Distribution network point cloud segmentation method based on environment fusion

Abstract

The present invention discloses a distribution network point cloud segmentation method based on environmental fusion, which relates to the technical field of target segmentation. The method comprises: collecting point cloud data, including fused environmental point cloud data. A hierarchical edge perception network is obtained, which comprises a local embedding module, a global embedding module, an attention mechanism module, and a multi-level downsampling and upsampling module. The jump connection between the attention mechanism module and the multi-level downsampling constitutes a multi-level network. The point cloud data is input into the hierarchical edge perception network, the local embedding module captures the local features and neighborhood features of each point, the global embedding module identifies the global features, and the attention mechanism module is used to fuse the local and global features to focus on the segmentation task features. Through multiple downsampling and upsampling of the multi-level network, the environmental features and the distribution network features are segmented, and the segmentation results of the distribution network are output. Thereby, the technical effect of improving the efficiency and accuracy of distribution network point cloud segmentation is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of target segmentation, and in particular to a distribution network point cloud segmentation method based on environment fusion. Background Art

[0002] Point cloud data has become an important tool in the maintenance and management of smart grids and distribution networks. Point cloud data can accurately describe the three-dimensional structure of the distribution network, providing rich information for the design, monitoring and maintenance of the power grid. However, with the increasing complexity of distribution networks, especially the more complex structure of distribution networks, the types of distribution networks are different, and the environments in which different types of distribution networks are located are also complex and changeable. Existing automatic segmentation methods for power point clouds are mainly suitable for high-voltage distribution networks with obvious features and low interference. The classification effect is poor for relatively low and complex distribution lines or certain low-voltage transmission lines, and there are technical problems such as low segmentation efficiency and high segmentation cost. Summary of the Invention

[0003] The present invention provides a distribution network point cloud segmentation method based on environmental fusion to solve the technical problems of low segmentation efficiency and high segmentation cost in the existing technology, and achieves the technical effect of improving the efficiency and accuracy of distribution network point cloud segmentation.

[0004] The distribution network point cloud segmentation method based on environment fusion provided by the present invention includes:

[0005] Collect distribution network point cloud data, wherein the point cloud data includes environmental point cloud data, and the environmental point cloud data is integrated with the distribution network point cloud data.

[0006] Obtain a pre-built hierarchical edge-perceiving network, wherein the hierarchical edge-perceiving network includes a local embedding module, a global embedding module, an attention mechanism module, global-to-local downsampling, and upsampling, wherein the attention mechanism module, global-to-local downsampling, and upsampling construct a multi-level network, and the multi-level network contains multiple jump connections, and the jump connection is a jump between the attention mechanism module and the global-to-local downsampling.

[0007] The distribution network point cloud data is input into the hierarchical edge perception network, the local features and local neighborhood features of each point in the distribution network point cloud data are captured by the local embedding module, global features are recognized by the global embedding module, the local features, local neighborhood features and global features are fused by the attention mechanism module, the segmentation task features are focused on, multiple downsampling and upsampling are performed through the multiple jump connections contained in the multi-level network, the environmental features and the distribution network features are segmented, and the segmentation results of the distribution network are output.

[0008] In a feasible implementation, the local embedding module captures the local features and local neighborhood features of each point in the distribution network point cloud data, including:

[0009] By using distance metric, the K neighboring points of each point in the point cloud data are determined.

[0010] The feature vectors of the K neighbor points are replicated K times to construct a feature matrix.

[0011] The K copied feature matrices are spliced ​​according to the channel dimension to obtain a stacked feature matrix, which includes the local features of the original point and the neighborhood.

[0012] The stacked feature matrix is ​​used as output to obtain the local features and local neighborhood features.

[0013] In a feasible implementation, before splicing the K copied feature matrices according to the channel dimension, the method further includes:

[0014] Configure the multi-scale K value.

[0015] Based on the multi-scale K value, multi-scale K neighbor points are determined.

[0016] Feature insertion is performed based on the feature vectors of the multi-scale K neighboring points to form multiple new feature dimensions.

[0017] Perform visual evaluation on the multiple new feature dimensions, determine the relationship between the feature distribution and the target feature variable, and screen new feature dimensions.

[0018] The new feature dimension is added to the concatenated feature.

[0019] In a feasible implementation, the fusing of local features, local neighborhood features, and global features through the attention mechanism module includes:

[0020] According to the local features and global features, they are mapped into query vectors, key vectors, and value vectors through a weight matrix.

[0021] Calculate the similarity scores between the query vector and all key vectors to obtain the attention score.

[0022] The similarity scores between each query vector and all key vectors are normalized by a normalization function, and a normalized weight distribution relationship between each query vector and the key vector is established.

[0023] The normalized weight is multiplied by the corresponding value vector to obtain the attention mechanism output.

[0024] In one feasible implementation, the normalized weight is multiplied by the corresponding value vector to obtain the attention mechanism output, including:

[0025] By formula: Calculate and obtain the output of the attention mechanism, where Q is the query vector, K is the key vector, V is the value vector, is the scaling factor.

[0026] In a feasible implementation, after outputting the segmentation result of the power distribution network, the method further includes:

[0027] The distribution line is identified based on the segmentation result of the distribution network, and the distribution location of the distribution lines is extracted.

[0028] A line smoothness test is performed based on the distribution line positioning to obtain a smoothness test result.

[0029] Abnormal nodes are identified according to the smoothness detection result.

[0030] When the abnormal node does not exist, a segmentation result of the power distribution network is determined.

[0031] In a feasible implementation, after identifying abnormal nodes, the method further includes:

[0032] When the abnormal node exists, the abnormal power distribution line and adjacent power distribution lines are obtained.

[0033] Alignment compensation for abnormal nodes on the line is performed according to the abnormal distribution line, and compensation characteristics are determined.

[0034] Based on the compensation feature, the abnormal environment feature is deducted.

[0035] Environmental features are extracted based on the adjacent power distribution lines, and correlation analysis is performed using the environmental features and abnormal environmental features to determine correlation identification features.

[0036] The abnormal environment feature is re-segmented using the associated identification feature, and the distribution line feature is restored and re-segmented based on the compensation feature.

[0037] In a feasible implementation, extracting environmental features based on the adjacent power distribution lines, performing correlation analysis using the environmental features and abnormal environmental features, and determining correlation identification features include:

[0038] According to the environmental features extracted from the adjacent power distribution lines, a feature initial point is set, where the feature initial point is a cluster of pixel points whose distance from the abnormal node meets the distance requirement.

[0039] Determine the neighborhood pixels of the initial feature point.

[0040] Pixel growth information is fitted based on the feature initial point, where the pixel growth information includes grayscale value, color, texture, and gradient.

[0041] Taking the feature initial point as the starting point, growth coverage judgment is performed on neighboring pixels based on the pixel growth information, and the pixel area that meets the pixel growth information is used as a new feature initial point.

[0042] New pixel growth information is fitted based on the new feature initial point, and the pixel growth information is iteratively used to determine the growth coverage of neighboring pixels until the abnormal environment feature is found, and the associated identification feature is determined based on the current pixel growth information.

[0043] The present invention discloses a distribution network point cloud segmentation method based on environmental fusion, comprising: collecting distribution network point cloud data and environmental point cloud data, and fusing the two. A pre-built hierarchical edge perception network is obtained, which includes a local embedding module, a global embedding module, an attention mechanism module, global to local downsampling and upsampling in a multi-level network, and contains multiple jump connections. The distribution network point cloud data is input into the hierarchical edge perception network, the local features and local neighborhood features of each point are captured by the local embedding module, the global feature recognition is performed by the global embedding module, the local and global features are fused by the attention mechanism module, and the segmentation task features are focused. Through multiple downsampling and upsampling in the multi-level network, the environmental features and the distribution network features are segmented, and the segmentation results of the distribution network are output. The distribution network point cloud segmentation method based on environmental fusion disclosed by the present invention solves the technical problems of low segmentation efficiency and high segmentation cost, and achieves the technical effect of improving the efficiency and accuracy of distribution network point cloud segmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the distribution network point cloud segmentation method based on environment fusion of the present invention.

[0045] Figure 2 This is a structural diagram of the hierarchical edge perception network in the distribution network point cloud segmentation method based on environment fusion of the present invention.

[0046] Explanation of the accompanying symbols: local embedding module LE, global embedding module GE, attention mechanism module TA, global to local downsampling GLD, upsampling UPS, hierarchical edge perception network HEA. DETAILED DESCRIPTION

[0047] The technical solutions provided in the embodiments of the present invention are designed to address the technical problems of high latency, lack of real-time performance, incomplete monitoring, and insufficient predictability in the prior art. The overall approach adopted is as follows:

[0048] First, the point cloud data of the distribution network is collected, which includes the environmental point cloud data, and these environmental point cloud data are fused with the distribution network point cloud data.

[0049] Next, we obtain a pre-built hierarchical edge-aware network, which includes a local embedding module, a global embedding module, an attention mechanism module, and global-to-local downsampling and upsampling. These modules, along with global-to-local downsampling and upsampling, form a multi-layer network with numerous skip connections, primarily between the attention mechanism module and the global-to-local downsampling.

[0050] The point cloud data of the power distribution network is then fed into this hierarchical edge-aware network. The local embedding module enables the network to capture the local features and local neighborhood features of each point; the global embedding module is used to identify global features. The attention mechanism module is responsible for fusing local features, local neighborhood features, and global features to focus on the features required for the segmentation task.

[0051] Finally, through multiple downsampling and upsampling in the multi-level network, the network is able to segment the environmental features and distribution network features, and output the segmentation results of the distribution network.

[0052] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention, rather than all, are shown in the drawings.

[0053] Figure 1 The figure is a flow chart of a method for segmenting a distribution network point cloud based on environment fusion according to the present invention, wherein the method comprises:

[0054] Collecting distribution network point cloud data, wherein the point cloud data includes environmental point cloud data, and the environmental point cloud data is integrated with the distribution network point cloud data;

[0055] Optionally, point cloud data of the distribution network can be collected using LiDAR, drones, or ground-based mobile scanning systems. The collected point cloud data includes both point cloud data related to the distribution network and point cloud data of the surrounding environment. The point cloud data related to the distribution network includes the 3D coordinate information of the distribution network's towers, conductors, transformers, and other ancillary equipment; the environmental point cloud data includes the 3D coordinate information of terrain, buildings, trees, and other natural or man-made obstacles.

[0056] Specifically, the environmental point cloud data and the distribution network point cloud data are mutually integrated point cloud data. In other words, the collected distribution network point cloud data contains the point cloud data of the distribution network neighborhood environmental objects collected synchronously, and the data points in the collected distribution network point cloud data have not been clearly classified and are in a mixed state.

[0057] Obtain a pre-built hierarchical edge-aware network, wherein the hierarchical edge-aware network includes a local embedding module, a global embedding module, an attention mechanism module, global-to-local downsampling, and upsampling, wherein the attention mechanism module, global-to-local downsampling, and upsampling construct a multi-level network, and the multi-level network contains multiple jump connections, wherein the jump connection is a jump between the attention mechanism module and the global-to-local downsampling;

[0058] Specifically, such as Figure 2 As shown, the pre-built Hierarchical Edge-Aware Network (HEA) consists of multiple modules, including a local embedding module (LE), a global embedding module (GE), an attention mechanism module (TA), a global-to-local downsampling module (GLD), and an upsampling module (USP). The local embedding module (LE) is responsible for extracting local features from the input image. Through convolution operations, it captures image details such as edges and textures. The global embedding module (GE) is responsible for extracting global features from the input image. Unlike local embedding, this module focuses on the overall context of the image, ensuring that the model can understand the overall image. The attention mechanism module (TA) is used to highlight features in important image regions and ignore less important parts. By calculating attention weights for input features, the model can focus more on key areas, improving the effectiveness of feature representation. The global-to-local downsampling module (GLD) downsamples the features, converting high-resolution feature maps to low-resolution feature maps. This helps reduce computational overhead while preserving important global and local information. The upsampling module (USP) restores low-resolution feature maps to high-resolution by upsampling the features, thereby restoring the resolution and increasing the clarity of the point cloud. The upsampling operation can be combined with the deconvolution operation to ensure that the restored feature map has better detail preservation.

[0059] The hierarchical edge perception network HEA divides the environmental point cloud features and the distribution network point cloud features by focusing on the edge features of the point cloud, especially combining global features with small-scale local features, thereby improving the efficiency and accuracy of distribution network point cloud segmentation, thereby solving the problem of deep fusion between the distribution network point cloud and the environmental point cloud.

[0060] Optionally, the hierarchical edge perception network HEA adopts a multi-level network structure, including multiple attention mechanism modules TA, global to local downsampling GLD and upsampling USP grouped and connected in a certain order. At the same time, there are multiple skip connections (Skip Connections) in the hierarchical edge perception network HEA. Among them, the skip connection refers to the direct connection between the attention mechanism module TA and the global to local downsampling GLD. Through the skip connection, feature information can be transferred between different layers of the network, the multi-scale information of the feature map can be retained, and the overall performance of the model can be improved.

[0061] The above method, through the combination of local and global embedding modules, enables the Hierarchical Edge Awareness (HEA) network to simultaneously capture local details and global contextual information, improving feature extraction accuracy. Global-to-local downsampling (GLD) reduces the resolution of feature maps, reducing computational effort and improving the model's computational efficiency. The attention mechanism (TA) performs weighted processing on feature maps, highlighting important areas and enhancing the effectiveness of feature representation. Skip connections are configured to preserve feature information at different levels, helping to improve the model's overall performance.

[0062] The distribution network point cloud data is input into the hierarchical edge perception network, the local features and local neighborhood features of each point in the distribution network point cloud data are captured by the local embedding module, global features are recognized by the global embedding module, the local features, local neighborhood features and global features are fused by the attention mechanism module, the segmentation task features are focused on, multiple downsampling and upsampling are performed through the multiple jump connections contained in the multi-level network, the environmental features and the distribution network features are segmented, and the segmentation results of the distribution network are output.

[0063] Specifically, first, the distribution network point cloud data (including environmental point cloud data) is input into the input end (Input) of the hierarchical edge perception network HEA; then, the local embedding module LE and the global embedding module GE respectively obtain the local neighborhood features of multiple points, the local neighborhood features and the global features of the entire point cloud data; then, the local features, local neighborhood features and global features are input into the attention mechanism module TA, focusing on the key features of the segmentation task for feature fusion; then, multiple downsampling operations and upsampling operations are performed through the global to local downsampling GLD and upsampling USP at different levels in the multi-level network, and features are transferred between different levels based on multiple jump connections to realize the segmentation of environmental features and distribution network features; finally, the segmented distribution network point cloud data is output from the output end (HEA Output) of the hierarchical edge perception network HEA to obtain the segmentation result of the distribution network and environmental features.

[0064] Through the above steps and modules, the hierarchical edge perception network HEA can efficiently and accurately segment the point cloud data of the distribution network and the environment, solve the problem of deep fusion between the distribution network point cloud and the environment point cloud, and improve the efficiency and accuracy of the distribution network point cloud segmentation.

[0065] In some embodiments, capturing the local features and local neighborhood features of each point in the power distribution network point cloud data by the local embedding module includes:

[0066] The K neighboring points of each point in the point cloud data are determined by distance measurement; the feature vectors of the K neighboring points are copied K times to construct a feature matrix; the K copied feature matrices are spliced ​​according to the channel dimension to obtain a stacked feature matrix, which includes the local features of the original point and the neighborhood; the stacked feature matrix is ​​used as output to obtain the local features and local neighborhood features.

[0067] Optionally, the local features and local neighborhood features of each point are captured by the local embedding module LE. First, the K nearest neighboring points of each point in the point cloud data are determined by a distance measurement method (such as Euclidean distance) to form the neighborhood of the point.

[0068] Furthermore, the eigenvectors of the K nearest neighbors are replicated K times, that is, the eigenvectors of each of the K nearest neighbors are replicated, to construct a feature matrix. Exemplarily, the feature matrix is ​​a KxC matrix, where the K rows of the feature matrix correspond to the eigenvectors of the K nearest neighbors, and the C columns correspond to the number of feature dimensions, C, of ​​the eigenvectors of the K nearest neighbors.

[0069] Optionally, the K copied feature matrices are concatenated according to the channel dimension to obtain a stacked feature matrix. This stacked feature matrix includes the local features of the K original points and their K neighborhoods. Each row or column of the stacked feature matrix contains the features of the original point and its neighborhood points. The K layers of feature matrices in the stacked feature matrix correspond to the K nearest neighbor points, forming a feature representation that incorporates local information. The stacked feature matrix is ​​used as the output of the local embedding module LE to obtain the local features and local neighborhood features.

[0070] Through the above steps, the local embedding module LE can effectively capture the local features of each point and its neighborhood features, so that the local information of the point cloud data is fully utilized, the local representation ability of the point cloud data is enhanced, and a basis for subsequent global feature extraction and feature fusion is provided.

[0071] In some implementations, before concatenating the K copied feature matrices according to the channel dimension, the process further includes:

[0072] Configure a multi-scale K value; determine multi-scale K neighbor points based on the multi-scale K value; perform feature insertion based on the feature vectors of the multi-scale K neighbor points to form multiple new feature dimensions; perform visual evaluation on the multiple new feature dimensions, determine the relationship between feature distribution and target feature variables, and screen new feature dimensions; and add the new feature dimensions to the splicing features.

[0073] Optionally, before concatenating the K replicated feature matrices along the channel dimension, configure a multi-scale K value. This means selecting multiple different K values ​​to capture neighborhood features at different scales. Specifically, a larger K value captures a wider range of local features and local neighborhood features, leading to a greater amount of data processing.

[0074] Optionally, based on the multi-scale K values, K multi-scale neighboring points are determined. Specifically, for each point, the set of neighboring points at different K values ​​is determined, where the feature vectors corresponding to the K multi-scale neighboring points have different feature dimensions. Furthermore, feature interpolation is performed based on the feature vectors of the K multi-scale neighboring points to form multiple new feature dimensions. These new feature dimensions reflect the differences in neighborhood features at different scales. For example, smaller scales may contain feature dimensions that are missing at larger scales, useful for identifying subtle textures or color variations in fabrics.

[0075] Optionally, use visualization tools such as scatter plots, heat maps, and boxplots to visually evaluate the feature distribution and its relationship with the target variable. Based on the visualization results, determine the relationship between the feature distribution and the target feature variable and select new feature dimensions. Specifically, select feature dimensions that have a significant correlation with the target variable (such as classification labels or regression values) as new feature dimensions.

[0076] Furthermore, the filtered new feature dimensions are added to the concatenated features to obtain the final stacked feature matrix. This ensures that the stacked feature matrix can represent richer and more diverse local features of the original points and their neighborhoods, making more comprehensive use of the local information of the point cloud data.

[0077] In some embodiments, fusing local features, local neighborhood features, and global features through an attention mechanism module includes:

[0078] According to the local features and global features, a weight matrix is ​​used to map them into a query vector, a key vector, and a value vector; the similarity score between the query vector and all key vectors is calculated to obtain an attention score; the similarity score between each query vector and all key vectors is normalized by a normalization function to establish a normalized weight distribution relationship between each query vector and the key vector; the normalized weight is multiplied by the corresponding value vector to obtain the attention mechanism output.

[0079] Specifically, the fusion of local features, local neighborhood features, and global features begins by performing linear transformations by multiplying the local features with a weight matrix and the global features with the weight matrix. This transforms the local and global features into their corresponding query vectors Q, key vectors K, and value vectors V, respectively. This unifies the feature dimensions and format to facilitate subsequent calculations. The query vector represents the target feature to be considered; the key vector represents the content of all features; and the value vector represents the content of the features to be extracted.

[0080] Furthermore, the similarity score between the query vector and all key vectors is calculated through a dot product operation and stored as the attention score. Then, the similarity score between each query vector and all key vectors is normalized using a normalization function, converting the similarity score into a normalized weight distribution so that the sum of all weights is 1.

[0081] Optionally, a normalized weight distribution is constructed based on a normalized probability distribution, reflecting the attention of the HEA to each position. Furthermore, the normalized weights in the normalized weight distribution are multiplied by the corresponding value vectors and summed to obtain the output of the attention mechanism. The attention mechanism output represents the fused feature information.

[0082] Through the attention mechanism module, TA focuses on important features and ignores irrelevant information. Local features, local neighborhood features and global features are integrated, making the feature representation more comprehensive and accurate, and improving the effect of feature extraction and classification.

[0083] In some implementations, obtaining an attention mechanism output by multiplying the normalized weights by the corresponding value vectors includes:

[0084] By formula: Calculate and obtain the output of the attention mechanism, where Q is the query vector, K is the key vector, V is the value vector, is the scaling factor.

[0085] Optionally, Q×K represents the similarity score, It is used to perform dimension division on the similarity score to prevent the dot product value from being too large, which would lead to gradient vanishing when calculating the softmax.

[0086] Furthermore, after processing Applying the softmax function, we can get the attention weight Finally, the attention weights are multiplied by the value vector V to obtain the output of the attention mechanism.

[0087] Furthermore, after outputting the segmentation result of the power distribution network, the method further includes:

[0088] The segmentation result of the distribution network is used to identify distribution lines and extract the distribution line location; a line smoothness test is performed based on the distribution line location to obtain a smoothness test result; an abnormal node is identified based on the smoothness test result; and when the abnormal node does not exist, the segmentation result of the distribution network is determined.

[0089] Optionally, after obtaining the segmentation results of the distribution network, distribution line identification is performed on the segmentation results. The segmentation results are point cloud data related to the distribution network, specifically including point cloud data of the distribution network's towers, distribution lines, transformers, and other ancillary equipment. Exemplarily, based on feature extraction algorithms (such as edge detection algorithms and Hough transforms) and target detection techniques, the distribution line point cloud from the segmentation results of the distribution network is extracted to generate distribution line location information.

[0090] Optionally, according to the distribution location of the distribution lines, the shape or layout of the distribution lines is analyzed based on computational geometry methods to detect the smoothness of the lines.

[0091] Exemplarily, a smoothness calculation method (such as polynomial fitting, B-spline curve, etc.) is used to detect the smoothness of the distribution line, and based on the smoothness detection results, abnormal nodes in the distribution line are identified, where the abnormal nodes may be points that deviate significantly from the smooth curve due to noise or data errors.

[0092] Optionally, when no abnormal nodes exist, the segmentation result of the distribution network can be considered valid. Correspondingly, if abnormal nodes exist, the segmentation result is reprocessed or manually checked and corrected.

[0093] In some embodiments, after identifying abnormal nodes, the method further includes:

[0094] When the abnormal node exists, the abnormal distribution line and the adjacent distribution lines are obtained; line abnormal node alignment compensation is performed according to the abnormal distribution line to determine the compensation characteristics; based on the compensation characteristics, the abnormal environmental characteristics are deducted; environmental characteristics are extracted according to the adjacent distribution line, and the environmental characteristics are used to perform correlation analysis with the abnormal environmental characteristics to determine the correlation identification characteristics; the abnormal environmental characteristics are re-segmented using the correlation identification characteristics, and the distribution line characteristics are restored and re-segmented based on the compensation characteristics.

[0095] Specifically, first, the distribution line containing the abnormal node is identified and extracted as the abnormal distribution line, and the distribution line directly adjacent to the abnormal distribution line is identified as the adjacent distribution line. For example, the spatial neighborhood search algorithm is used to find the distribution line adjacent to the abnormal node.

[0096] Furthermore, alignment compensation is performed on abnormal nodes in the abnormal distribution line to eliminate the deviation caused by the abnormal nodes. At the same time, based on the alignment compensation results, compensation features are determined, where the compensation features include position, direction, and morphology features.

[0097] Optionally, an interpolation algorithm is used to align and compensate abnormal nodes, including linear interpolation and spline interpolation.

[0098] Optionally, use a smoothing algorithm (such as Gaussian smoothing) to process abnormal nodes and achieve alignment compensation

[0099] Furthermore, based on the compensation features, a compensation vector is generated, starting from the abnormal node and combining the compensation features. The neighborhood features of the compensation vector are then extracted as abnormal environmental features. The compensation vector represents the vector from the abnormal node to the compensation location, containing both direction and magnitude information. This method accurately extracts environmental features surrounding the abnormal node, providing data support for subsequent correlation analysis.

[0100] Optionally, correlation analysis is performed using abnormal environmental features and environmental features of adjacent distribution lines, including calculating the correlation coefficient between the abnormal environmental features and the environmental features of adjacent distribution lines, determining their correlation, and screening out correlation identification features.

[0101] Optionally, the abnormal environmental features are re-segmented based on the associated identification features, dividing them into environmental point cloud data and power supply data related to the distribution network. Then, based on the compensation features, the distribution line features are restored to the abnormal node, and the neighborhood space of the abnormal node is re-segmented to correct the abnormal distribution line containing the abnormal node.

[0102] Through the above process, abnormal distribution lines can be effectively handled and the reliability and accuracy of distribution network point cloud segmentation can be improved.

[0103] In some implementations, environmental features are extracted based on the adjacent distribution lines, and correlation analysis is performed using the environmental features and abnormal environmental features to determine associated identification features, including: setting a feature initial point based on the environmental features extracted from the adjacent distribution lines, the feature initial point being a cluster of pixel points whose distance from the abnormal node meets the distance requirement; determining the neighborhood pixels of the feature initial point; fitting pixel growth information based on the feature initial point, the pixel growth information including grayscale value, color, texture, and gradient; taking the feature initial point as the starting point, performing growth coverage judgment on the neighborhood pixels based on the pixel growth information, and using the pixel area that meets the pixel growth information as a new feature initial point; fitting new pixel growth information based on the new feature initial point, iteratively using the pixel growth information to perform growth coverage judgment on the neighborhood pixels until the abnormal environmental features are reached, and determining the associated identification features based on the current pixel growth information.

[0104] Specifically, first, environmental features are extracted from adjacent distribution lines. These environmental features include information such as the grayscale value, color, texture, and gradient of the pixels, reflecting the environmental characteristics during normal segmentation. Then, based on the location of the abnormal node, pixels that meet certain distance requirements from the abnormal node are selected as feature initial points. The feature initial point is the starting point for growth and analysis, and it needs to maintain a certain distance from the abnormal node to distinguish normal and abnormal environmental features. Next, the pixels surrounding the feature initial point are identified as neighborhood pixels. The neighborhood pixels contain pixel information related to the feature initial point. Then, based on the feature initial point, pixel growth information, including grayscale value, color, texture, and gradient, is fitted, reflecting the growth trend and characteristics of the pixels, which help to identify and distinguish normal and abnormal environmental features.

[0105] Furthermore, starting from the feature initial point, the pixel growth information is used to determine the growth coverage of neighboring pixels. This determines which pixel regions meet the requirements of the pixel growth information and can be used as new feature initial points. In other words, pixel regions that meet the pixel growth information are regions with continuous pixel types.

[0106] Furthermore, based on the new feature initial point, new pixel growth information is fitted and iterative growth coverage judgment is performed. The iterative process continuously expands the feature initial point and grows to cover new pixel areas until the abnormal environmental features are covered. Based on the current pixel growth information, the associated recognition feature is determined.

[0107] In summary, the distribution network point cloud segmentation method based on environment fusion provided by the present invention has the following technical effects:

[0108] By collecting distribution network point cloud data, wherein the point cloud data includes environmental point cloud data, and the environmental point cloud data is integrated with the distribution network point cloud data; obtaining a pre-built hierarchical edge perception network, the hierarchical edge perception network includes a local embedding module, a global embedding module, an attention mechanism module, global to local downsampling, and upsampling, wherein the attention mechanism module, global to local downsampling, and upsampling construct a multi-level network, and the multi-level network contains multiple jump connections, and the jump connection is the jump between the attention mechanism module and the global to local downsampling; inputting the distribution network point cloud data into the hierarchical edge perception network, using the local embedding module to capture the local features and local neighborhood features of each point in the distribution network point cloud data, using the global embedding module to perform global feature recognition, using the attention mechanism module to fuse the local features, local neighborhood features, and global features, focusing on the segmentation task features, and using the multi-level network to contain multiple jump connections to perform multiple downsampling and upsampling, segmenting the environmental features and the distribution network features, and outputting the distribution network segmentation results. Thus, the technical effect of improving the efficiency and accuracy of distribution network point cloud segmentation is achieved.

[0109] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the embodiments mentioned above. It should be understood that those skilled in the art can still modify the technical solutions described in the above embodiments or replace some of the technical features therein with equivalents; and such modifications or replacements do not deviate from the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention and are all included in the scope of protection of the present invention.

Claims

1. A point cloud segmentation method for distribution network based on environment fusion, characterized by: The method comprises: Collecting distribution network point cloud data, wherein the point cloud data includes environmental point cloud data, and the environmental point cloud data is integrated with the distribution network point cloud data; Obtain a pre-built hierarchical edge-aware network, wherein the hierarchical edge-aware network includes a local embedding module, a global embedding module, an attention mechanism module, global-to-local downsampling, and upsampling, wherein the attention mechanism module, global-to-local downsampling, and upsampling construct a multi-level network, and the multi-level network contains multiple jump connections, wherein the jump connection is a jump between the attention mechanism module and the global-to-local downsampling; Input the distribution network point cloud data into the hierarchical edge perception network, capture the local features and local neighborhood features of each point in the distribution network point cloud data through the local embedding module, perform global feature recognition through the global embedding module, fuse the local features, local neighborhood features and global features through the attention mechanism module, focus on the segmentation task features, perform multiple downsampling and upsampling through the multiple jump connections included in the multi-level network, segment the environmental features and the distribution network features, and output the segmentation result of the distribution network; After outputting the segmentation result of the power distribution network, the method further includes: Performing distribution line identification on the segmentation result of the distribution network and extracting the distribution location of the distribution lines; Performing line smoothness detection according to the distribution line positioning to obtain a smoothness detection result; Identifying abnormal nodes based on the smoothness detection result; When the abnormal node does not exist, determining a segmentation result of the power distribution network; After identifying abnormal nodes, the method further includes: When the abnormal node exists, obtaining the abnormal distribution line and adjacent distribution lines; Perform line abnormal node alignment compensation according to the abnormal distribution line and determine compensation characteristics; Deducting abnormal environmental features based on the compensation features; Extracting environmental features based on the adjacent power distribution lines, performing correlation analysis using the environmental features and abnormal environmental features, and determining correlation identification features; The abnormal environment feature is re-segmented using the associated identification feature, and the distribution line feature is restored and re-segmented based on the compensation feature.

2. The method for point cloud segmentation of a power distribution network based on environment fusion according to claim 1, characterized in that: The local embedding module captures the local features and local neighborhood features of each point in the distribution network point cloud data, including: Determine the K neighbor points of each point in the point cloud data through distance measurement; The feature vectors of the K neighbor points are copied K times to construct a feature matrix; The K copied feature matrices are spliced ​​according to the channel dimension to obtain a stacked feature matrix, which includes the local features of the original point and the neighborhood; The stacked feature matrix is ​​used as output to obtain the local features and local neighborhood features.

3. The method for point cloud segmentation of a power distribution network based on environment fusion according to claim 2, characterized in that: Before splicing the K copied feature matrices according to the channel dimension, the method further includes: Configure multi-scale K value; Determining multi-scale K neighbor points based on the multi-scale K value; Perform feature insertion based on the feature vectors of the multi-scale K neighboring points to form multiple new feature dimensions; Performing visual evaluation on the multiple new feature dimensions, determining the relationship between feature distribution and target feature variables, and screening new feature dimensions; The new feature dimension is added to the concatenated feature.

4. The method for point cloud segmentation of a power distribution network based on environment fusion according to claim 1, characterized in that: The attention mechanism module is used to fuse local features, local neighborhood features, and global features, including: According to the local features and global features, the weight matrix is ​​used to map the query vector, key vector, and value vector; Calculate the similarity scores between the query vector and all key vectors to obtain an attention score; The similarity scores between each query vector and all key vectors are normalized by a normalization function, and a normalized weight distribution relationship between each query vector and the key vector is established; The normalized weight is multiplied by the corresponding value vector to obtain the attention mechanism output.

5. The method for point cloud segmentation of a power distribution network based on environment fusion according to claim 4, characterized in that: The normalized weight is multiplied by the corresponding value vector to obtain the attention mechanism output, including: By formula: Calculate and obtain the output of the attention mechanism, where Q is the query vector, K is the key vector, V is the value vector, is the scaling factor.

6. The method for point cloud segmentation of a power distribution network based on environment fusion according to claim 1, characterized in that: Extracting environmental features based on the adjacent power distribution lines, performing correlation analysis using the environmental features and abnormal environmental features, and determining correlation identification features, including: Setting a feature initial point based on the environmental features extracted from the adjacent power distribution lines, wherein the feature initial point is a cluster of pixel points whose distance from the abnormal node meets the distance requirement; Determine the neighborhood pixels of the feature initial point; Fitting pixel growth information based on the feature initial point, wherein the pixel growth information includes grayscale value, color, texture, and gradient; Taking the feature initial point as the starting point, performing growth coverage judgment on neighboring pixels based on the pixel growth information, and taking the pixel area that meets the pixel growth information as the new feature initial point; New pixel growth information is fitted based on the new feature initial point, and the pixel growth information is iteratively used to determine the growth coverage of neighboring pixels until the abnormal environment feature is found, and the associated identification feature is determined based on the current pixel growth information.

Images