Post-Processing Method and Device, Equipment, and Medium for Power Line Classification Based on Point Cloud

Abstract

The present disclosure provides a post - processing method and device for power line classification based on point cloud, an electronic device, and a storage medium. First, power line point cloud is extracted from the original point cloud data; then, the power line point cloud is segmented and clustered to determine the target number of strands of the power line. After that, the line type is determined according to the target number of strands, and different methods are used to obtain the single - power - line point cloud according to different line types; then, the single - power - line point cloud is fitted according to a parabola, and the fitted point cloud is sampled and filled at a preset step length to obtain a simulated power line point cloud; finally, a KD - tree is used to search for laser points at a preset distance from the simulated power line point cloud, and the searched laser points are used as the points corresponding to the root simulated power line point cloud, and the category of the laser points is set to power line. The solution of the present disclosure reduces the workload of manual inspection and correction of the power line classification result in the later stage, and improves the accuracy and efficiency of power line point cloud classification.

Description

Technical Field

[0001] The present disclosure relates to the field of power technologies, and particularly to the field of laser point cloud processing technologies. It discloses a post-processing method and device for power line classification based on point cloud, an electronic device, and a storage medium. Background Art

[0002] Currently, there are many automatic classification schemes for power line point cloud data. However, due to the complexity of power scenarios and the particularity of point cloud data, the current automatic classification schemes often have large deviations. One of the main challenges is that there are complex ground object structures, occlusions, and noises in the point cloud data, which make it difficult for automatic classification algorithms to identify and classify power lines. In addition, the current automatic classification schemes often face limitations in terms of accuracy and integrity when processing power line point cloud data. For example, in different power line scenarios, the distribution and quantity distribution of power lines are diverse, making it difficult for the current schemes to accurately distinguish different types of power lines. Moreover, there may be incomplete or missing parts in the point cloud data, which further increases the difficulty of automatic classification. These factors lead to certain limitations in the classification accuracy and robustness of automatic classification algorithms.

[0003] Due to the limitations of the above automatic classification schemes, manual post-processing is usually required at present to improve the accuracy and integrity of the classification results. Manual post-processing requires professionals to check, correct, and supplement the automatic classification results to ensure a high degree of accuracy in power line classification. However, this manual processing method is not only time-consuming and laborious but also has low efficiency in processing large-scale data. Summary of the Invention

[0004] The present disclosure provides at least a post-processing method and device for power line classification based on point cloud, an electronic device, and a storage medium, so as to reduce the workload of later manual inspection and correction of power line classification results and improve the accuracy and efficiency of power line point cloud classification.

[0005] According to one aspect of the present disclosure, a post-processing method for power line classification based on point cloud is provided, including:

[0006] Obtain point cloud data within a target power line area, and extract power line point cloud from the point cloud data;

[0007] Along the laying direction of the power line, divide the power line point cloud into multiple segments according to a length of N meters, perform clustering processing on each segment of the power line point cloud respectively, determine the number of power line strands corresponding to each segment of the power line point cloud, and take the number of power line strands that appears the most as the target number of strands; where N is a positive number;

[0008] Determine the line type according to the target number of strands;

[0009] In the case where the line type is a double - circuit line, for each segment of the power line point cloud, determine the average elevation value of each cluster of point clouds corresponding to the segment of the power line point cloud, and determine the layering result of each cluster of point clouds according to the average elevation values; according to the layering results corresponding to each segment of the power line point cloud, combine the point cloud clusters of the same layer to obtain the single - root power line point cloud of the corresponding layer;

[0010] In the case where the line type is a single - circuit line, for each segment of the power line point cloud, divide the segment of the power line point cloud in a direction parallel to the laying direction according to the target number of strands; combine the point clouds on the same layer of each segment of the power line point cloud to obtain the single - root power line point cloud corresponding to each layer;

[0011] For each single - root power line point cloud, fit the point cloud of the single - root power line according to a parabola, and then sample and fill the point cloud obtained by fitting at a preset step size to obtain a simulated power line point cloud;

[0012] For each simulated power line point cloud, establish a KD - tree using the point cloud of the single - root simulated power line, and use the KD - tree to search for laser points at a preset distance from the point cloud of the single - root simulated power line in the point cloud data, and set the category of the searched laser points to power line.

[0013] In a possible implementation manner, the determining the line type according to the target number of strands includes:

[0014] In the case where the target number of strands is odd, determine that the line type is a single - circuit line;

[0015] In the case where the target number of strands is even, determine that the line type is a double - circuit line.

[0016] In a possible implementation manner, the extracting the power line point cloud from the point cloud data includes:

[0017] Perform denoising processing on the point cloud data;

[0018] Segment the denoised point cloud data according to the distance between adjacent power line towers;

[0019] Extract the power line point cloud from each segment of the point cloud data respectively.

[0020] In a possible implementation manner, the extracting the power line point cloud from each segment of the point cloud data respectively includes:

[0021] Use the method of deep learning or the method of point cloud clustering to extract the power line point cloud from each segment of the point cloud data respectively.

[0022] In a possible implementation, the laying direction of the power line is from the power line pole with a smaller number to the power line pole with a larger number.

[0023] In a possible implementation, the number of circuit strands of the single-circuit line is 3.

[0024] In a possible implementation, the laser points at a preset distance from the point cloud of the simulated power line are the laser points at a preset distance from the center of the point cloud of the simulated power line.

[0025] According to another aspect of the present disclosure, there is provided a post-processing device for power line classification based on point cloud, including:

[0026] A point cloud classification module, configured to obtain point cloud data within a target power line area and extract the power line point cloud from the point cloud data;

[0027] A strand number determination module, configured to divide the power line point cloud into multiple segments along the laying direction of the power line at a length of N meters, perform clustering processing on each segment of the power line point cloud respectively, determine the number of power line strands corresponding to each segment of the power line point cloud, and use the power line strand number with the most occurrences as the target strand number; where N is a positive number;

[0028] A line type determination module, configured to determine the line type according to the target strand number;

[0029] A single-line point cloud processing module, configured to, when the line type is a double-circuit line, for each segment of the power line point cloud, determine the average elevation value of each cluster of point clouds corresponding to the segment of the power line point cloud, and determine the layering result of each cluster of point clouds according to the average elevation values; according to the layering results corresponding to each segment of the power line point cloud, combine the point cloud clusters of the same layer to obtain the single power line point cloud of the corresponding layer;

[0030] The single-line point cloud processing module is further configured to, when the line type is a single-circuit line, for each segment of the power line point cloud, divide the segment of the power line point cloud in a direction parallel to the laying direction according to the target strand number; combine the point clouds on the same layer of each segment of the power line point cloud to obtain the single power line point cloud corresponding to each layer;

[0031] A point cloud simulation module, configured to, for each power line point cloud, fit the power line point cloud according to a parabola, and then sample and fill the point cloud obtained by fitting at a preset step length to obtain a simulated power line point cloud;

[0032] A leakage point search module, which is used to build a KD tree for each simulated power line point cloud, search for laser points within a preset distance from the simulated power line point cloud in the point cloud data by using the KD tree, use the searched laser points as points of the simulated power line point cloud, and set the category of the searched laser points to power line.

[0033] According to another aspect of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored on the memory. When the processor executes the computer program, the method described in any one of the above is implemented.

[0034] According to another aspect of the present disclosure, a computer-readable storage medium is provided. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the method described in any one of the above is implemented.

[0035] The power line classification post-processing method, device, electronic device, and storage medium based on point cloud provided by the present disclosure first extract power line point clouds from the original point cloud data; then segment and cluster the power line point clouds to determine the target number of strands of the power line. After that, determine the line type according to the target number of strands, and obtain single power line point clouds in different ways according to different line types; then, fit the single power line point clouds according to a parabola, and then sample and fill the point clouds obtained by fitting according to a preset step length to obtain simulated power line point clouds; finally, use the KD tree to search for laser points within a preset distance from the simulated power line point cloud in the point cloud data, use the searched laser points as points of the corresponding simulated power line point cloud, and set the category of the laser points to power line. The solution of the present disclosure performs parabola fitting, sampling filling, and leakage point search on the classification results of power lines, reduces the workload of later manual inspection and correction of the classification results of power lines, improves the accuracy and efficiency of power line point cloud classification, further improves the automation degree of point cloud classification in power scenarios, and can more effectively support other business requirements based on the automated classification results.

[0036] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. Brief Description of the Drawings

[0037] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0038] Figure 1 is a flowchart of the power line classification post-processing method based on point cloud according to the present disclosure;

[0039] Figure 2It is a schematic structural diagram of a post - processing device for power line classification based on point cloud according to the present disclosure;

[0040] Figure 3 It is a schematic structural diagram of an electronic device according to the present disclosure. Detailed implementation manners

[0041] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well - known functions and structures are omitted below.

[0042] In view of the defects in the current post - processing of power line classification results, such as high labor costs, time - consuming and laborious, and low efficiency in large - scale data processing, the present disclosure provides a method and device for post - processing power line classification based on point cloud, an electronic device, and a storage medium. The solution of the present disclosure performs parabolic fitting, sampling filling, and missing point search on the classification results of power lines, reducing the workload of manual inspection and correction of power line classification results in the later stage, improving the accuracy and efficiency of power line point cloud classification, further increasing the degree of automation of point cloud classification in power scenarios, and being able to more effectively support other business requirements based on automated classification results.

[0043] The technical solutions of the present disclosure will be described below through specific embodiments.

[0044] As shown in Figure 1 , it is a flowchart of the method for post - processing power line classification based on point cloud in this embodiment. The execution subject of this embodiment is a computing device or component with data - processing capabilities. Specifically, the method of this embodiment may include the following steps:

[0045] S110. Obtain point cloud data within a target power line area, and extract power line point cloud from the point cloud data.

[0046] Scan the target power line area with a lidar to obtain the above - mentioned point cloud data. The target power line area is the area where the power lines are located, and the collected point cloud data includes point clouds of power lines, towers, ground, buildings, vegetation, and other objects.

[0047] After collecting the above - mentioned point cloud data, automatically classify the point cloud data and extract the power line point cloud.

[0048] S120. Along the laying direction of the power line, divide the power line point cloud into multiple segments according to a length of N meters. Perform clustering processing on each segment of the power line point cloud respectively, determine the number of power line strands corresponding to each segment of the power line point cloud, and take the number of power line strands with the most occurrences as the target number of strands; where N is a positive number.

[0049] The laying direction of the power line is from the power line tower pole with a smaller number to the power line tower pole with a larger number. The above N can be set according to the actual scenario. For example, set N to 2 meters.

[0050] The above clustering processing can be Euclidean clustering. According to the number of clusters, the number of power line strands can be obtained. In this step, after clustering to obtain multiple numbers of power line strands, the target number of strands is determined by using the mode method.

[0051] S130. Determine the line type according to the target number of strands.

[0052] When the target number of strands is odd, determine that the line type is a single-circuit line; for example, when the target number of strands is 3, the type of the power line is a single-circuit line. When the target number of strands is even, determine that the line type is a double-circuit line.

[0053] Generally, an even number of power lines is a double-circuit line, and the number of power lines in a single-circuit line is 3.

[0054] S140. When the line type is a double-circuit line, for each segment of the power line point cloud, determine the average elevation value of each cluster of point clouds corresponding to the segment of the power line point cloud, and determine the layering result of each cluster of point clouds according to the average elevation values; according to the layering results corresponding to each segment of the power line point cloud, combine the clusters of point clouds in the same layer to obtain the single power line point cloud corresponding to the layer.

[0055] In a double-circuit line, take out the clustering results of each segment of the power line point cloud in the previous step S120. According to the three-dimensional coordinate information of each cluster of point clouds in each segment of the power line point cloud, count the elevation of each point and calculate the average elevation value of each cluster, so as to obtain the relative position of each cluster of point clouds in the elevation direction, thereby determining the layering situation of the segment of the power line point cloud in the elevation, and then combining the layering results of each segment to obtain the single power line point cloud corresponding to each layer.

[0056] S150. When the line type is a single-circuit line, for each segment of the power line point cloud, divide the segment of the power line point cloud in parallel to the laying direction according to the target number of strands; combine the point clouds on the same layer of each segment of the power line point cloud to obtain the single power line point cloud corresponding to each layer.

[0057] When the target number of strands is 3, in a single-circuit line, the point clouds of each section of the power line are divided into upper, middle, and lower parts in a direction parallel to the laying direction, and then the point clouds of 3 power lines can be obtained by combining the point clouds of each section of the power line.

[0058] Steps S140 - S150 determine whether the current line is a single-circuit line or a double-circuit line, and then combine the point clouds of each section of the power line according to the line characteristics of different line types to obtain the point cloud of a single power line.

[0059] S160. For each point cloud of a power line, fit the point cloud of this power line according to a parabola, and then sample and fill the fitted point cloud at a preset step size to obtain the simulated power line point cloud.

[0060] Specifically, since the distance between adjacent poles of the power line is relatively long, it is suitable to use the parabola equation for parabola fitting. Because there may be misclassifications or omissions in the middle of the power line after automatic classification, the point cloud of a single power line obtained in the above steps may be broken or discontinuous. Therefore, after fitting the point cloud of a single power line with the parabola equation, sample and fill the fitted point cloud at a step size of 0.1 meters to obtain the simulated point cloud of this power line, that is, the above-mentioned simulated power line point cloud.

[0061] S170. For each simulated power line point cloud, establish a KD tree using this simulated power line point cloud, and use the KD tree to search for laser points at a preset distance from this simulated power line point cloud in the point cloud data, and set the category of the searched laser points as the power line.

[0062] The laser points at a preset distance from this simulated power line point cloud are the laser points at a first preset distance from the center of this simulated power line point cloud, or the laser points at a second preset distance from the edge of this simulated power line point cloud.

[0063] Specifically, establish a KD tree for the fitted point cloud of a single simulated power line point cloud obtained in step S160, and search for the point cloud within 0.75 meters (the search threshold is generally set slightly larger than the radius of the power line and will not search for other power lines) around in the original point cloud data, and classify the categories of the searched laser points as the category of the power line, so as to achieve the purpose of correctly classifying the misclassified or omitted power line laser points.

[0064] The present disclosure obtains the simulated power line point cloud corresponding to a single power line by segmenting, clustering, and combining the point cloud of the power line between tower poles, and then searches for laser points within a certain threshold range around each power line and classifies them into the category of power lines. This solves the problem of misclassification or omission of power lines after automatic classification of point clouds in the power scenario, reduces the workload of subsequent manual inspection and processing, thereby improving the processing efficiency of point cloud data, and improving the accuracy of classification results and the degree of automation of data production.

[0065] In some embodiments, the extraction of the power line point cloud from the point cloud data can be implemented by the following steps:

[0066] First, denoise the point cloud data; then, segment the denoised point cloud data according to the distance between adjacent power line towers; then, extract the power line point cloud from each segment of the point cloud data respectively.

[0067] Specifically, the method of deep learning or the method of point cloud clustering can be used to extract the power line point cloud from each segment of the point cloud data respectively.

[0068] In some embodiments, the collected three-dimensional point cloud, that is, the point cloud data, is denoised, segmented, and classified. Denoising is to exclude the influence of noise points and discrete points in the point cloud data. Segmentation is to divide the point cloud data into multiple segments according to every two adjacent towers, which reduces the data volume and facilitates subsequent separate processing. Finally, deep learning is used to automatically classify the segmented point cloud. Methods such as deep learning methods or point cloud clustering are used to separate information such as power lines and tower poles to achieve power line classification. Point cloud classification helps to filter out useless points and obtain the required point cloud data.

[0069] Based on the same inventive concept, the present disclosure provides a post-processing device for power line classification based on point cloud. The steps executed by the components of this device are the same as or similar to the above method, so similar parts will not be elaborated. As Figure 2 shown, the post-processing device for power line classification based on point cloud in this embodiment includes:

[0070] A point cloud classification module 210, configured to obtain point cloud data within the target power line area and extract the power line point cloud from the point cloud data.

[0071] A strand number determination module 220, configured to divide the power line point cloud into multiple segments along the laying direction of the power line at a length of N meters, perform clustering processing on each segment of the power line point cloud respectively, determine the number of strands of the power line corresponding to each segment of the power line point cloud, and use the number of strands of the power line that appears the most times as the target number of strands; where N is a positive number.

[0072] A line type determination module 230, configured to determine the line type according to the target number of strands.

[0073] The single-line point cloud processing module 240 is used to, when the line type is a double-circuit line, for each section of the power line point cloud, determine the average elevation value of each cluster of point clouds corresponding to this section of the power line point cloud, and determine the layering result of each cluster of point clouds according to each average elevation value; according to the layering results corresponding to each section of the power line point cloud, combine the point cloud clusters of the same layer to obtain the single power line point cloud of the corresponding layer.

[0074] The single-line point cloud processing module 240 is further used to, when the line type is a single-circuit line, for each section of the power line point cloud, divide this section of the power line point cloud in a direction parallel to the laying direction according to the target number of strands; combine the point clouds of each section of the power line point cloud on the same layer to obtain the single power line point cloud corresponding to each layer.

[0075] The point cloud simulation module 250 is used to, for each power line point cloud, sample and fill this power line point cloud according to a preset step size, and fit it according to a parabola to obtain a simulated power line point cloud.

[0076] The leakage point search module 260 is used to, for each simulated power line point cloud, establish a KD tree using this simulated power line point cloud, and use the KD tree to search for laser points at a preset distance from this simulated power line point cloud in the power line point cloud, and use the searched laser points as the points of this simulated power line point cloud, and set the category of the searched laser points to power line.

[0077] In some embodiments, when the line type determination module 230 determines the line type according to the target number of strands, it is specifically used for:

[0078] When the target number of strands is odd, determine that the line type is a single-circuit line;

[0079] When the target number of strands is even, determine that the line type is a double-circuit line.

[0080] In some embodiments, when the point cloud classification module 210 extracts the power line point cloud in the point cloud data, it is used for:

[0081] Perform denoising processing on the point cloud data;

[0082] Segment the denoised point cloud data according to the distance between adjacent power line towers;

[0083] Extract the power line point cloud in each section of the point cloud data respectively.

[0084] In some embodiments, when the point cloud classification module 210 extracts the power line point cloud in each section of the point cloud data respectively, it is used for:

[0085] Using deep learning methods or point cloud clustering methods, extract the power line point clouds in each segment of point cloud data respectively.

[0086] In some embodiments, the laying direction of the power line is from the power line tower pole with a smaller number to the power line tower pole with a larger number.

[0087] In some embodiments, the number of wire strands of the single - circuit line is 3.

[0088] In some embodiments, the laser points at a preset distance from the root simulated power line point cloud are the laser points at a preset distance from the center of the root simulated power line point cloud.

[0089] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device and a computer - readable storage medium.

[0090] Figure 3 FIG. shows a schematic block diagram of an exemplary electronic device 300 that can be used to implement the embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, 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 disclosure described and / or claimed herein.

[0091] As Figure 3 shown, the device 300 includes a computing unit 310, which can perform various appropriate actions and processes according to a computer program stored in a read - only memory (ROM) 320 or a computer program loaded from a storage unit 380 into a random - access memory (RAM) 330. In the RAM 330, various programs and data required for the operation of the device 300 can also be stored. The computing unit 310, the ROM 320, and the RAM 330 are connected to each other via a bus 340. An input / output (I / O) interface 350 is also connected to the bus 340.

[0092] Multiple components in the device 300 are connected to the I / O interface 350, including: an input unit 360, such as a keyboard, a mouse, etc.; an output unit 370, such as various types of displays, speakers, etc.; a storage unit 380, such as a magnetic disk, an optical disk, etc.; and a communication unit 390, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 390 allows the device 300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0093] The computing unit 310 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 310 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated 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 310 executes the various methods and processes described above. For example, in some embodiments, any of the above methods can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 380. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 300 via the ROM 320 and / or the communication unit 390. When the computer program is loaded into the RAM 330 and executed by the computing unit 310, one or more steps of any of the methods described above can be executed. Alternatively, in other embodiments, the computing unit 310 can be configured to execute any of the methods described above in any other suitable manner (e.g., by means of firmware).

[0094] The various embodiments of the systems and techniques described above in this document 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), complex 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 can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0095] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0096] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0097] In order 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 a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds 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, speech input, or tactile input).

[0098] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend 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.

[0099] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, can also be a server of a distributed system, or a server incorporating a blockchain.

[0100] It should be understood that the various forms of processes shown above can be used, with steps reordered, added or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in this disclosure can be achieved, and no limitations are imposed herein.

[0101] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A point cloud-based power line classification post-processing method, characterized in that: include: Acquire point cloud data within a target power line area, and extract the power line point cloud from the point cloud data; Along the laying direction of the power line, the power line point cloud is divided into multiple sections according to the length of N meters, and each section of the power line point cloud is clustered to determine the number of power line strands corresponding to each section of the power line point cloud, and the number of power line strands with the largest number of occurrences is taken as the target number of strands; wherein N is a positive number; Determining the line type according to the target number of strands; In the case where the line type is a double-circuit line, for each section of the power line point cloud, the average elevation value of each cluster point cloud corresponding to the section of the power line point cloud is determined, and the stratification results of each cluster point cloud are determined according to each average elevation value; according to the stratification results corresponding to each section of the power line point cloud, the point cloud clusters of the same layer are combined to obtain a single power line point cloud of the corresponding layer; In the case where the line type is a single-circuit line, for each section of the power line point cloud, the point cloud of the section of the power line is divided in parallel to the laying direction according to the target number of strands; the point clouds of the sections of the power line point clouds on the same layer are combined to obtain the single power line point clouds corresponding to each layer; For each power line point cloud, the power line point cloud is fitted according to a parabola, and then the fitted point cloud is sampled and filled according to a preset step size to obtain a simulated power line point cloud; For each simulated power line point cloud, a KD tree is established using the root simulated power line point cloud, and the KD tree is used to search for a laser point at a preset distance from the root simulated power line point cloud in the point cloud data, and the searched laser point is used as a point of the root simulated power line point cloud, and the category of the searched laser point is set to power line; Wherein, extracting the power line point cloud from the point cloud data comprises: Performing denoising processing on the point cloud data; According to the distance between adjacent power line towers, the denoised point cloud data is segmented; Extract the power line point clouds from each section of point cloud data respectively; The laser point at a preset distance from the simulated power line point cloud is a laser point at a preset distance from the center of the simulated power line point cloud.

2. The method according to claim 1, characterized in that The determining of the line type according to the target number of strands includes: When the target number of strands is an odd number, determining the line type to be a single-circuit line; When the target number of strands is an even number, the line type is determined to be a double-circuit line.

3. The method according to claim 1, characterized in that The extracting of the power line point clouds from each segment of point cloud data comprises: The power line point clouds in each segment of point cloud data are extracted respectively by using deep learning methods or point cloud clustering methods.

4. The method according to claim 1, characterized in that The laying direction of the power line is from the power line tower pole with a smaller number to the power line tower pole with a larger number.

5. The method according to claim 1, characterized in that: The number of circuit wire strands of the single-circuit line is 3.

6. A point cloud-based power line classification post-processing device, characterized in that: include: A point cloud classification module, used to obtain point cloud data within a target power line area and extract power line point clouds from the point cloud data; The strand number determination module is used to divide the power line point cloud into multiple segments according to the length of N meters along the laying direction of the power line, perform clustering processing on each segment of the power line point cloud, determine the number of power line strands corresponding to each segment of the power line point cloud, and take the number of power line strands with the largest number of occurrences as the target number of strands; wherein N is a positive number; A line type determination module, used to determine the line type according to the target number of strands; The single-line point cloud processing module is used to determine, for each section of the power line point cloud, the average elevation value of each cluster of point clouds corresponding to the section of the power line point cloud, and determine the stratification results of each cluster of point clouds according to each average elevation value; according to the stratification results corresponding to each section of the power line point cloud, the point cloud clusters of the same layer are combined to obtain the single power line point cloud of the corresponding layer; The single-line point cloud processing module is further used to, when the line type is a single-circuit line, for each section of the power line point cloud, divide the section of the power line point cloud in parallel to the laying direction according to the target number of strands; combine the point clouds of the sections of the power line point clouds on the same layer to obtain the single power line point clouds corresponding to each layer; The point cloud simulation module is used to fit each power line point cloud according to a parabola, and then sample and fill the fitted point cloud according to a preset step size to obtain a simulated power line point cloud; A leakage point search module is used to establish a KD tree for each simulated power line point cloud using the root simulated power line point cloud, and use the KD tree to search for a laser point at a preset distance from the root simulated power line point cloud in the point cloud data, and use the searched laser point as a point of the root simulated power line point cloud, and set the category of the searched laser point to power line; Wherein, when extracting the power line point cloud in the point cloud data, the point cloud classification module is specifically used to: Performing denoising processing on the point cloud data; According to the distance between adjacent power line towers, the denoised point cloud data is segmented; Extract the power line point clouds from each section of point cloud data respectively; The laser point at a preset distance from the simulated power line point cloud is a laser point at a preset distance from the center of the simulated power line point cloud.

7. An electronic device comprising a memory, a processor and a computer program stored in the memory, wherein the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images