Power transmission line monitoring method and device based on machine vision

Abstract

The invention discloses a power transmission line monitoring method and device based on machine vision, and relates to the technical field of machine vision. Acquiring a line image including a target power transmission line, and preprocessing the line image to obtain an initial image; performing edge enhancement on the initial image to obtain a first enhanced image; determining a target foreground image only containing the target power transmission line according to the line image and the first enhanced image; performing boundary enhancement on the target foreground image through a preset algorithm to obtain a second enhanced image; and substituting the second enhanced image into a preset model to obtain a target feature, and searching a preset database according to the target feature to determine the line state of the target power transmission line. By performing multiple enhancement on the image, key features of the target power transmission line can be highlighted, noise and interference can be reduced, so that the accuracy of target identification is improved, detailed identification can be performed on the power transmission line through image processing and the preset model, external interference caused by snow accumulation and icing is reduced, and the line monitoring efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine vision, and in particular relates to a method and device for monitoring a power transmission line based on machine vision. Background Art

[0002] As the scale of power grid continues to expand, the safe operation of transmission lines is of vital importance. Snowfall in winter poses a serious threat to the normal operation of transmission lines. Snow accumulation and ice cover may cause line breakage, insulator damage, wire overload and other problems. At present, the inspection of transmission lines mainly relies on manual inspection, but manual inspection has problems such as heavy workload, low efficiency and susceptibility to environmental factors under severe weather conditions, making it difficult to achieve efficient and intelligent monitoring.

[0003] Patent CN111402248B discloses a method for detecting defects in power transmission line conductors based on machine vision, including: obtaining on-site image data and making a training data set; constructing and training an instance segmentation network to obtain a prediction model; model-deriving input images to obtain rectangular area images and binary mask images of the conductors; using a skeleton algorithm to extract the conductor skeleton, and calculating the average width of the conductor to reconstruct the binary mask image; using a homomorphic filtering algorithm to eliminate the influence of uneven illumination of the rectangular area image, and extracting the segmented conductor area image in combination with the reconstructed binary mask image; generating a large number of rectangular frames on the conductor area for screening; making a classification training data set, constructing and training a shallow classification network, and obtaining a classification prediction model; inputting the conductor segment area image into the classification prediction model, and counting the defect type and defect ratio of the conductor segment state. The present invention can accurately segment the conductor and detect the state of the conductor in segments, and judge the type and degree of conductor defects. However, the scheme is not accurate enough when facing lines with ice or snow, which makes the line monitoring efficiency low. Summary of the invention

[0004] The purpose of the present invention is to solve the problem and to propose a transmission line monitoring method and device based on machine vision.

[0005] In a first aspect of the present invention, a method for monitoring a power transmission line based on machine vision is first proposed, the method comprising: Acquire a line image including a target power transmission line, and preprocess the line image to obtain an initial image; Performing edge enhancement on the initial image to obtain a first enhanced image; Determining a target foreground image that only includes the target power transmission line according to the line image and the first enhanced image; Performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image; Substituting the second enhanced image into a preset model to obtain target features, searching a preset database according to the target features to determine the line status of the target power transmission line.

[0006] Optionally, preprocessing the line image to obtain an initial image includes: Performing non-local mean denoising on the line image and then reconstructing the resolution to obtain a first preset image; Performing adaptive local contrast enhancement on the first preset image to obtain a second preset image; The second preset image is subjected to grayscale conversion and then multi-scale bilateral filtering to obtain an initial image.

[0007] Optionally, determining a target foreground image that only includes the target power transmission line according to the line image and the first enhanced image includes: Performing multi-Otsu threshold and connectivity map analysis on the first enhanced image to obtain a target recognition area; The line image is cut according to the target recognition area to obtain a target foreground image.

[0008] Optionally, performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image includes: Performing grayscale conversion on the target foreground image to obtain a target grayscale image, and performing gradient morphological operation on the target grayscale image to obtain a target edge image; Performing non-maximum suppression on the target edge image to obtain a target fused image; The target foreground image and the target fused image are weightedly fused to obtain a second enhanced image.

[0009] Optionally, substituting the second enhanced image into a preset model to obtain a target feature includes: The second enhanced image is passed through a first convolution module and a second convolution module to obtain a first convolution feature, and the first convolution feature is passed through a first C2f module to obtain a first extraction feature; The first extracted feature image is passed through a third convolution module to obtain a second convolution feature, the first convolution feature and the second convolution feature are weightedly fused to obtain a first fused feature, and the first fused feature is passed through a second C2f module to obtain a second extracted feature; The second extracted feature is passed through a first attention module to obtain a first attention feature; The second extracted feature is passed through a fourth convolution module to obtain a third convolution feature, the second convolution feature and the third convolution feature are weightedly fused to obtain a second fused feature, and the second fused feature is passed through a third C2f module to obtain a third extracted feature; The third extracted feature is passed through a second attention module to obtain a second attention feature; The third extracted feature is subjected to a feature enhancement module to obtain a first enhanced feature; The first attention feature, the second attention feature and the first enhanced feature are weightedly fused to obtain a target feature.

[0010] In a second aspect of the present invention, a machine vision-based power transmission line monitoring device is provided, comprising: A preprocessing module, used to obtain a line image including a target power transmission line, and preprocess the line image to obtain an initial image; An edge enhancement module, used for performing edge enhancement on the initial image to obtain a first enhanced image; a target foreground image determination module, configured to determine a target foreground image containing only the target power transmission line according to the line image and the first enhanced image; A boundary enhancement module, used for performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image; The target feature determination module is used to substitute the second enhanced image into a preset model to obtain the target feature, and to search a preset database according to the target feature to determine the line state of the target transmission line.

[0011] Optionally, the preprocessing module includes: A resolution reconstruction module, used for performing non-local mean denoising on the line image and then reconstructing the resolution to obtain a first preset image; A contrast enhancement module, configured to perform adaptive local contrast enhancement on the first preset image to obtain a second preset image; A grayscale conversion module is used to perform grayscale conversion on the second preset image and then perform multi-scale bilateral filtering to obtain an initial image.

[0012] Optionally, the target foreground image determination module includes: A target recognition region determination module, used for performing multi-Otsu threshold and connectivity graph analysis on the first enhanced image to obtain a target recognition region; A region cutting module is used to cut the line image according to the target recognition region to obtain a target foreground image.

[0013] Optionally, the boundary enhancement module includes: A target edge image determination module is used to perform grayscale conversion on the target foreground image to obtain a target grayscale image, and perform gradient morphological operation on the target grayscale image to obtain a target edge image; A target fused image determination module, used for performing non-maximum suppression on the target edge image to obtain a target fused image; The first weighted fusion module is used to perform weighted fusion on the target foreground image and the target fused image to obtain a second enhanced image.

[0014] Optionally, the target feature determination module includes: A first extraction feature determination module, used for obtaining a first convolution feature from the second enhanced image through a first convolution module and a second convolution module, and obtaining a first extraction feature through a first C2f module from the first convolution feature; A second extraction feature determination module, used for obtaining a second convolution feature by passing the first extraction feature image through a third convolution module, obtaining a first fusion feature by weighted fusion of the first convolution feature and the second convolution feature, and obtaining a second extraction feature by passing the first fusion feature through a second C2f module; A first attention feature determination module, used for obtaining a first attention feature through the first attention module from the second extracted feature; A third extraction feature determination module, used for obtaining a third convolution feature by passing the second extraction feature through a fourth convolution module, obtaining a second fusion feature by weighted fusion of the second convolution feature and the third convolution feature, and obtaining a third extraction feature by passing the second fusion feature through a third C2f module; A second attention feature determination module, used for obtaining a second attention feature through the second attention module from the third extracted feature; A first enhanced feature determination module, used for obtaining a first enhanced feature through the feature enhancement module of the third extracted feature; The second weighted fusion module is used to perform weighted fusion on the first attention feature, the second attention feature and the first enhanced feature to obtain a target feature.

[0015] Beneficial effects of the present invention: The present invention proposes a transmission line monitoring method based on machine vision, which obtains a line image containing a target transmission line, pre-processes the line image to obtain an initial image; performs edge enhancement on the initial image to obtain a first enhanced image; determines a target foreground image containing only the target transmission line according to the line image and the first enhanced image; performs boundary enhancement on the target foreground image through a preset algorithm to obtain a second enhanced image; substitutes the second enhanced image into a preset model to obtain target features, and searches a preset database according to the target features to determine the line status of the target transmission line. By performing multiple enhancements on the image, the key features of the target transmission line can be highlighted, noise and interference can be reduced, thereby improving the accuracy of target recognition. Through image processing and preset models, the transmission line can be identified in detail, reducing external interference caused by snow and ice, and improving the efficiency of line monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 A flowchart of a method for monitoring a power transmission line based on machine vision provided by an embodiment of the present invention; Figure 2 A schematic structural diagram of a power transmission line monitoring device based on machine vision provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the present invention, the description of "first", "second", etc. is only used for descriptive purposes, and cannot be understood as indicating or implying its relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0019] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0020] The embodiment of the present invention provides a method for monitoring a power transmission line based on machine vision. Figure 1 , Figure 1 A flow chart of a method for monitoring a power transmission line based on machine vision provided by an embodiment of the present invention. The method comprises the following steps: S101, acquiring a line image including a target transmission line, and preprocessing the line image to obtain an initial image; S102, performing edge enhancement on the initial image to obtain a first enhanced image; S103, determining a target foreground image that only includes the target power transmission line according to the line image and the first enhanced image; S104, performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image; S105, substituting the second enhanced image into a preset model to obtain target features, and searching a preset database according to the target features to determine the line status of the target transmission line.

[0021] A machine vision-based transmission line monitoring method provided in an embodiment of the present invention can highlight the key features of the target transmission line and reduce noise and interference by performing multiple enhancements on the image, thereby improving the accuracy of target recognition. Through image processing and preset models, the transmission line can be identified in detail, reducing external interference caused by snow and ice accumulation, thereby improving the line monitoring efficiency.

[0022] In one implementation, preprocessing can remove noise and interference in the image and improve the quality of the image. In complex weather conditions such as snow, fog, low light, etc., preprocessing can enhance the usability of the image and ensure that the lines and key features in the image are still clearly presented even in unideal environments.

[0023] In one implementation, the line image containing the target transmission line is an RGB image taken by a drone. The preset database is used to save different line states and corresponding features. If the similarity between two features is greater than a similarity threshold, the two features are considered to be the same. The similarity threshold is determined by a technician.

[0024] In one implementation, edge enhancement can improve the recognizability of details in an image, especially under visually complex conditions such as snow cover or ice, helping the system to accurately identify line anomalies; improving the clarity of important boundaries and contours in the image helps reduce the impact of background noise and irrelevant information on the results, thereby reducing the probability of misidentification.

[0025] In one implementation, by combining the initial image with the enhanced image, the target transmission line can be accurately extracted, eliminating background interference or non-target areas. After the target area is extracted, subsequent image processing can focus more on key content, reducing computational complexity and improving processing efficiency and accuracy.

[0026] In one implementation, the boundary information in the image is further strengthened by enhancing the boundaries of the target foreground image, making the precise shape of the line and structural anomalies such as fractures or damages clearer, and enabling accurate identification of line details. In complex environments such as snow and ice, the enhanced boundaries can better help the system identify parts of the line that may be covered or damaged, thereby improving its ability to adapt to severe weather.

[0027] In one implementation, by comparing historical data or standardized models in the database, the health of the transmission line can be accurately determined and potential problems can be quickly identified.

[0028] In one embodiment, preprocessing the line image to obtain the initial image includes: Performing non-local mean denoising on the line image and then reconstructing the resolution to obtain a first preset image; Performing adaptive local contrast enhancement on the first preset image to obtain a second preset image; The second preset image is gray-scale converted and then subjected to multi-scale bilateral filtering to obtain an initial image.

[0029] In one implementation, non-local mean denoising is an information denoising method based on the overall similarity of an image, which can effectively reduce the noise in an image while retaining the details of the image. By processing the entire image, the influence of noise in a single local area is avoided.

[0030] In one implementation, the clarity of the image can be improved through resolution reconstruction, especially for small line structures and potential problems such as fractures and snow accumulation, which helps to observe details more accurately and improve the effect of subsequent processing.

[0031] In one implementation, non-local mean denoising is performed on the line image as follows: first, a search window (an area of ​​size S×S) is defined, and all pixel points similar to pixel i are found in this window. A small area Patch of size f×f (the size of the small area is smaller than the search window) is taken with pixel i as the center. For each candidate pixel i in the search window, a patch of the same size is also taken for comparison, and the Euclidean distance between the two patches is calculated. The Euclidean distance is converted into a weight using a Gaussian kernel function. Finally, pixel i is denoised by weighted averaging.

[0032] In one implementation, the specific method of adaptive local contrast enhancement is to convert the first preset image into the LAB color space, perform L channel enhancement, divide the image into multiple small sub-blocks (such as 8×8), perform histogram equalization on each area separately, count the pixel intensity distribution in each small area, calculate its cumulative distribution function, and then smooth the transition between different areas through bilinear interpolation, and then convert the processed image into an RGB image to obtain the second preset image; by enhancing the L channel and then performing contrast enhancement on different areas, local details will not be lost like global histogram equalization, and bilinear interpolation smoothing can be used for the transition between different areas to make the image more natural.

[0033] In one implementation, the resolution reconstruction is performed by super-resolution reconstruction (such as bicubic interpolation, deep learning SRGAN, ESRGAN, etc.).

[0034] In one implementation, local contrast enhancement can improve detail contrast in local areas, especially for low-contrast areas such as snow-covered wires, which helps to highlight the features of these areas and improve the visibility of details; the adaptive method can dynamically adjust according to the contrast requirements of different areas in the image to avoid over-enhancement or under-enhancement and ensure a balanced improvement in image quality; after contrast enhancement, the key elements of the image, wires and ice and snow accumulation, become more prominent, facilitating subsequent analysis and problem detection.

[0035] In one implementation, grayscale helps to remove color information from an image and simplify the processing process. In particular, when color information is not a key factor, grayscale images can highlight brightness and texture details, making lines and structures clearer. Grayscale images can enhance line structures through brightness contrast, avoid color interference, and help focus on the shape and structural features of the image during subsequent processing.

[0036] In one embodiment, determining the target foreground image containing only the target power transmission line according to the line image and the first enhanced image comprises: Performing multi-Otsu threshold and connectivity map analysis on the first enhanced image to obtain a target recognition area; The target foreground image is obtained by cutting the line image according to the target recognition area.

[0037] In one implementation, under changeable weather conditions such as ice and snow, low light, haze, etc., the grayscale distribution of the image may be more complex. The multiple segmentation of the Otsu threshold can automatically adjust according to the different grayscale levels of the image to improve the adaptability and accuracy of target recognition. Using multiple thresholds for segmentation can more accurately identify the target area, especially in complex images, where there may be multiple target areas of different levels such as background, wires, snow, etc. This method can enhance the image segmentation effect and avoid information loss caused by a single threshold.

[0038] In one implementation, connectivity graph analysis can help identify connected areas in an image and exclude irrelevant scattered noise. For power line images, connectivity analysis can effectively identify areas of lines that are connected in a line, ignoring irrelevant debris or background information. By cutting the image according to the target identification area, the area containing only the target power line can be accurately extracted, removing the background and other irrelevant parts. In this way, subsequent image processing and analysis can focus on key content and improve processing efficiency and accuracy.

[0039] In one embodiment, performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image includes: Perform grayscale conversion on the target foreground image to obtain a target grayscale image, and perform gradient morphological operation on the target grayscale image to obtain a target edge image; Perform non-maximum suppression on the target edge image to obtain the target fusion image; The target foreground image and the target fusion image are weightedly fused to obtain a second enhanced image.

[0040] In one implementation, gradient morphological operations can emphasize the edge information of the image, especially for structures with obvious edges such as transmission lines, which can highlight the details of the edge part and help subsequent detection and analysis; non-maximum suppression can remove non-edge pixels from the edge image, making the edges in the image more detailed and clear, which can help to highlight the edge lines of the transmission line and provide a clearer outline for subsequent defect analysis and fault diagnosis.

[0041] In one implementation, the specific operation of non-maximum suppression is to use the Sobel operator to calculate the gradient, calculate the gradient direction, suppress the non-maximum along the gradient direction, retain only the local maximum gradient value, remove adjacent non-maximum pixels, and finally output the target fused image; add and divide the target foreground image and the target fused image by 2 to obtain a second enhanced image.

[0042] In one implementation, non-maximum suppression can be used to remove redundant information in edge areas, making the edges of images clearer and more precise, and avoiding the impact of errors and noise on subsequent analysis.

[0043] In one implementation, weighted fusion combines the target foreground image and the target fused image to take into account the advantages of both. The foreground image provides the overall information of the line, and the fused image provides more detailed edge information, generating a clearer and more complete image, which is helpful for subsequent line status judgment, fault detection and other tasks.

[0044] In one embodiment, substituting the second enhanced image into a preset model to obtain the target feature includes: The second enhanced image is passed through the first convolution module and the second convolution module to obtain a first convolution feature, and the first convolution feature is passed through the first C2f module to obtain a first extraction feature; The first extracted feature image is passed through the third convolution module to obtain the second convolution feature, the first convolution feature and the second convolution feature are weightedly fused to obtain the first fused feature, and the first fused feature is passed through the second C2f module to obtain the second extracted feature; The second extracted feature is passed through the first attention module to obtain the first attention feature; The second extracted feature is passed through the fourth convolution module to obtain the third convolution feature, the second convolution feature and the third convolution feature are weightedly fused to obtain the second fused feature, and the second fused feature is passed through the third C2f module to obtain the third extracted feature; The third extracted feature is passed through the second attention module to obtain the second attention feature; The third extracted feature is subjected to a feature enhancement module to obtain a first enhanced feature; The first attention feature, the second attention feature and the first enhanced feature are weightedly fused to obtain the target feature.

[0045] In one implementation, the convolution operation helps obtain effective information from the original image by extracting local features, which can effectively enhance the local features in the input image and enable the model to better capture details. Through multiple convolution modules, the model can gradually extract multi-level features in the image, from low-level features to high-level features, providing richer information for subsequent analysis.

[0046] In one implementation, the introduction of the C2f (Convolutional to Feature) module enables the features extracted by convolution to be further optimized and extracted, which can reduce the interference of irrelevant information and improve the expressiveness of the features. The first C2f module further refines the features to ensure that the model can focus on the most important parts when processing images, remove redundant information, and make the features more compact and effective.

[0047] In one implementation, the fused features are obtained by weighted fusion of features, and are all fused after being aligned based on the maximum size, and the fusion is performed for averaging.

[0048] In one implementation, the first convolution module in the scheme is (convolution kernel size: 3×3, stride: 2, number of output channels: 64), the second convolution module (convolution kernel size: 3×3, stride: 2, number of output channels: 128), the third convolution module (convolution kernel size: 3×3, stride: 2, number of output channels: 256) and the fourth convolution module (convolution kernel size: 3×3, stride: 2, number of output channels: 512); the first C2f module, the second C2f module and the third C2f module are commonly used modules in YOLOv8; the first attention module is the same as the second attention module, and the attention module consists of three branches, each of which is responsible for calculating the weighted feature map in the three dimensions of width, length and channel. In each branch, the feature map is first rotated, then processed by the pooling layer and the convolution layer, and then the attention weight is obtained by the Sigmoid activation function. Next, the rotated feature map is weighted using the attention weights obtained from the three branches and restored to the original shape of the input feature map. Finally, the weighted feature maps of the three branches are averaged to obtain the final output feature map. The feature enhancement module includes the backbone network (ResNet+deformable convolution), C2f module, convolution layer, pooling layer, fully connected layer and output layer, among which (ResNet+deformable convolution), convolution layer, pooling layer, fully connected layer and output layer are the model parameters of the deformable convolution network.

[0049] In one implementation, the second convolution further enhances the expressiveness of features, especially in complex images, and can capture deeper feature information and enhance the accuracy of subsequent processing. Through multiple convolutions, the model can understand the details of the image more comprehensively, thereby improving the model's robustness and adaptability to complex situations.

[0050] In one implementation, weighted fusion of features from different convolutional modules can combine their respective advantages, fuse low-level and high-level features, and enhance the expressiveness of information. Such fusion can ensure that the useful information extracted by each module can be effectively retained; by weighted fusion of different features, the possible deviations or deficiencies of a single module can be eliminated, making the final features more comprehensive and accurate.

[0051] In one implementation, the second C2f module further processes the fused features to further remove redundant information and improve the distinguishing ability of the features, which helps to extract more accurate and key features in subsequent analysis.

[0052] In one implementation, the attention mechanism can dynamically adjust the focus on features according to the importance of different areas, so that the model can focus on more important parts of the image, such as power lines, snow-covered areas, etc., to improve the recognition accuracy of the model; by introducing the attention module, the model can more accurately process images in complex environments such as snow, frost and other weather interferences, and improve its sensitivity to key features.

[0053] In one implementation, the fourth convolution module continues to strengthen the feature extraction of the image, especially in more complex scenarios, to further capture details and enhance the expressiveness of the model; through continuous convolution modules, the model can extract features at different levels, so that complex images can be fully analyzed, helping to make subsequent analysis more accurate.

[0054] In one implementation, by weighted fusion of the second convolutional features and the third convolutional features, the model can combine the advantages of both, eliminate their shortcomings, and improve the accuracy and expressiveness of the features. The third C2f module further refines the fused features, removes redundant information, retains the most representative parts, and improves the discriminative ability and expressiveness of the features.

[0055] In one implementation, through the second attention module, the model can further focus on the most important features in the image, especially for small changes or key areas of the target, thereby improving the accuracy of target detection; through the processing of the enhancement module, the model can maintain high accuracy in a variety of complex situations and improve the ability to detect complex faults.

[0056] In one implementation, by weighted fusion of multiple attention features and enhancement features, the model can integrate all extracted useful information to ensure that the final target feature contains all key content in the image; the weighted method of fusing different features can effectively improve the overall performance of the model, enhance the system's robustness in complex environments, reduce misjudgments, and improve the reliability of target detection.

[0057] Based on the same inventive concept, the embodiment of the present invention also provides a power transmission line monitoring device based on machine vision. Figure 2 , Figure 2 A schematic diagram of the structure of a power transmission line monitoring device based on machine vision provided by an embodiment of the present invention includes: A preprocessing module, used to obtain a line image including a target transmission line, and preprocess the line image to obtain an initial image; An edge enhancement module, used for performing edge enhancement on the initial image to obtain a first enhanced image; A target foreground image determination module, used to determine a target foreground image containing only a target power transmission line according to the line image and the first enhanced image; A boundary enhancement module, used for performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image; The target feature determination module is used to substitute the second enhanced image into a preset model to obtain the target feature, and to search a preset database according to the target feature to determine the line state of the target transmission line.

[0058] A machine vision-based power transmission line monitoring device provided in an embodiment of the present invention can highlight key features of a target power transmission line and reduce noise and interference by performing multiple image enhancements, thereby improving the accuracy of target recognition. Through image processing and preset models, the power transmission line can be identified in detail, reducing external interference caused by snow and ice accumulation, thereby improving line monitoring efficiency.

[0059] In one embodiment, the pre-processing module includes: A resolution reconstruction module, used for performing non-local mean denoising on the line image and then reconstructing the resolution to obtain a first preset image; A contrast enhancement module, used for adaptively performing local contrast enhancement on the first preset image to obtain a second preset image; The grayscale conversion module is used to perform grayscale conversion on the second preset image and then perform multi-scale bilateral filtering to obtain an initial image.

[0060] In one embodiment, the target foreground image determination module includes: A target recognition region determination module, used for performing multi-Otsu threshold and connectivity graph analysis on the first enhanced image to obtain a target recognition region; The region cutting module is used to cut the line image according to the target recognition region to obtain the target foreground image.

[0061] In one embodiment, the border enhancement module includes: A target edge image determination module is used to perform grayscale conversion on the target foreground image to obtain a target grayscale image, and to perform gradient morphological operations on the target grayscale image to obtain a target edge image; A target fusion image determination module is used to perform non-maximum suppression on the target edge image to obtain a target fusion image; The first weighted fusion module is used to perform weighted fusion on the target foreground image and the target fusion image to obtain a second enhanced image.

[0062] In one embodiment, the target feature determination module includes: A first extraction feature determination module, used for obtaining a first convolution feature from the second enhanced image through the first convolution module and the second convolution module, and obtaining a first extraction feature through the first convolution feature through the first C2f module; A second extraction feature determination module is used for obtaining a second convolution feature by passing the first extraction feature image through a third convolution module, obtaining a first fusion feature by weighted fusion of the first convolution feature and the second convolution feature, and obtaining a second extraction feature by passing the first fusion feature through a second C2f module; A first attention feature determination module, used for obtaining a first attention feature through the first attention module for the second extracted feature; A third extraction feature determination module is used for obtaining a third convolution feature by passing the second extraction feature through a fourth convolution module, obtaining a second fusion feature by weighted fusion of the second convolution feature and the third convolution feature, and obtaining a third extraction feature by passing the second fusion feature through a third C2f module; A second attention feature determination module, used for obtaining a second attention feature through the second attention module using the third extracted feature; A first enhanced feature determination module, used for obtaining a first enhanced feature through a feature enhancement module using the third extracted feature; The second weighted fusion module is used to perform weighted fusion on the first attention feature, the second attention feature and the first enhanced feature to obtain the target feature.

[0063] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for monitoring a power transmission line based on machine vision, characterized in that: The method comprises: Acquire a line image including a target power transmission line, and preprocess the line image to obtain an initial image; Performing edge enhancement on the initial image to obtain a first enhanced image; Determining a target foreground image that only includes the target power transmission line according to the line image and the first enhanced image; Performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image; Substituting the second enhanced image into a preset model to obtain target features, searching a preset database according to the target features to determine the line status of the target power transmission line.

2. A method for monitoring a power transmission line based on machine vision according to claim 1, characterized in that: Preprocessing the line image to obtain an initial image includes: Performing non-local mean denoising on the line image and then reconstructing the resolution to obtain a first preset image; Performing adaptive local contrast enhancement on the first preset image to obtain a second preset image; The second preset image is subjected to grayscale conversion and then multi-scale bilateral filtering to obtain an initial image.

3. The method for monitoring a power transmission line based on machine vision according to claim 1, characterized in that: Determining a target foreground image that only includes the target power transmission line according to the line image and the first enhanced image includes: Performing multi-Otsu threshold and connectivity map analysis on the first enhanced image to obtain a target recognition area; The line image is cut according to the target recognition area to obtain a target foreground image.

4. The method for monitoring a power transmission line based on machine vision according to claim 1, characterized in that: Performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image includes: Performing grayscale conversion on the target foreground image to obtain a target grayscale image, and performing gradient morphological operation on the target grayscale image to obtain a target edge image; Performing non-maximum suppression on the target edge image to obtain a target fused image; The target foreground image and the target fused image are weightedly fused to obtain a second enhanced image.

5. The method for monitoring a power transmission line based on machine vision according to claim 1, characterized in that: Substituting the second enhanced image into the preset model to obtain the target features includes: The second enhanced image is passed through a first convolution module and a second convolution module to obtain a first convolution feature, and the first convolution feature is passed through a first C2f module to obtain a first extraction feature; The first extracted feature image is passed through a third convolution module to obtain a second convolution feature, the first convolution feature and the second convolution feature are weightedly fused to obtain a first fused feature, and the first fused feature is passed through a second C2f module to obtain a second extracted feature; The second extracted feature is passed through a first attention module to obtain a first attention feature; The second extracted feature is passed through a fourth convolution module to obtain a third convolution feature, the second convolution feature and the third convolution feature are weightedly fused to obtain a second fused feature, and the second fused feature is passed through a third C2f module to obtain a third extracted feature; The third extracted feature is passed through a second attention module to obtain a second attention feature; The third extracted feature is subjected to a feature enhancement module to obtain a first enhanced feature; The first attention feature, the second attention feature and the first enhanced feature are weightedly fused to obtain a target feature.

6. A power transmission line monitoring device based on machine vision, characterized in that: The device comprises: A preprocessing module, used to obtain a line image including a target power transmission line, and preprocess the line image to obtain an initial image; An edge enhancement module, used for performing edge enhancement on the initial image to obtain a first enhanced image; a target foreground image determination module, configured to determine a target foreground image containing only the target power transmission line according to the line image and the first enhanced image; A boundary enhancement module, used for performing boundary enhancement on the target foreground image by a preset algorithm to obtain a second enhanced image; The target feature determination module is used to substitute the second enhanced image into a preset model to obtain the target feature, and to search a preset database according to the target feature to determine the line state of the target transmission line.

7. The power transmission line monitoring device based on machine vision according to claim 6, characterized in that: The preprocessing module comprises: A resolution reconstruction module, used for performing non-local mean denoising on the line image and then reconstructing the resolution to obtain a first preset image; A contrast enhancement module, configured to perform adaptive local contrast enhancement on the first preset image to obtain a second preset image; A grayscale conversion module is used to perform grayscale conversion on the second preset image and then perform multi-scale bilateral filtering to obtain an initial image.

8. The power transmission line monitoring device based on machine vision according to claim 6, characterized in that: The target foreground image determination module comprises: A target recognition region determination module, used for performing multi-Otsu threshold and connectivity graph analysis on the first enhanced image to obtain a target recognition region; A region cutting module is used to cut the line image according to the target recognition region to obtain a target foreground image.

9. The power transmission line monitoring device based on machine vision according to claim 6, characterized in that: The boundary enhancement module comprises: A target edge image determination module is used to perform grayscale conversion on the target foreground image to obtain a target grayscale image, and perform gradient morphological operation on the target grayscale image to obtain a target edge image; A target fused image determination module, used for performing non-maximum suppression on the target edge image to obtain a target fused image; The first weighted fusion module is used to perform weighted fusion on the target foreground image and the target fused image to obtain a second enhanced image.

10. The power transmission line monitoring device based on machine vision according to claim 6, characterized in that: The target feature determination module comprises: A first extraction feature determination module, used for obtaining a first convolution feature from the second enhanced image through a first convolution module and a second convolution module, and obtaining a first extraction feature through a first C2f module from the first convolution feature; A second extraction feature determination module, used for obtaining a second convolution feature by passing the first extraction feature image through a third convolution module, obtaining a first fusion feature by weighted fusion of the first convolution feature and the second convolution feature, and obtaining a second extraction feature by passing the first fusion feature through a second C2f module; A first attention feature determination module, used for obtaining a first attention feature through the first attention module from the second extracted feature; A third extraction feature determination module, used for obtaining a third convolution feature by passing the second extraction feature through a fourth convolution module, obtaining a second fusion feature by weighted fusion of the second convolution feature and the third convolution feature, and obtaining a third extraction feature by passing the second fusion feature through a third C2f module; A second attention feature determination module, used for obtaining a second attention feature through the second attention module from the third extracted feature; A first enhanced feature determination module, used for obtaining a first enhanced feature through the feature enhancement module of the third extracted feature; The second weighted fusion module is used to perform weighted fusion on the first attention feature, the second attention feature and the first enhanced feature to obtain a target feature.

Images