0133] S708 performs detail enhancement and / or brightness adjustment on the enhanced original power transmission inspection image to obtain the image enhancement result of the original power transmission inspection image.

[0134] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed 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 performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0135] Based on the same inventive concept, this application also provides an image enhancement apparatus for implementing the image enhancement method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more image enhancement apparatus embodiments provided below can be found in the limitations of the image enhancement method described above, and will not be repeated here.

[0136] In one exemplary embodiment, such as Figure 9 As shown, an image enhancement device is provided, including: an acquisition module 11, a quantization module 12, a processing module 13, and an enhancement module 14, wherein:

[0137] The acquisition module 11 is used to acquire the original transmission line inspection image of the target transmission line;

[0138] The quantization module 12 is used to extract image features of the original power transmission inspection image using a preset image enhancement model, and to perform fuzzy quantization on the extracted image features using a membership function to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0139] The sorting module 13 is used to integrate and process the fuzzy feature information corresponding to the original power transmission inspection image to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image; the high-level feature information characterizes the key attributes of the target power transmission line.

[0140] Enhancement module 14 is used to perform image enhancement processing on the original power transmission inspection image based on multi-layer high-level feature information to obtain the image enhancement result of the original power transmission inspection image.

[0141] In an exemplary embodiment, the quantization module described above includes a mapping unit and a determination unit, wherein:

[0142] The mapping unit is used to perform membership degree mapping on the extracted image features using membership functions to determine the fuzzy membership degree value of each pixel in the original power transmission inspection image;

[0143] The determining unit is used to obtain the fuzzy feature information corresponding to the original power transmission inspection image based on the fuzzy membership value of each pixel in the original power transmission inspection image.

[0144] In an exemplary embodiment, the mapping unit is further configured to: blur the RGB values ​​of each pixel in the original power transmission inspection image using membership functions to determine the color blur membership value of each pixel in the original power transmission inspection image; blur different texture regions in the original power transmission inspection image using membership functions to determine the texture blur membership value of each pixel in the original power transmission inspection image; blur the brightness values ​​of each pixel in the original power transmission inspection image using membership functions to determine the brightness blur membership value of each pixel in the original power transmission inspection image; blur the edge values ​​of edge pixels in the original power transmission inspection image using membership functions to determine the edge blur membership value of edge pixels in the original power transmission inspection image; and use each color blur membership value, each texture blur membership value, each brightness blur membership value, and each edge blur membership value as the blur membership value of each pixel in the original power transmission inspection image.

[0145] In an exemplary embodiment, the determining unit is further configured to perform weighted processing on the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image based on the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image, so as to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

[0146] In an exemplary embodiment, the determining unit is further configured to calculate the activation degree of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image; the activation degree is used to describe the joint strength between the feature type and other feature types; and when the activation degree of each feature type satisfies the preset conditions, the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image is assigned based on the minimum activation degree among the activation degrees.

[0147] In an exemplary embodiment, the above-described integration module includes a screening unit and an integration unit, wherein:

[0148] The filtering unit is used to filter out redundant fuzzy feature information from the fuzzy feature information corresponding to the original power transmission inspection image, and obtain the target fuzzy feature information related to the target power transmission line.

[0149] The integration unit is used to integrate the target fuzzy feature information related to the target transmission line to obtain multi-layer high-level feature information corresponding to the original transmission line inspection image.

[0150] In an exemplary embodiment, the enhancement module includes an enhancement unit and an adjustment unit, wherein:

[0151] The enhancement unit is used to perform image enhancement processing on the original power transmission inspection image based on multi-layer high-level feature information to generate an enhanced original power transmission inspection image.

[0152] The adjustment unit is used to perform detail enhancement and / or brightness adjustment on the enhanced original power transmission inspection image to obtain the image enhancement result of the original power transmission inspection image.

[0153] Each module in the aforementioned image enhancement device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0154] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the content of any of the embodiments of the image enhancement methods described above.

[0155] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the content of any of the embodiments of the above-described image enhancement methods.

[0156] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the content of any one of the embodiments of the image enhancement methods described above.

[0157] 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 used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0158] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, database, or other media 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), magnetic 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 take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0159] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0160] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An image enhancement method, characterized in that, The method includes: Acquire raw transmission line inspection images of the target transmission line; The image features of the original power transmission inspection image are extracted using a preset image enhancement model, and the extracted image features are fuzzy quantized using a membership function to obtain the fuzzy feature information corresponding to the original power transmission inspection image. The fuzzy feature information corresponding to the original power transmission inspection image is integrated and processed to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image; the high-level feature information characterizes the key attributes of the target power transmission line. Based on the multi-layer high-level feature information, the original power transmission inspection image is subjected to image enhancement processing to obtain the image enhancement result of the original power transmission inspection image.

2. The method according to claim 1, characterized in that, The step of using membership functions to perform fuzzy quantization on the extracted image features to obtain fuzzy feature information corresponding to the original power transmission inspection image includes: The membership function is used to perform membership degree mapping on the extracted image features to determine the fuzzy membership degree value of each pixel in the original power transmission inspection image; Based on the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy feature information corresponding to the original power transmission inspection image is obtained.

3. The method according to claim 2, characterized in that, The image features include at least color features, texture features, brightness features, and edge features; The step of using the membership function to perform membership degree mapping on the extracted image features to determine the fuzzy membership degree value of each pixel in the original power transmission inspection image includes: The membership function is used to blur the RGB values ​​of each pixel in the original power transmission inspection image to determine the color blur membership value of each pixel in the original power transmission inspection image; the membership function is used to blur different texture regions in the original power transmission inspection image to determine the texture blur membership value of each pixel in the original power transmission inspection image; the membership function is used to blur the brightness values ​​of each pixel in the original power transmission inspection image to determine the brightness blur membership value of each pixel in the original power transmission inspection image; and the membership function is used to blur the edge values ​​of edge pixels in the original power transmission inspection image to determine the edge blur membership value of edge pixels in the original power transmission inspection image. The color fuzz membership value, the texture fuzz membership value, the brightness fuzz membership value, and the edge fuzz membership value are used as the fuzz membership value of each pixel in the original power transmission inspection image.

4. The method according to claim 3, characterized in that, The process of obtaining fuzzy feature information corresponding to the original power transmission inspection image based on the fuzzy membership value of each pixel in the original power transmission inspection image includes: Based on the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image, the fuzzy membership values ​​of multiple feature types of each pixel in the original power transmission inspection image are weighted to obtain the fuzzy feature information corresponding to the original power transmission inspection image.

5. The method according to claim 4, characterized in that, The method for obtaining the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image includes: Calculate the activation degree of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image; the activation degree is used to describe the joint strength between the pixel and other feature types. When the activation degree of each feature type meets the preset conditions, the weight of each feature type in the fuzzy membership value of each pixel in the original power transmission inspection image is assigned based on the minimum activation degree among the activation degrees.

6. The method according to any one of claims 1-5, characterized in that, The process of integrating the fuzzy feature information corresponding to the original power transmission inspection image to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image includes: Redundant fuzzy feature information is filtered out from the fuzzy feature information corresponding to the original power transmission inspection image to obtain target fuzzy feature information related to the target power transmission line; The target fuzzy feature information related to the target transmission line is integrated to obtain multi-layer high-level feature information corresponding to the original transmission line inspection image.

7. The method according to any one of claims 1-5, characterized in that, The image enhancement processing of the original power transmission inspection image based on the multi-layer high-level feature information, to obtain the image enhancement result of the original power transmission inspection image, includes: Based on the multi-layer high-level feature information, the original power transmission inspection image is enhanced to generate the enhanced original power transmission inspection image. The enhanced original power transmission inspection image is subjected to detail enhancement and / or brightness adjustment to obtain the image enhancement result of the original power transmission inspection image.

8. An image enhancement device, characterized in that, The device includes: The acquisition module is used to acquire the original transmission line inspection images of the target transmission line; The quantization module is used to extract image features from the original power transmission inspection image using a preset image enhancement model, and to perform fuzzy quantization on the extracted image features using a membership function to obtain the fuzzy feature information corresponding to the original power transmission inspection image. The processing module is used to integrate and process the fuzzy feature information corresponding to the original power transmission inspection image to obtain multi-layer high-level feature information corresponding to the original power transmission inspection image; the high-level feature information characterizes the key attributes of the target power transmission line. An enhancement module is used to perform image enhancement processing on the original power transmission inspection image based on the multi-layer high-level feature information to obtain the image enhancement result of the original power transmission inspection image.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images