Power transmission line insulator defect detection method of adaptive enhancement module

Abstract

The invention provides a power transmission line insulator defect detection method based on an adaptive enhancement module, and the method comprises the steps: obtaining high-altitude power transmission line insulator image data: employing a high-definition camera of an unmanned plane to shoot insulator defect images at different angles, and carrying out the preprocessing, and forming an insulator defect image data set; constructing an insulator defect target detection model: constructing a high-altitude power transmission line insulator defect target detection model based on DyHead and an information adaptive enhancement module; training an insulator defect target detection model; and insulator image defect detection: inputting the preprocessed to-be-detected insulator defect image into the trained target detection model to obtain classification and positioning structures of various insulator defects in the defect image, and detecting defect positions.

Description

Technical Field

[0001] The invention relates to the field of power transmission line defect detection, and in particular to a power transmission line insulator defect detection method with an adaptive enhancement module. Background Art

[0002] Overhead transmission lines are an important part of modern power systems, and their stable operation is directly related to the safe transmission of electricity. However, with the long-term operation of transmission lines and the influence of external environments such as wind, rain, ice, snow, and dirt, insulators on the lines are prone to various defects, such as cracks, breakage, and pollution. These defects may cause the performance of insulators to deteriorate, and in severe cases may even cause power system failures and affect the normal operation of the power grid. Therefore, timely and accurate detection and diagnosis of insulator defects are crucial to ensure the safety and reliability of the power system.

[0003] Traditional insulator defect detection methods often rely on manual inspections and simple electrical tests, which usually require professional staff to work at height to carry out repairs and replacements. This is not only inefficient and costly, but also easily affected by human factors.

[0004] In recent years, deep learning technology has made significant progress in the field of image recognition and classification, providing a new solution for the detection of insulator defects in high-altitude transmission lines. And with the development of drone technology, in actual applications, researchers usually use drones to shoot high-altitude transmission lines first. The high-definition camera carried by the drone can capture high-resolution images, and then use the pre-trained deep learning model to classify and identify insulator defects. However, when the background is complex and diverse, and the insulators are of different types and irregular shapes, it will lead to weak resolution, missed detection and false detection problems, making it difficult for the model to detect the target.

[0005] Therefore, how to make the network have stronger detection and differentiation capabilities for insulator defects and improve the detection accuracy of insulators in complex scenes has become a technical problem that needs to be solved urgently. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and provide a method for detecting defects in insulators of power transmission lines with an adaptive enhancement module, thereby solving the problems of difficulty in identifying defect models of insulators of power transmission lines and low detection accuracy in the prior art.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] The present invention provides a method for detecting defects in insulators of a power transmission line using an adaptive enhanced module, comprising:

[0009] S1. Acquisition of image data of insulators of high-altitude transmission lines: Use the high-definition camera of the drone to take pictures of insulator defects at different angles, perform preprocessing, and form an insulator defect image dataset;

[0010] S2. Constructing an insulator defect target detection model: an insulator defect target detection model for high-altitude transmission lines based on DyHead and information adaptive enhancement module;

[0011] S3, training of insulator defect target detection model;

[0012] S4. Insulator image defect detection: The preprocessed insulator defect image to be detected is input into the trained target detection model to obtain the classification and positioning structure of various insulator defects in the defect image and detect the defect location.

[0013] Further, in said S2, constructing an insulator defect target detection model includes a deep residual network, a feature fusion network and a detection head network;

[0014] The deep residual network is based on the construction of the deep convolutional neural network ResNet50, specifically:

[0015] The deep convolutional neural network ResNet50 introduces a residual block, which includes a convolutional layer and an activation function, specifically:

[0016] y=F(x)+x (1);

[0017] Among them, x is the input; F(x) is the output after the convolution layer and activation function, through the addition operation;

[0018] The feature fusion network is based on the construction of the FPN network and the information adaptive enhancement module, specifically:

[0019] In the information adaptive enhancement module, the feature map is processed by adaptive average pooling AdaptiveAvgPool2d(×). For a feature map X of a certain layer, ij :

[0020]

[0021] in, is the feature after pooling; i, j are the input feature level ordinal and enhanced feature level ordinal respectively; σ and μ are the mean and standard deviation in x; ε is a small positive number added for stable division; γ and β are trainable parameters;

[0022] For features lower than the feature layer, a 3×3 convolution with a stride of 2 is used to reduce the feature until the feature scale is the same; for features higher than the feature layer, a 3×3 transposed convolution with a stride of 2 is used to enlarge the feature until the feature scale is the same:

[0023]

[0024] Among them, conv(·) is a 3×3 convolution with a stride of 2; conv trans (·) is a 3×3 transposed convolution with a stride of 2; is the output feature;

[0025] right Perform point-by-point convolution PWconv(·) to obtain the weight parameter W ij , to represent the attention of the feature map of the level to a certain level:

[0026]

[0027] After obtaining the initial weight parameters of all layers, they are mapped to the range (0, 1) through the softmax function to obtain the final weight parameters

[0028]

[0029] Among them, a and b are the level ordinal numbers;

[0030] The scaled hierarchical features and the corresponding final weight parameters are weighted and summed to obtain the enhanced feature map Y j :

[0031]

[0032] in, is the element-wise multiplication; c is the total number of all levels;

[0033] Add a shortcut connection path to the enhanced features to obtain the final output features

[0034] Among them, X j It is the input feature of the information adaptive enhancement module.

[0035] Furthermore, the detection head network is constructed based on the DyHead module;

[0036] The DyHead module includes a scale-aware attention module, a space-aware attention module, and a task-aware attention module;

[0037] The scale-aware attention module π L(·) Dynamically fuse features based on the semantic importance of features at different scales:

[0038]

[0039] Where F is the input feature vector; S and C are the total number of width-height products and the total number of channels, respectively; f(·) is a linear function approximated by a 1×1 convolution; α(·) is a hard-sigmoid function;

[0040] The spatial perception attention module π S (·) Specifically:

[0041]

[0042] Where L is the number of feature levels; l is the number of the current feature level; K is the number of sparse sampling positions; k is the number of the current sparse sampling position; p k +Δp k Focus on a specific area by position shift; Δm k is the position p k A self-learnable importance metric factor;

[0043] The task-aware attention module π C (·) Select different tasks by dynamically opening or closing feature channels:

[0044] π L (F)·F=max(α 1 (F)·F+β 1 (F),α 2 (F)·F+β 2 (F));

[0045] Among them, [α 1 ,β 1 ,α 2 ,β 2 ] T =θ(·) is a hyperfunction used to learn and control the activation threshold.

[0046] Furthermore, in S3, the training of the insulator defect target detection model includes processing of the data set, definition of the loss function and training of the model parameters:

[0047] The definition of the loss function includes classification loss and regression loss;

[0048] The classification loss uses the cross entropy loss L cls :

[0049]

[0050] Among them, yi is the label of the real category; p i is the category probability predicted by the model; C is the target for each prediction;

[0051] The regression loss uses the GIoU loss function:

[0052]

[0053] Among them, A and B are the areas of the predicted bounding box and the true bounding box respectively, and C is the area of ​​the minimum enclosing rectangle that can contain A and B.

[0054] Furthermore, the training of the model parameters adopts the method of stochastic gradient descent, and only part of the data is used for calculation each time the parameters are updated, and the parameters are updated by calculating the gradient of the loss:

[0055]

[0056] Among them, θ is the model parameter; η is the learning rate; is the gradient of the loss function with respect to the parameters.

[0057] The beneficial effects of the present invention are: by utilizing the information adaptive enhancement module, various types of insulator defects can be effectively learned;

[0058] For the insulator defect detection task, a multi-dimensional attention mechanism is introduced into the method, which can accurately identify the location of insulator defects in complex scenes and at different angles. Compared with existing methods, the recognition accuracy is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 The present invention is a flow chart of a method for detecting defects in insulators of a transmission line using an adaptive enhanced module. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0061] See also Figure 1 , a transmission line insulator defect detection method with an adaptive enhanced module, comprising:

[0062] S1. Acquisition of image data of insulators of high-altitude transmission lines: Use the high-definition camera of the drone to take pictures of insulator defects at different angles, perform preprocessing, and form an insulator defect image dataset;

[0063] In a specific implementation, obtaining the image data of the insulator of the overhead transmission line includes the following steps:

[0064] First, fully plan the terrain, meteorological conditions and line conditions of the transmission line; specify the flight route through the existing transmission line map; ensure that the drone and the load are in good condition and have sufficient battery power;

[0065] After that, choose a multi-rotor drone or fixed-wing drone equipped with a high-definition camera, fly along the preset route, and maintain a safe flight altitude and distance;

[0066] Finally, the drone hovers at the best shooting angle at the insulator location to take high-definition photos. After the shooting is completed, the data is downloaded from the drone to the computer to build an insulator defect dataset.

[0067] Before the UAV flies, a route survey is carried out to formulate a suitable flight route. The UAV flies automatically on the set route while maintaining a fixed distance from the transmission line to ensure that images can be accurately obtained at each insulator. After obtaining the image data, the defect locations of the insulators are marked to create a data set, which is then randomly selected in a ratio of 8:2 as the training set and test set, respectively.

[0068] S2. Constructing an insulator defect target detection model: an insulator defect target detection model for high-altitude transmission lines based on DyHead and information adaptive enhancement module;

[0069] S3, training of insulator defect target detection model;

[0070] S4. Insulator image defect detection: The preprocessed insulator defect image to be detected is input into the trained target detection model to obtain the classification and positioning structure of various insulator defects in the defect image and detect the defect location.

[0071] In S2, an insulator defect target detection model is constructed including a deep residual network, a feature fusion network and a detection head network;

[0072] In a specific implementation, a deep parameter network is used to extract features for an image, which are then input into a feature fusion network to further fuse the rich feature information, and finally enter a detection head network to complete the classification and positioning of the task.

[0073] The deep residual network is based on the construction of the deep convolutional neural network ResNet50.

[0074] Among them, the deep residual network is based on the deep convolutional neural network ResNet50, which consists of a 50-layer deep network. It is used to solve the problems of gradient disappearance and gradient explosion in deep neural networks, so that the network can be trained deeper.

[0075] Specifically:

[0076] (1) Construct a deep residual network.

[0077] The deep residual network uses ResNet50, which is mainly composed of convolutional layers, batch normalization, ReLU activation function, pooling layers and residual blocks. It can be divided into the following stages:

[0078] 11) The first convolution layer.

[0079] The first convolution layer uses a large 7×7 convolution kernel, 64 channels, and a step size of 2, which is mainly used to perform preliminary feature extraction on the input image. It is followed by a 3×3 maximum pooling layer with a step size of 2 to reduce the size of the feature map.

[0080] 12) Convolutional layers and residual blocks.

[0081] ResNet50 consists of 4 stages, each of which consists of multiple residual blocks, and each residual block has multiple convolutional layers:

[0082] Stage 1: 3 residual blocks, each block consists of 1×1 convolution, 3×3 convolution and 1×1 convolution component;

[0083] Stage 2: 4 residual blocks, the number of convolution channels gradually increases to 64, 128, 256 and 512 channels respectively;

[0084] Stage 3: 6 residual blocks, gradually increasing the depth of the network;

[0085] Stage 4: 3 residual blocks to extract deep features;

[0086] 13) Quick connection path

[0087] The deep convolutional neural network ResNet50 introduces a residual block, which includes a convolutional layer and an activation function, and directly adds the input to the output, allowing the network to directly transmit the input signal, avoiding the gradient problem caused by the deep network.

[0088] Specifically:

[0089] y=F(x)+x (1);

[0090] Among them, x is the input; F(x) is the output after the convolution layer and activation function. Through the addition operation, the network can be more easily optimized and trained.

[0091] 14) Global Average Pooling

[0092] After the last convolution operation, global average pooling is used to reduce the dimension of the feature map to a fixed-size feature vector;

[0093] 15) Fully connected layer

[0094] After global average pooling, it is connected to a fully connected layer and outputs a vector of length equal to the number of categories, which is usually the number of image categories in classification tasks.

[0095] 16) Activation function and normalization.

[0096] Each convolutional layer is followed by a ReLU activation function and batch normalization, which together help speed up the training process and stabilize the performance of the model.

[0097] The feature fusion network is based on the construction of FPN network and information adaptive enhancement module.

[0098] Among them, the FPN network consists of a bottom-up feature extraction path and a top-down feature fusion path, and then the information adaptive enhancement module is used to further enhance the fused features;

[0099] The bottom-up feature extraction path uses the deep residual network ResNet50 as the backbone network. The input image is processed through multiple consecutive convolution kernel pooling operations to generate feature maps of different scales, namely C2, C3, C4, and C5. These feature images gradually reduce the resolution and have deeper and deeper semantic information.

[0100] In a specific implementation, FPN uses several important feature layers of ResNet50 as the input of FPN, which come from convolution blocks C2, C3, C4, and C5 of different depths. The resolutions of these feature maps decrease successively, while the number of channels increases successively, which are 1 / 4, 1 / 8, 1 / 16, and 1 / 32 of the input resolution respectively.

[0101] The top-down feature fusion path amplifies the high-level feature map by upsampling, and then fuses it with the feature map of the corresponding level from the bottom to the top. The fusion method is element-by-element addition. In order to ensure that the fused feature maps have the same number of channels, 1╳1 convolution is used to exchange the channels of the bottom-up feature maps C2, C3, C4, and C5 to match the number of channels of the upsampled feature maps.

[0102] In a specific implementation, upsampling is performed starting from C5, and fused layer by layer with the feature maps of the shallow layers. The output P5 of C5 is upsampled by a factor of 2 (Upsample) to match its resolution with that of C4:

[0103] P5 ↑ =Upsample(P5);

[0104] Use 1x1 convolution to transform the channel of C4 and reduce the number of channels to 256:

[0105] L4 = conv(C4, 256);

[0106] Add the upsampled P5 to the 1x1 convolution-processed C4 to generate P4:

[0107] P4=P5 ↑ +L4;

[0108] In this way, P4 inherits deep semantic information while maintaining a high spatial resolution. By repeating similar steps, feature maps of different scales of P2, P3, P4, and P5 can be generated.

[0109] This path can transfer high-level semantic feature maps to low-level feature maps layer by layer, so that the deep semantic information in high-level features can be integrated with the low-level detail features.

[0110] The information adaptive enhancement module is used to further process the features after top-down path fusion, enhance the information content of the input features, and enhance the low-resolution high-level features with the target position information of the low-level features. At the same time, each feature is quickly linked to ensure the original information of each feature while enhancing the perception of other layers of information, thereby improving the model's ability to recognize the type and location of insulator defects in the responsible environment, improving the representativeness of insulator defect features, and facilitating the subsequent processing of the DyHead detection head.

[0111] Specifically, in the information adaptive enhancement module, the feature map is processed using adaptive average pooling AdaptiveAvgPool2d(×). For a feature map X of a certain layer, ij :

[0112]

[0113] in, is the feature after pooling; i, j are the input feature level ordinal and enhanced feature level ordinal respectively; σ and μ are the mean and standard deviation in x; ε is a small positive number added for stable division; γ and β are trainable parameters;

[0114] For features lower than the feature layer, a 3×3 convolution with a stride of 2 is used to reduce the feature until the feature scale is the same; for features higher than the feature layer, a 3×3 transposed convolution with a stride of 2 is used to enlarge the feature until the feature scale is the same:

[0115]

[0116] Among them, conv(·) is a 3×3 convolution with a stride of 2; conv trans (·) is a 3×3 transposed convolution with a stride of 2; is the output feature;

[0117] right Perform point-by-point convolution PWconv(·) to obtain the weight parameter W ij , to represent the attention of the feature map of the level to a certain level:

[0118]

[0119] After obtaining the initial weight parameters of all layers, they are mapped to the range (0, 1) through the softmax function to obtain the final weight parameters

[0120]

[0121] Among them, a and b are the level ordinal numbers;

[0122] The scaled hierarchical features and the corresponding final weight parameters are weighted and summed to obtain the enhanced feature map Y j :

[0123]

[0124] in, is the element-wise multiplication; c is the total number of all levels;

[0125] Add a shortcut connection path to the enhanced features to obtain the final output features

[0126] Among them, X j It is the input feature of the information adaptive enhancement module.

[0127] The detection head network is based on the construction of the DyHead module;

[0128] The DyHead module includes a scale-aware attention module, a space-aware attention module and a task-aware attention module; the scale-aware attention module, the space-aware attention module and the task-aware attention module are stacked to realize the attention operation on the input features, and enhance the perception ability of the target detector for the scale-aware attention module, the space-aware attention module and the task-aware attention module.

[0129] The scale-aware attention module π L (·) Dynamically fuse features based on the semantic importance of features at different scales:

[0130]

[0131]

[0132] Where F is the input feature vector; S and C are the total number of width-height products and the total number of channels, respectively; f(·) is a linear function approximated by a 1×1 convolution; α(·) is a hard-sigmoid function;

[0133] Among them, the spatial perception attention module consists of two steps: first, deformable convolution is used for sparsification, and then features are aggregated across levels at the same spatial position.

[0134] The spatial perception attention module π S (·) Specifically:

[0135]

[0136] Where L is the number of feature levels; l is the number of the current feature level; K is the number of sparse sampling positions; k is the number of the current sparse sampling position; p k +Δp k Focus on a specific area by position shift; Δm k is the position p k A self-learnable importance metric factor;

[0137] The task-aware attention module π C (·) Select different tasks by dynamically opening or closing feature channels:

[0138] π L (F)·F=max(α 1 (F)·F+β 1 (F),α 2 (F)·F+β 2 (F));

[0139] Among them, [α 1 ,β 1 ,α 2 ,β 2 ] T =θ(·) is a hyperfunction used to learn and control the activation threshold. Specifically, global pooling is first performed in the L×S dimension, then two fully connected layers and a normalization layer are used, and finally the sigmoid function is used for normalized output.

[0140] The dataset was processed by taking pictures of insulators of high-altitude power transmission lines using drones. Professionals were then asked to manually annotate the insulators and their defects in the images, including the bounding boxes of the insulators and the labels of the defect types. The images were then converted into a dataset in COCO format.

[0141] In S3, the training of the insulator defect target detection model includes the processing of the data set, the definition of the loss function and the training of the model parameters:

[0142] The definition of the loss function includes classification loss and regression loss;

[0143] The classification loss uses the cross entropy loss L cls , to measure the difference between the predicted probability of each category and the true label;

[0144]

[0145] Among them, y i is the label of the real category; p i is the category probability predicted by the model; C is the target for each prediction;

[0146] The regression loss uses the GIoU loss function:

[0147]

[0148] Among them, A and B are the areas of the predicted bounding box and the true bounding box respectively, and C is the area of ​​the minimum enclosing rectangle that can contain A and B.

[0149] The training of the model parameters adopts the method of stochastic gradient descent. Only part of the data is used for calculation each time the parameters are updated, and the parameters are updated by calculating the gradient of the loss:

[0150]

[0151] Among them, θ is the model parameter; η is the learning rate; is the gradient of the loss function with respect to the parameters.

[0152] The model parameter update process is repeated in multiple training rounds. In each round, the model iterates multiple times on all training data until the loss function converges or the preset number of training rounds is reached.

[0153] The above-mentioned embodiments only express the implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be based on the attached claims.

Claims

1. A method for detecting defects in insulators of power transmission lines with an adaptive enhanced module, characterized in that: include: S1. Acquisition of image data of insulators of high-altitude transmission lines: Use the high-definition camera of the drone to take pictures of insulator defects at different angles, perform preprocessing, and form an insulator defect image dataset; S2. Constructing an insulator defect target detection model: an insulator defect target detection model for high-altitude transmission lines based on DyHead and information adaptive enhancement module; S3, training of insulator defect target detection model; S4. Insulator image defect detection: The preprocessed insulator defect image to be detected is input into the trained target detection model to obtain the classification and positioning structure of various insulator defects in the defect image and detect the defect location.

2. The method for detecting defects in insulators of a power transmission line using an adaptive enhanced module according to claim 1, characterized in that: In S2, an insulator defect target detection model is constructed including a deep residual network, a feature fusion network and a detection head network; The deep residual network is based on the construction of the deep convolutional neural network ResNet50, specifically: The deep convolutional neural network ResNet50 introduces a residual block, which includes a convolutional layer and an activation function, specifically: y=F(x)+x (1); Among them, x is the input; F(x) is the output after the convolution layer and activation function, through the addition operation; The feature fusion network is based on the construction of the FPN network and the information adaptive enhancement module, specifically: In the information adaptive enhancement module, the feature map is processed by adaptive average pooling AdaptiveAvgPool2d(×). For a feature map X of a certain layer, ij : in, is the feature after pooling; i, j are the input feature level ordinal and enhanced feature level ordinal respectively; σ and μ are the mean and standard deviation in x; ε is a small positive number added for stable division; γ and β are trainable parameters; For features lower than the feature layer, a 3×3 convolution with a stride of 2 is used to reduce the feature until the feature scale is the same; for features higher than the feature layer, a 3×3 transposed convolution with a stride of 2 is used to enlarge the feature until the feature scale is the same: Among them, conv(·) is a 3×3 convolution with a stride of 2; conv trans (·) is a 3×3 transposed convolution with a stride of 2; is the output feature; right Perform point-by-point convolution PWconv(·) to obtain the weight parameter W ij , to represent the attention of the feature map of the level to a certain level: After obtaining the initial weight parameters of all layers, they are mapped to the range (0, 1) through the softmax function to obtain the final weight parameters Among them, a and b are the level ordinal numbers; The scaled hierarchical features and the corresponding final weight parameters are weighted and summed to obtain the enhanced feature map Y j : in, is the element-wise multiplication; c is the total number of all levels; Add a shortcut connection path to the enhanced features to obtain the final output features 3. The method for detecting defects in insulators of a power transmission line using an adaptive enhanced module according to claim 2, characterized in that: The detection head network is based on the construction of the DyHead module; The DyHead module includes a scale-aware attention module, a space-aware attention module, and a task-aware attention module; The scale-aware attention module π L (·) Dynamically fuse features based on the semantic importance of features at different scales: Where F is the input feature vector; S and C are the total number of width-height products and the total number of channels, respectively; f(·) is a linear function approximated by a 1×1 convolution; α(·) is a hard-sigmoid function; The spatial perception attention module π S (·) Specifically: Where L is the number of feature levels; l is the number of the current feature level; K is the number of sparse sampling positions; k is the number of the current sparse sampling position; p k +Δp k Focus on a specific area by position shift; Δm k is the position p k A self-learnable importance metric factor; The task-aware attention module π C (·) Select different tasks by dynamically opening or closing feature channels: p L (F)·F=max(α 1 (F)·F+β 1 (F),a 2 (F)·F+β 2 (F)); Among them, [α 1 ,β 1 ,α 2 ,β 2 ] T =θ(·) is a hyperfunction used to learn and control the activation threshold.

4. The method for detecting defects in insulators of a power transmission line using an adaptive enhanced module according to claim 3, characterized in that: In S3, the training of the insulator defect target detection model includes the processing of the data set, the definition of the loss function and the training of the model parameters: The definition of the loss function includes classification loss and regression loss; The classification loss uses the cross entropy loss L cls : Among them, y i is the label of the real category; p i is the category probability predicted by the model; C is the target for each prediction; The regression loss uses the GIoU loss function: Among them, A and B are the areas of the predicted bounding box and the true bounding box respectively, and C is the area of ​​the minimum enclosing rectangle that can contain A and B.

5. The method for detecting defects in insulators of a power transmission line using an adaptive enhanced module according to claim 4, characterized in that: The training of the model parameters adopts the method of stochastic gradient descent. Only part of the data is used for calculation each time the parameters are updated, and the parameters are updated by calculating the gradient of the loss: Among them, θ is the model parameter; η is the learning rate; is the gradient of the loss function with respect to the parameters.

Images