Method and system for separating branches and leaves from vegetation point clouds around transmission lines

Abstract

The present invention provides a branch and leaf separation method and system for vegetation point clouds around transmission lines, which relates to the field of data processing technology. The implementation scheme is as follows: based on the curvature of each point cloud in the vegetation point cloud around the target transmission line, the surrounding vegetation point cloud is divided into a first branch point cloud set and a first leaf point cloud set; the first branch point cloud set is segmented into a connected domain to obtain multiple connected components; based on the linear feature significance of the geometric distribution characteristics of each connected component, the first branch point cloud set is divided into a second branch point cloud set and a branch and leaf mixed point cloud set; with each branch point in the second branch point cloud set as a seed point, the branch and leaf mixed point cloud set is divided into a regional growth point cloud set and a non-regional growth point cloud set by regional growth; based on the union of the second branch point cloud set and the regional growth point cloud set, the target branch point cloud set is determined; based on the union of the first leaf point cloud set and the non-regional growth point cloud set, the target leaf point cloud set is determined.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for separating branches and leaves from a vegetation point cloud around a power transmission line. Background Art

[0002] With the continued development of modern power systems, the safety and reliability of transmission lines have become a key technical research area. Light Detection and Ranging (LiDAR) technology, with its high-precision three-dimensional data acquisition capabilities, has become a core tool for monitoring transmission corridors. By acquiring point cloud data around transmission lines, researchers can perform detailed modeling of vegetation and infrastructure within complex geographic environments. This detailed modeling is crucial for ensuring the safe operation of transmission lines, as it helps identify potential risk factors, such as excessive vegetation growth or damaged infrastructure.

[0003] However, processing vegetation point cloud data presents significant challenges in separating branches and leaves from trees. The complexity and unstructured nature of point cloud data surrounding transmission lines makes it difficult for traditional analysis methods to accurately identify and extract branch and leaf structures. The complex spatial distribution and significant noise of the data, coupled with the diverse and dynamic nature of tree morphology, further complicate accurate separation.

[0004] Furthermore, noise and occlusion in point cloud data significantly complicate the separation of tree branches and leaves. The complex three-dimensional structure of trees, the intertwining of branches and leaves, and the morphological differences between species pose significant challenges to existing separation algorithms. Traditional methods are prone to misclassification when dealing with small branches and leaves, compromising the accuracy of subsequent analyses of ecosystems near power transmission lines. This not only leads to biased data analysis results but can also negatively impact subsequent decision support systems. Summary of the Invention

[0005] The present invention provides a method and system for separating branches and leaves from vegetation point clouds around transmission lines, which can solve at least one of the above technical problems.

[0006] According to one aspect of the present invention, a method for separating branches and leaves from a vegetation point cloud around a transmission line is provided, comprising:

[0007] Based on the curvature of each point cloud in the vegetation point cloud surrounding the target transmission line, the surrounding vegetation point cloud is divided into a first branch point cloud set and a first leaf point cloud set;

[0008] Performing connected domain segmentation on the first branch point cloud set to obtain multiple connected components;

[0009] Based on the linear feature significance of the geometric distribution features of each of the connected components, dividing the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set;

[0010] Taking each branch point in the second branch point cloud as a seed point, dividing the branch-leaf mixed point cloud into a regional growth point cloud and a non-regional growth point cloud by region growing;

[0011] Determining a target branch point cloud set of the target transmission line based on a union of the second branch point cloud set and the regional growth point cloud set;

[0012] A target blade point cloud set of the target transmission line is determined based on a union of the first blade point cloud set and the non-region growth point cloud set.

[0013] According to another aspect of the present invention, a device for separating branches and leaves from a point cloud of vegetation around a transmission line is provided, the method comprising:

[0014] a first point cloud classification module, configured to divide the surrounding vegetation point cloud of the target transmission line into a first branch point cloud set and a first leaf point cloud set based on the curvature of each point cloud in the surrounding vegetation point cloud;

[0015] A connected domain segmentation module, configured to perform connected domain segmentation on the first branch point cloud set to obtain a plurality of connected components;

[0016] A second point cloud classification module is configured to divide the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set based on the linear feature significance of the geometric distribution features of each of the connected components;

[0017] a third point cloud classification module, configured to use each branch point in the second branch point cloud set as a seed point and divide the branch-leaf mixed point cloud set into a regional growth point cloud set and a non-regional growth point cloud set by region growing;

[0018] A first union module is configured to determine a target branch point cloud set of the target transmission line based on a union of the second branch point cloud set and the regional growth point cloud set;

[0019] The second union module is configured to determine a target blade point cloud set of the target transmission line based on a union of the first blade point cloud set and the non-regional growth point cloud set.

[0020] By adopting the technical solution of the present invention, the surrounding vegetation point cloud of the target transmission line is divided into a first branch point cloud set and a first leaf point cloud set based on the curvature of each point cloud in the surrounding vegetation point cloud. In this way, the surrounding vegetation point cloud of the transmission line is classified into branches and leaves point by point. Then, the first branch point cloud set is segmented into a connected domain to obtain multiple connected components; based on the linear feature significance of the geometric distribution characteristics of each connected component, the first branch point cloud set is divided into a second branch point cloud set and a branch and leaf mixed point cloud set. In this way, the branch point cloud is further classified into branches and leaves section by section. Moreover, with each branch point in the second branch point cloud set as a seed point, the branch and leaf mixed point cloud set is divided into a regional growth point cloud set and a non-regional growth point cloud set through regional growth. In this way, the branch and leaf mixed points are classified point by point again. Finally, the target branch point cloud for the target transmission line is determined based on the union of the second branch point cloud and the regional growth point cloud. The target leaf point cloud for the target transmission line is determined based on the union of the first leaf point cloud and the non-regional growth point cloud. Therefore, by combining point-by-point and segment-by-segment classification strategies, we can accurately separate branches and leaves from complex vegetation structures, reducing the occurrence of misclassification.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are provided for a better understanding of the present invention and do not constitute a limitation of the present invention.

[0023] Figure 1 This is a flow chart of a method for separating branches and leaves from a vegetation point cloud around a transmission line according to an embodiment of the present invention;

[0024] Figure 2 is a flow chart of a method for separating branches and leaves from a vegetation point cloud around a transmission line according to another embodiment of the present invention;

[0025] Figure 3A is a schematic diagram of an original point cloud according to an embodiment of the present invention;

[0026] Figure 3B is a schematic diagram of a first branch point cloud according to an embodiment of the present invention;

[0027] Figure 3C is a schematic diagram of a first blade point cloud set according to an embodiment of the present invention;

[0028] Figure 4 This is a structural block diagram of a branch and leaf separation device for vegetation point clouds around power transmission lines according to an embodiment of the present invention;

[0029] Figure 5is a block diagram of an electronic device for implementing the method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following description of exemplary embodiments of the present invention is made in conjunction with the accompanying drawings, and various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0031] Figure 1 This is a flow chart of a method for separating branches and leaves from a vegetation point cloud around a transmission line according to an embodiment of the present invention.

[0032] like Figure 1 As shown, the method for separating branches and leaves from vegetation point clouds around transmission lines may include:

[0033] S110, dividing the surrounding vegetation point cloud of the target transmission line into a first branch point cloud set and a first leaf point cloud set based on the curvature of each point cloud in the surrounding vegetation point cloud;

[0034] S120, performing connected domain segmentation on the first branch point cloud to obtain multiple connected components;

[0035] S130, dividing the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set based on the linear feature significance of the geometric distribution features of each connected component;

[0036] S140, using each branch point in the second branch point cloud as a seed point, dividing the branch-leaf mixed point cloud into regional growth point cloud sets and non-regional growth point cloud sets by region growing;

[0037] S150, determining a target branch point cloud set for the target transmission line based on the union of the second branch point cloud set and the regional growth point cloud set;

[0038] S160 : Determine a target blade point cloud set of the target transmission line based on the union of the first blade point cloud set and the non-region growth point cloud set.

[0039] For example, the vegetation point cloud surrounding the target transmission line may be a point cloud within a specified neighborhood.

[0040] For example, the surface normal change rate can be used to calculate the curvature NCR of each point cloud. This describes the change in the point cloud along the surface normal and can reflect the degree to which the point cloud deviates from the tangent plane. Using the curvature of each point cloud, each point cloud can be classified as a branch point cloud or a leaf point cloud. If it is classified as a branch point cloud, it is added to the first branch point cloud set. If it is classified as a leaf point cloud, it is added to the first leaf point cloud set.

[0041] For example, a connected domain refers to a set of points connected by a neighborhood. For a three-dimensional point cloud, if two points are sufficiently close to each other in space, for example, within a certain threshold range, they are determined to be adjacent. These adjacent points can then be connected to form a connected region, which is then used as a connected component. In this way, the first branch point cloud set can be divided into multiple connected components. Each connected component includes multiple adjacent point clouds.

[0042] Exemplarily, the geometric distribution characteristics of the connected components may include linear characteristics, planar characteristics, and scattering characteristics. For different connected components, their geometric distribution characteristics are different. For example, for the connected components of the branch category, the significance of their linear characteristics is higher, while the significance of their planar characteristics and scattering characteristics is lower. For the connected components of the leaf category or the mixed category of branches and leaves, the significance of their linear characteristics is lower, while the significance of their planar characteristics and scattering characteristics is higher.

[0043] Therefore, we can determine whether a connected component belongs to the branch category by the significance of the linear features in its geometric distribution. If it belongs to the branch category, the connected component is added to the second branch point cloud set. If it does not belong to the branch category, the connected component is added to the branch-leaf mixed point cloud set. Subsequently, the point clouds in the branch-leaf mixed point cloud set are subjected to branch-leaf separation.

[0044] For example, using each branch point in the second branch point cloud set as a seed point, regional growing is used to identify each point cloud in the branch-leaf mixed point cloud set. Thus, the regional growing point identified as a seed point is added to the regional growing point cloud set and deleted from the branch-leaf mixed point cloud set. After each seed point has identified each point cloud in the branch-leaf mixed point cloud set, the remaining point clouds in the branch-leaf mixed point cloud set are added to the non-regional growing point cloud set.

[0045] For example, the normal vector angle between the seed point and any point cloud in the branch-leaf mixed point cloud set is calculated to determine whether the point cloud is a regional growth point of the seed point.

[0046] For example, by calculating the distance between the seed point and any point cloud in the branch-leaf mixed point cloud set, it can be determined whether the point cloud is a regional growth point of the seed point.

[0047] Exemplarily, the union of the second branch point cloud set and the regional growth point cloud set is used as the target branch point cloud set of the target transmission line.

[0048] Exemplarily, the union of the first blade point cloud set and the non-region growth point cloud set is used as the target blade point cloud set of the target transmission line.

[0049] For example, before step S140, the first leaf point cloud set may be merged into the branch-leaf mixed point cloud set, and the first leaf point cloud set may be changed to an empty set, and then step S140 may be performed. Furthermore, in step S160, the non-region growth point cloud set is actually used as the target leaf point cloud set.

[0050] For example, based on the target branch point cloud set and the target leaf point cloud set, a target tree can be accurately constructed. Thus, by analyzing the target tree, it can be evaluated whether the target tree will cause power transmission hazards to the target transmission line.

[0051] According to the above embodiment, based on the curvature of each point cloud in the vegetation point cloud surrounding the target transmission line, the surrounding vegetation point cloud is divided into a first branch point cloud set and a first leaf point cloud set. In this way, the vegetation point cloud surrounding the transmission line is classified into branches and leaves point by point. Then, the first branch point cloud set is segmented into a connected domain to obtain multiple connected components; based on the linear feature significance of the geometric distribution characteristics of each connected component, the first branch point cloud set is divided into a second branch point cloud set and a branch and leaf mixed point cloud set. In this way, the branch point cloud is further classified into branches and leaves section by section. Moreover, with each branch point in the second branch point cloud set as a seed point, the branch and leaf mixed point cloud set is divided into a regional growth point cloud set and a non-regional growth point cloud set by regional growth. In this way, the branch and leaf mixed points are classified point by point again. Finally, the target branch point cloud for the target transmission line is determined based on the union of the second branch point cloud and the regional growth point cloud. The target leaf point cloud for the target transmission line is determined based on the union of the first leaf point cloud and the non-regional growth point cloud. Therefore, by combining point-by-point and segment-by-segment classification strategies, we can accurately separate branches and leaves from complex vegetation structures, reducing the occurrence of misclassification.

[0052] In one embodiment, the above method may also include: determining the covariance matrix of each point cloud based on the position information of the geometric center of the target transmission line in the surrounding vegetation point cloud within a specified neighborhood, and the position information of each point cloud in the surrounding vegetation point cloud; performing singular value decomposition on the covariance matrix of each point cloud to obtain the singular eigenvalues ​​of each point cloud; and determining the curvature of each point cloud based on the singular eigenvalues ​​of each point cloud.

[0053] For example, for point p i = (x i ,yi ,z i ), its covariance matrix C is defined as follows:

[0054] ;

[0055] in, is the geometric center of neighborhood p, and n is the number of point clouds in neighborhood p.

[0056] For example, singular value decomposition of the covariance matrix of the point cloud can be performed to obtain singular eigenvalues ​​of the point cloud, which may include three eigenvalues, including a first eigenvalue of rank 0, a second eigenvalue of rank 1, and a third eigenvalue of rank 2.

[0057] Exemplarily, the ratio between the first eigenvalue of the point cloud and the sum of the three eigenvalues ​​is used as the curvature of the point cloud.

[0058] For example, the point cloud p is calculated using the following formula: i Curvature:

[0059] ;

[0060] in, represents the first eigenvalue of rank 0, represents the second eigenvalue of rank 1, represents the third eigenvalue of rank 2.

[0061] In this example, the covariance matrix of the point cloud is determined based on its position relative to the geometric center of the point cloud in the neighborhood, and then the singular value decomposition of the covariance matrix is ​​performed to obtain the singular eigenvalues ​​of the point cloud. Using the singular eigenvalues ​​of the point cloud, the curvature of the point cloud can be accurately calculated.

[0062] In one embodiment, the above-mentioned vegetation point cloud around the target transmission line is divided into a first branch point cloud set and a first leaf point cloud set based on the local curvature NCR of each point cloud in the vegetation point cloud around the target transmission line, including one of the following: when the local curvature NCR of the point cloud is greater than a preset local curvature threshold T NCR In the case of , the point cloud is added to the first leaf point cloud set; when the local curvature NCR of the point cloud is less than the preset local curvature threshold T NCR , add the point cloud to the first branch point cloud set.

[0063] For example, the curvature of any point cloud ranges from 0 to 1 / 3, reflecting the degree of change along the surface normal. Higher values ​​indicate more pronounced curvature changes, while lower values ​​indicate less pronounced curvature changes.

[0064] It can be understood that the leaf point cloud has a higher curvature. The branch point cloud has a lower curvature. Although the spatial distribution of leaf points is complex, resulting in the curvature NCR of some leaf point clouds being similar to that of the branch point cloud, by setting the curvature threshold T NCR , which can effectively separate leaf points with large local curvature changes from the point cloud.

[0065] For example, Figure 2 As shown, the curvature of the original point cloud data is estimated, and the curvature is used to divide it into a first branch point cloud set and a first leaf point cloud set. The connected domain segmentation and branch and leaf separation of the connected components of the above-mentioned steps S120 and S130 are performed on the first branch point cloud set to obtain a second branch point cloud set and a branch and leaf mixed point cloud set. Then, the first leaf point cloud set is merged into the branch and leaf mixed point cloud set to obtain a new branch and leaf mixed point cloud set. Finally, the above-mentioned step S140 is adopted, and each branch point in the second branch point cloud set is used as a seed point, and the new branch and leaf mixed point cloud set is divided into a regional growth point cloud set and a non-regional growth point cloud set by regional growth. Finally, in step S160, the non-regional growth point cloud set is used as the target leaf point cloud set.

[0066] For example, when the curvature threshold is 0.1, the above curvature method is used to Figure 3A The original point cloud shown in the figure is separated into branches and leaves, and the first branch point cloud set can be obtained as follows Figure 3B As shown, the first leaf point cloud is as follows Figure 3C shown.

[0067] Based on the local curvature of each point cloud in the vegetation point cloud around the target transmission line, the surrounding vegetation point cloud is divided into the first branch point cloud set and the first leaf point cloud set.

[0068] In one embodiment, the above method also includes: determining the linear characteristics, planar characteristics and scattering characteristics of the connected components based on the geometric distribution characteristics of the connected components; processing the linear characteristics, planar characteristics and scattering characteristics based on the nonlinear adjustment coefficient, planarity weight and scattering weight to obtain the linear feature significance value; processing the planar characteristics and scattering characteristics based on the nonlinear adjustment coefficient, planarity weight and scattering weight to obtain the nonlinear feature significance value; determining the linear feature significance degree of the connected components based on the ratio between the linear feature significance value and the nonlinear feature significance value.

[0069] For example, the singular eigenvalues ​​of the connected components can be calculated and the linear features, planar features and scattering features of the connected components can be calculated using the singular eigenvalues ​​of the connected components.

[0070] Exemplarily, singular value decomposition is performed on the covariance matrix of the connected component to obtain three eigenvalues ​​of the connected component with ranks of 0, 1 and 2, namely the first eigenvalue, the second eigenvalue and the third eigenvalue.

[0071] Exemplarily, the ratio of the difference between the first eigenvalue and the second eigenvalue of the connected component to the first eigenvalue is used as the linear feature of the connected component.

[0072] Exemplarily, the ratio of the difference between the second eigenvalue and the third eigenvalue of the connected component to the first eigenvalue is used as the plane feature of the connected component.

[0073] Exemplarily, the ratio between the third eigenvalue and the first eigenvalue of the connected component is used as the scattering feature of the connected component.

[0074] For example, the linear feature significance of the connected component is calculated using the following formula: :

[0075] ;

[0076] in, represents linear features, represents the nonlinear adjustment coefficient, Represents a plane feature, represents the scattering characteristics, represents the planarity weight, represents the scattering weight.

[0077] For example, and Represent the planarity and scattering weights respectively, and satisfy the constraints + = 1. By adjusting the weight, the importance of planarity and scattering can be more accurately reflected. In order to adapt to the situation where the linear characteristics are not significant, a nonlinear adjustment coefficient is introduced. ,dynamically adjusts the contribution of the nonlinear part, making the above formula more flexible and accurate in distinguishing branches from leaves.,LWSR values ​​range from 0 to positive infinity, where the larger the value, the more likely the segment is dominated by linear features.

[0078] For example, to precisely determine the optimal parameter combination, a grid search method can be used to traverse multiple parameter combinations to find the optimal parameters that maximize the classification accuracy based on the LWSR value, thereby effectively distinguishing branches from leaves. The values ​​of WP and λ range from 0 to 1, WS is 1 - WP, and the LWSR threshold ranges from 0.5 to 10. Through a systematic search within the preset parameter space, the optimal combination of (WP = 0.2), (WS = 0.8), (λ = 0.8), and an LWSR threshold of 2 was ultimately determined. This combination achieves the highest classification accuracy based on the LWSR metric and further reduces the omission error of branch point detection in subsequent processing.

[0079] According to the above embodiment, the linear feature significance of the connected components can be accurately calculated.

[0080] In one embodiment, the division of the first branch point cloud set into the second branch point cloud set and the branch-leaf mixed point cloud set based on the linear feature significance of the geometric distribution features of each connected component includes one of the following:

[0081] When the linear feature significance of the connected component is greater than a preset linear feature significance threshold, the connected component is added to the second branch point cloud set;

[0082] When the linear feature significance of the connected component is less than a preset linear feature significance threshold, the connected component is added to the branch-leaf mixed point cloud set.

[0083] In this example, the first branch point cloud is first divided into multiple connected components, and the multiple connected components are identified as branches or leaves respectively. In this way, during calculation, the geometric distribution characteristics of each connected component rather than each point are calculated, which greatly reduces the amount of calculation and improves the classification efficiency.

[0084] In one embodiment, the above method uses each branch point in the second branch point cloud as a seed point and divides the branch-leaf mixed point cloud into a regional growth point cloud and a non-regional growth point cloud by regional growing, including:

[0085] Extract seed points from the second branch point cloud set, delete the seed points from the second branch point cloud set, and perform the following first operation on the seed points, return to continue extracting seed points, and stop extracting seed points when the second branch point cloud set is empty, and use the branch-leaf mixed point cloud set as the non-region growth point cloud set;

[0086] The first operation includes:

[0087] Determine whether each branch-and-leaf mixed point cloud is a regional growth point of the seed point based on the normal angle between the normal vector of the seed point and the normal vectors of each branch-and-leaf mixed point cloud in the branch-and-leaf mixed point cloud set;

[0088] For any branch-and-leaf mixed point cloud in the branch-and-leaf mixed point cloud set, if the branch-and-leaf mixed point cloud is a regional growth point of a seed point, the branch-and-leaf mixed point cloud is added to the regional growth point cloud set, and the branch-and-leaf mixed point cloud is deleted from the branch-and-leaf mixed point cloud set.

[0089] Exemplarily, based on the normal angle between the normal vector of the seed point and the normal vectors of each branch and leaf mixed point cloud in the branch and leaf mixed point cloud set, as well as the distance between the seed point and the branch and leaf mixed point cloud, it is determined whether each branch and leaf mixed point cloud is the regional growth point of the seed point.

[0090] Exemplarily, if the distance between the seed point and the branch-leaf mixed point cloud is relatively close, and the normal angle between the normal vector of the seed point and the normal vector of the branch-leaf mixed point cloud is less than the smoothing threshold, then the branch-leaf mixed point cloud is the regional growth point of the seed point.

[0091] Based on the above example, the branch points in the second branch point cloud set are used as seed points, and neighboring points (irregular branch points, i.e., branch-leaf mixed point clouds) are extracted from the branch-leaf mixed point cloud set through region growing. This process defines relevant constraint rules to determine whether neighboring points (irregular branch points) can be merged with the current region (seed point). This example determines whether to classify neighboring points (irregular branch points) into the same region by checking the angle between the normal vectors of the seed point and its neighboring points (irregular branch points). If the normal vector angle is less than the smoothing threshold θ, the neighboring point is considered a branch point and continues to grow using it as a seed point. This stage divides neighboring points with smooth surfaces and minimal normal changes into the same region.

[0092] In one embodiment, the angle between the normal vector of the seed point and the normal vectors of each branch-leaf mixed point cloud in the branch-leaf mixed point cloud set is determined to determine whether each branch-leaf mixed point cloud is a regional growth point of the seed point, including:

[0093] When the normal angle between the normal vector of the seed point and the normal vector of the adjacent branch-leaf mixed point cloud is less than the preset smoothing threshold, or the difference between 180 degrees and the normal angle is less than the smoothing threshold, the branch-leaf mixed point cloud is determined to be the regional growth point of the seed point;

[0094] When the normal angle between the normal vector of the seed point and the normal vector of the adjacent branch-leaf mixed point cloud is greater than the preset smoothing threshold, or the difference between 180 degrees and the normal angle is greater than the smoothing threshold, the branch-leaf mixed point cloud is determined to be the regional growth point of the seed point.

[0095] According to the above embodiment, adjacent points with smooth surfaces can be divided into the same branch area, and the normal line changes very little.

[0096] Figure 4This is a structural block diagram of a branch and leaf separation device for vegetation point clouds around transmission lines according to an embodiment of the present invention.

[0097] like Figure 4 As shown, the branch and leaf separation device for the vegetation point cloud around the transmission line may include:

[0098] A first point cloud classification module 410 is configured to divide the surrounding vegetation point cloud of the target transmission line into a first branch point cloud set and a first leaf point cloud set based on the curvature of each point cloud in the surrounding vegetation point cloud;

[0099] A connected domain segmentation module 420 is configured to perform connected domain segmentation on the first branch point cloud set to obtain a plurality of connected components;

[0100] A second point cloud classification module 430 is configured to divide the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set based on the linear feature significance of the geometric distribution features of each of the connected components;

[0101] The third point cloud classification module 440 is configured to use each branch point in the second branch point cloud set as a seed point and divide the branch-leaf mixed point cloud set into regional growth point cloud sets and non-regional growth point cloud sets by region growing;

[0102] A first union module 450 is configured to determine a target branch point cloud set of the target transmission line based on a union of the second branch point cloud set and the regional growth point cloud set;

[0103] The second union module 460 is configured to determine a target blade point cloud set for the target transmission line based on the union of the first blade point cloud set and the non-region growth point cloud set.

[0104] In one embodiment, the above device further comprises:

[0105] a covariance matrix determination module, configured to determine the covariance matrix of each point cloud based on the position information of the geometric center of the vegetation point cloud surrounding the target transmission line within a specified neighborhood and the position information of each point cloud in the vegetation point cloud;

[0106] A singular value decomposition module, configured to perform singular value decomposition on the covariance matrix of each point cloud to obtain a singular eigenvalue of each point cloud;

[0107] The curvature determination module is configured to determine the curvature of each point cloud based on the singular eigenvalue of each point cloud.

[0108] In one embodiment, the first point cloud classification module 410 includes:

[0109] a first classification unit, configured to add the point cloud to the first leaf point cloud set if the local curvature of the point cloud is greater than a preset local curvature threshold;

[0110] The second classification unit is configured to add the point cloud to the first branch point cloud set when the local curvature of the point cloud is less than a preset local curvature threshold.

[0111] In one embodiment, the above device further comprises:

[0112] A geometric feature determination module, configured to determine the linear feature, the planar feature, and the scattering feature of the connected component based on the geometric distribution feature of the connected component;

[0113] a linear feature determination module, configured to process the linear feature, the planar feature, and the scattering feature based on a nonlinear adjustment coefficient, a planarity weight, and a scattering weight to obtain a linear feature significance value;

[0114] a nonlinear feature determination module, configured to process the planar feature and the scattering feature based on the nonlinear adjustment coefficient, the planarity weight, and the scattering weight to obtain a nonlinear feature significance value;

[0115] The linear significance determination module is used to determine the linear feature significance of the connected component based on the ratio between the linear feature significance value and the nonlinear feature significance value.

[0116] In one embodiment, the second point cloud classification module 430 includes:

[0117] a third classification unit, configured to add the connected component to the second branch point cloud set if the linear feature significance of the connected component is greater than a preset linear feature significance threshold;

[0118] The fourth classification unit is used to add the connected component to the branch-leaf mixed point cloud set when the linear feature significance of the connected component is less than a preset linear feature significance threshold.

[0119] In one embodiment, the third point cloud classification module 440 is specifically configured to:

[0120] Extracting seed points from the second branch point cloud set, deleting the seed points from the second branch point cloud set, and performing the following first operation on the seed points, returning to continue extracting the seed points until the second branch point cloud set is an empty set, stopping extracting the seed points, and using the branch-leaf mixed point cloud set as the non-region growth point cloud set;

[0121] The first operation includes:

[0122] Determining whether each of the branch-leaf mixed point clouds is a regional growth point of the seed point based on a normal angle between the normal vector of the seed point and the normal vectors of each branch-leaf mixed point cloud in the branch-leaf mixed point cloud set;

[0123] For any branch-leaf mixed point cloud in the branch-leaf mixed point cloud set, if the branch-leaf mixed point cloud is the regional growth point of the seed point, the branch-leaf mixed point cloud is added to the regional growth point cloud set, and the branch-leaf mixed point cloud is deleted from the branch-leaf mixed point cloud set.

[0124] In one embodiment, the determining whether each of the branch-leaf mixed point clouds is a regional growth point of the seed point based on the angle between the normal vector of the seed point and the normal vector of each branch-leaf mixed point cloud in the branch-leaf mixed point cloud set includes:

[0125] When the normal angle between the normal vector of the seed point and the normal vector of the adjacent branch-leaf mixed point cloud is less than a preset smoothing threshold, or the difference between 180 degrees and the normal angle is less than the smoothing threshold, the branch-leaf mixed point cloud is determined to be the regional growth point of the seed point;

[0126] When the normal angle between the normal vector of the seed point and the normal vector of the adjacent branch-leaf mixed point cloud is greater than a preset smoothing threshold, or the difference between 180 degrees and the normal angle is greater than the smoothing threshold, the branch-leaf mixed point cloud is determined to be the regional growth point of the seed point.

[0127] For the description of specific functions and examples of each module and submodule of the system in the embodiment of the present invention, please refer to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0128] In the technical solution of the present invention, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0129] According to an embodiment of the present invention, the present invention further provides a system and a readable storage medium.

[0130] Figure 5A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0131] like Figure 5 As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. Computing unit 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to bus 804.

[0132] Various components in device 800 are connected to I / O interface 805, including an input unit 806, such as a keyboard, mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 809, such as a network card, modem, wireless communication transceiver, etc. The communication unit 809 allows device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0133] The computing unit 801 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the branch and leaf separation method for a point cloud of vegetation around a power transmission line. For example, in some embodiments, the branch and leaf separation method for a point cloud of vegetation around a power transmission line can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the branch and leaf separation method for a point cloud of vegetation around a power transmission line described above can be performed. Alternatively, in other embodiments, the computing unit 801 may be configured in any other appropriate manner (for example, by means of firmware) to execute the branch and leaf separation method for the vegetation point cloud around the transmission line.

[0134] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0136] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0138] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0139] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0140] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present invention can be achieved. This is not limited herein.

[0141] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for separating branches and leaves from vegetation point clouds around power transmission lines, characterized in that: include: Based on the curvature of each point cloud in the vegetation point cloud surrounding the target transmission line, dividing the vegetation point cloud surrounding the target transmission line into a first branch point cloud set and a first leaf point cloud set; Performing connected domain segmentation on the first branch point cloud set to obtain multiple connected components; The linear feature significance based on the geometric distribution characteristics of each connected component , dividing the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set; Merging the point clouds in the first leaf point cloud set into the branch-leaf mixed point cloud set, so that the first leaf point cloud set is an empty set; Taking each branch point in the second branch point cloud set as a seed point, the branch-leaf mixed point cloud set is divided into a regional growth point cloud set and a non-regional growth point cloud set by regional growing, including: extracting seed points from the second branch point cloud set, deleting the seed points in the second branch point cloud set, and performing the following first operation on the seed points, returning to continue extracting the seed points until the second branch point cloud set is an empty set, stopping extracting the seed points, and using the branch-leaf mixed point cloud set as the non-regional growth point cloud set; the first operation includes: determining whether each branch-leaf mixed point cloud is a regional growth point of the seed point based on the normal angle between the normal vector of the seed point and the normal vector of each branch-leaf mixed point cloud in the branch-leaf mixed point cloud set; for any branch-leaf mixed point cloud in the branch-leaf mixed point cloud set, if the branch-leaf mixed point cloud is the regional growth point of the seed point, adding the branch-leaf mixed point cloud to the regional growth point cloud set, and deleting the branch-leaf mixed point cloud in the branch-leaf mixed point cloud set; Determining a target branch point cloud set of the target transmission line based on a union of the second branch point cloud set and the regional growth point cloud set; Determining a target blade point cloud set of the target transmission line based on a union of the first blade point cloud set of an empty set and the non-region growth point cloud set; constructing a target tree based on the target branch point cloud set and the target leaf point cloud set, and analyzing the target tree to evaluate whether the target tree poses a power transmission hazard to the target transmission line; Among them, the linear feature significance of the connected component : ； in, represents linear features, represents the nonlinear adjustment coefficient, Represents a plane feature, represents the scattering characteristics, represents the planarity weight, represents the scattering weight; The planarity weight, the scattering weight, the nonlinear adjustment coefficient and the linear feature significance are The parameter combination of the threshold is the combination parameter that makes the classification accuracy the highest among multiple parameter combinations.

2. The method according to claim 1, characterized in that Also includes: Determining the covariance matrix of each point cloud based on the position information of the geometric center of the target transmission line in the surrounding vegetation point cloud within the specified neighborhood and the position information of each point cloud in the surrounding vegetation point cloud; Performing singular value decomposition on the covariance matrix of each point cloud to obtain a singular eigenvalue of each point cloud; The curvature of each of the point clouds is determined based on the singular eigenvalues ​​of each of the point clouds.

3. The method according to claim 1, characterized in that The dividing of the surrounding vegetation point cloud into a first branch point cloud set and a first leaf point cloud set based on the local curvature of each point cloud in the surrounding vegetation point cloud of the target transmission line comprises one of the following: When the local curvature of the point cloud is greater than a preset local curvature threshold, adding the point cloud to the first blade point cloud set; When the local curvature of the point cloud is less than a preset local curvature threshold, the point cloud is added to the first branch point cloud set.

4. The method according to claim 1, wherein Also includes: Determining linear features, planar features, and scattering features of the connected components based on geometric distribution features of the connected components; Processing the linear feature, the planar feature, and the scattering feature based on a nonlinear adjustment coefficient, a planarity weight, and a scattering weight to obtain a linear feature significance value; Processing the planar feature and the scattering feature based on the nonlinear adjustment coefficient, the planarity weight, and the scattering weight to obtain a nonlinear feature significance value; The linear feature significance level of the connected component is determined based on the ratio between the linear feature significance value and the nonlinear feature significance value.

5. The method according to claim 1, characterized in that The dividing of the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set based on the linear feature significance of the geometric distribution features of each of the connected components includes one of the following: When the linear feature significance of the connected component is greater than a preset linear feature significance threshold, adding the connected component to the second branch point cloud set; When the linear feature significance of the connected component is less than a preset linear feature significance threshold, the connected component is added to the branch-leaf mixed point cloud set.

6. The method according to claim 1, wherein The determining whether each of the branch-leaf mixed point clouds is a regional growth point of the seed point based on the angle between the normal vector of the seed point and the normal vector of each branch-leaf mixed point cloud in the branch-leaf mixed point cloud set includes: When the normal angle between the normal vector of the seed point and the normal vector of the adjacent branch-leaf mixed point cloud is less than a preset smoothing threshold, or the difference between 180 degrees and the normal angle is less than the smoothing threshold, the branch-leaf mixed point cloud is determined to be the regional growth point of the seed point; When the normal angle between the normal vector of the seed point and the normal vector of the adjacent branch-leaf mixed point cloud is greater than a preset smoothing threshold, or the difference between 180 degrees and the normal angle is greater than the smoothing threshold, the branch-leaf mixed point cloud is determined to be the regional growth point of the seed point.

7. A branch and leaf separation device for vegetation point cloud around power transmission lines, characterized in that: include: a first point cloud classification module, configured to divide the vegetation point cloud surrounding the target transmission line into a first branch point cloud set and a first leaf point cloud set based on the curvature of each point cloud in the vegetation point cloud surrounding the target transmission line; A connected domain segmentation module, configured to perform connected domain segmentation on the first branch point cloud set to obtain a plurality of connected components; The second point cloud classification module is used to classify the linear features based on the geometric distribution characteristics of each connected component. , dividing the first branch point cloud set into a second branch point cloud set and a branch-leaf mixed point cloud set; a module for merging the point clouds in the first leaf point cloud set into the branch-leaf mixed point cloud set, so that the first leaf point cloud set is an empty set; The third point cloud classification module is used to use each branch point in the second branch point cloud set as a seed point, and divide the branch-leaf mixed point cloud set into a regional growth point cloud set and a non-regional growth point cloud set through regional growing, including: extracting seed points from the second branch point cloud set, deleting the seed points in the second branch point cloud set, and performing the following first operation on the seed points, returning to continue extracting the seed points until the second branch point cloud set is an empty set, stopping extracting the seed points, and using the branch-leaf mixed point cloud set as the non-regional growth point cloud set; the first operation includes: determining whether each branch-leaf mixed point cloud is a regional growth point of the seed point based on the normal angle between the normal vector of the seed point and the normal vector of each branch-leaf mixed point cloud in the branch-leaf mixed point cloud set; for any branch-leaf mixed point cloud in the branch-leaf mixed point cloud set, if the branch-leaf mixed point cloud is the regional growth point of the seed point, adding the branch-leaf mixed point cloud to the regional growth point cloud set, and deleting the branch-leaf mixed point cloud in the branch-leaf mixed point cloud set; A first union module is configured to determine a target branch point cloud set of the target transmission line based on a union of the second branch point cloud set and the regional growth point cloud set; A second union module is configured to determine a target blade point cloud set of the target transmission line based on a union of the first blade point cloud set and the non-region growth point cloud set; A module for constructing a target tree based on the target branch point cloud set and the target leaf point cloud set, and analyzing the target tree to evaluate whether the target tree poses a power transmission hazard to the target transmission line; Among them, the linear feature significance of the connected component : ； in, represents linear features, represents the nonlinear adjustment coefficient, Represents a plane feature, represents the scattering characteristics, represents the planarity weight, represents the scattering weight; The planarity weight, the scattering weight, the nonlinear adjustment coefficient and the linear feature significance are The parameter combination of the threshold is the combination parameter that makes the classification accuracy the highest among multiple parameter combinations.

8. A system for separating branches and leaves from vegetation point clouds around power transmission lines, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

Images