Power wire data processing method, device, equipment, medium and program product

Abstract

The invention provides an electric power wire data processing method and device, equipment, a medium and a program product, and relates to the technical field of data processing, and the method comprises the steps: obtaining the point cloud data of at least two electric power wires; the point cloud data is acquired by the acquisition equipment based on a motion track; the acquisition equipment is located at a lower position between the at least two power wires; the motion trail rotates in the first direction point by point based on the position starting point to collect point cloud data of part of the power conductors, and then rotates in the second direction point by point to collect point cloud data of the remaining power conductors; the first direction is opposite to the second direction, and three-dimensional reconstruction of the power wire is carried out according to the point cloud data, so that the point-by-point rotation scanning mode covers the collection range of the whole power wire, it is ensured that a sufficient amount of point cloud data is collected, and the data collection efficiency is improved through the ordered scanning path; and three-dimensional reconstruction of the power conductor is carried out based on the point cloud data collected by the motion trail, so that the accuracy of reconstruction of the power conductor is improved.

Description

[0001] This application claims priority to Chinese Patent Application No. 202410516502.3, filed on April 25, 2024, entitled “Power Conductor Data Processing Method, Apparatus, Equipment, Medium and Program Product”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of data processing technology, and in particular to a method, apparatus, equipment, medium, and program product for processing data of power conductors. Background Technology

[0003] LiDAR (Light Detection and Ranging) technology is a high-precision measurement technology that measures the distance, velocity, and orientation of a target object by emitting a laser beam and receiving the reflected echo. LiDAR can generate high-resolution three-dimensional point cloud data, thereby constructing accurate three-dimensional models. These three-dimensional models can be used to analyze the spatial position, shape, and surrounding environment of power lines, which helps to optimize the design and maintenance of power systems.

[0004] In existing technologies, lidar can scan power lines at a relatively long distance, enabling rapid and high-precision measurement and 3D modeling of the lines. By scanning the lines, information such as their position, height, direction, and bending radius can be obtained, thereby achieving comprehensive monitoring and evaluation of the lines.

[0005] However, when it comes to LiDAR scanning data mounted on a pan-tilt unit, this type of scanning data often only contains point cloud data of no more than 15% of the total length of the power line. If this type of scanning data is used for 3D reconstruction of the power line in cases where point cloud data for the missing part of the power line is missing, the accuracy of the 3D reconstruction will be low. Summary of the Invention

[0006] This application provides a method, apparatus, device, medium, and program product for processing power conductor data, which addresses the problem that existing point cloud data on missing sections of conductors results in low accuracy in 3D reconstruction of power conductors when such scanned data is used.

[0007] In a first aspect, this application provides a method for processing power conductor data, the method comprising:

[0008] The point cloud data of at least two power lines is acquired; the point cloud data is acquired by the acquisition device based on a motion trajectory; the acquisition device is positioned below the at least two power lines; the motion trajectory is based on a point-by-point rotation in a first direction to acquire point cloud data of a portion of the power lines, and then point-by-point rotation in a second direction to acquire point cloud data of the remaining power lines; the first direction and the second direction are opposite.

[0009] Three-dimensional reconstruction of power lines is performed based on the point cloud data.

[0010] Optionally, the first direction is upward and the second direction is downward; the motion trajectory is based on the starting point of the position, rotating vertically upward point by point until reaching the position point above the power pole, and then rotating vertically downward point by point until returning to the starting point of the position; during the rotation process, horizontal rotation is performed at each position point so that the acquisition device can acquire point cloud data of the power conductor at the position point.

[0011] Optionally, acquire point cloud data for at least two power lines, including:

[0012] Acquire the first point cloud data of at least two power lines within a preset time period;

[0013] The first point cloud data is spliced ​​and deduplicated to obtain the second point cloud data;

[0014] Based on a preset spatial bounding box, the second point cloud data is processed to obtain point cloud data of power lines located within the spatial bounding box.

[0015] Optionally, before performing three-dimensional reconstruction of the power conductors based on the point cloud data, the method further includes:

[0016] A statistical filtering algorithm is used to filter the point cloud data to remove abnormal data.

[0017] Optionally, performing three-dimensional reconstruction of the power conductors based on the point cloud data includes:

[0018] The point cloud data is clustered to obtain at least one set of point cloud cluster data;

[0019] Three-dimensional reconstruction of power conductors is performed based on at least one set of point cloud cluster data.

[0020] Optionally, three-dimensional reconstruction of power conductors is performed based on the at least one set of point cloud cluster data, including:

[0021] For each set of point cloud clusters, calculate the rotation angle required for the orthographic projection of the power conductor;

[0022] Based on the rotation angle, a spatial transformation matrix is ​​constructed, and the point cloud cluster data is transformed by projection coordinates based on the spatial transformation matrix to obtain the target point cloud data;

[0023] The target point cloud data is segmented based on the number of power lines to obtain at least one set of point cloud clusters; the target point cloud data in each set of point cloud clusters corresponds to the same power line.

[0024] The first equation and the second equation of the at least one set of point cloud clusters are calculated on the first projection plane and the second equation on the second projection plane, respectively, and the three-dimensional reconstruction of the power conductor is performed based on the first equation and the second equation.

[0025] Optionally, the target point cloud data is segmented based on the number of power lines to obtain at least one set of point cloud clusters, including:

[0026] The target point cloud data is segmented based on a preset step size to obtain at least one segment, and the maximum value of the point cloud data in each segment is obtained.

[0027] Based on the maximum value of the point cloud data, a clustering algorithm is used to perform cluster analysis on the target point cloud data in each segment to obtain multiple initial point cloud clusters;

[0028] Calculate the three-dimensional bounding box of each initial point cloud cluster, and use the three-dimensional bounding box to determine the coordinates of the plane center point;

[0029] Obtain the distance threshold between power lines, and classify the multiple initial point cloud clusters based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold to obtain at least one point cloud cluster group.

[0030] Optionally, based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold, the multiple initial point cloud clusters are classified to obtain at least one point cloud cluster group, including:

[0031] For each initial point cloud cluster, determine whether the coordinates of the plane center point meet a preset condition; the preset condition is determined based on the distance threshold.

[0032] If so, the initial point cloud cluster corresponding to the coordinates of the plane center point is added to the first point cloud cluster;

[0033] If not, then create a second point cloud cluster for the initial point cloud cluster corresponding to the coordinates of the center point of the plane, and add the initial point cloud cluster to the second point cloud cluster;

[0034] The first point cloud cluster and the second point cloud cluster are summarized and filtered to obtain at least one set of point cloud clusters.

[0035] Optionally, calculating the first equation and the second equation of the at least one set of point cloud clusters in the first projection plane and the second equation in the second projection plane respectively includes:

[0036] For each point cloud cluster, calculate the cross-sectional area of ​​the point cloud cluster on the third projection plane;

[0037] The target point cloud data in the point cloud cluster is filtered based on the cross-sectional area to obtain the third point cloud data.

[0038] Using the third point cloud data, the first equation of the first projection plane and the second equation of the second projection plane are obtained by fitting.

[0039] Optionally, a three-dimensional reconstruction of the power conductor is performed based on the first equation and the second equation, including:

[0040] At preset intervals, the fourth point cloud data of a single-strand power conductor is determined based on the first equation and the second equation, respectively;

[0041] The fourth point cloud data is used for fitting to generate the power conductor equation.

[0042] Optionally, the method further includes:

[0043] Acquire edge point cloud data and 3D reconstruction results of power poles on both sides of any two sides;

[0044] Based on the edge point cloud data, the point cloud data, and the results of the 3D reconstruction, supplementary point cloud data is determined;

[0045] The supplementary point cloud data and the point cloud data are labeled and stored respectively.

[0046] Optionally, the method further includes:

[0047] Acquire edge point cloud data and 3D reconstruction results of power poles on both sides of any two sides;

[0048] Based on the edge point cloud data, the point cloud data, and the results of the 3D reconstruction, supplementary point cloud data is determined;

[0049] The supplementary point cloud data and the point cloud data are labeled and stored respectively.

[0050] Secondly, this application provides a power conductor data processing apparatus, the apparatus comprising:

[0051] An acquisition module is used to acquire point cloud data of at least two power conductors; the point cloud data is acquired by an acquisition device based on a motion trajectory; the acquisition device is positioned below the at least two power conductors; the motion trajectory is based on a point-by-point rotation in a first direction from a starting point to acquire point cloud data of a portion of the power conductors, and then point-by-point rotation in a second direction to acquire point cloud data of the remaining power conductors; the first direction and the second direction are opposite.

[0052] The reconstruction module is used to perform three-dimensional reconstruction of power lines based on the point cloud data.

[0053] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0054] The memory stores computer-executed instructions;

[0055] The processor executes computer execution instructions stored in the memory to implement the method as described in any one of the first aspects.

[0056] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the method as described in any one of the first aspects.

[0057] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method as described in any one of the first aspects.

[0058] In summary, this application provides a method, apparatus, device, medium, and program product for processing power line data. By utilizing the characteristics of a data acquisition device, such as a motion lidar, mounted on a rotatable platform, a rotatable motion trajectory is designed. Based on this trajectory, comprehensive and accurate point cloud data can be acquired. Specifically, the acquisition device is positioned below at least two power lines. If the acquisition device is positioned to capture point cloud data from the furthest power line, choosing to acquire data below the power lines provides a better viewing angle, ensuring clear and complete point cloud data. Step by step, after completing the initial fixed-point scan at the lower position, the acquisition device begins to rotate point by point in the first direction based on the starting point, and then controls the rotation point by point in the second direction. The first and second directions are opposite. In this way, the point-by-point rotation scanning method can cover the entire acquisition range of the power line, ensuring that a sufficient amount of point cloud data is collected. This orderly scanning path can improve the efficiency of data acquisition, reduce unnecessary movement and adjustment, and reduce repeated scanning and omissions. Furthermore, the three-dimensional reconstruction of the power line is performed based on the point cloud data collected by this motion trajectory, which greatly improves the accuracy of the three-dimensional reconstruction of the power line. Attached Figure Description

[0059] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0060] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of this application;

[0061] Figure 2 A flowchart illustrating a power conductor data processing method provided in an embodiment of this application;

[0062] Figure 3 A schematic diagram of a projection onto the XOY plane is provided as an embodiment of this application;

[0063] Figure 4 A schematic diagram of a projection onto the XOZ plane is provided as an embodiment of this application;

[0064] Figure 5 A flowchart illustrating the process of segmenting point cloud data for power lines is provided in an embodiment of this application.

[0065] Figure 6 This is a schematic diagram of the structure of a power conductor data processing device provided in an embodiment of this application;

[0066] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0067] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0068] To facilitate a clear description of the technical solutions in the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish identical or similar items with essentially the same function and purpose. For example, "first device" and "second device" are merely used to distinguish different devices and do not limit their order of execution. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that "first" and "second" do not necessarily imply that they are different.

[0069] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0070] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0071] In power line scanning, lidar technology can achieve high-precision measurement and 3D modeling of the conductors, thus providing more accurate and comprehensive conductor information.

[0072] In one possible implementation, scanning the conductor can acquire information such as its position, height, direction, and bending radius, thereby enabling comprehensive monitoring and evaluation of the conductor. Furthermore, lidar technology can be combined with other sensors and data processing technologies to achieve automated inspection and intelligent monitoring of power conductors, further improving the operational efficiency and safety of power equipment.

[0073] However, when it comes to LiDAR scanning data mounted on a pan-tilt unit, this type of scanning data often only contains point cloud data of no more than 15% of the total length of the power line. If this type of scanning data is used for 3D reconstruction of the power line in cases where point cloud data for the missing part of the power line is missing, the accuracy of the 3D reconstruction will be low.

[0074] In another possible implementation, after acquiring the 3D point cloud data of the power line using the LiDAR system, the 3D point cloud data is preprocessed, such as through denoising and filtering, to improve data quality. Furthermore, feature extraction techniques are used to identify and segment the point cloud data related to the power line. Then, curve fitting or surface fitting algorithms (such as least squares method, spline interpolation, etc.) are applied to perform high-precision geometric modeling on the segmented point cloud data related to the power line. Combined with the physical characteristics of the power line, the model parameters are optimized to achieve 3D reconstruction of the power conductor, further improving the model accuracy.

[0075] In another possible implementation, power line point cloud data acquired by an airborne LiDAR system is used to achieve accurate 3D reconstruction of a single power conductor between individual towers (i.e., a single span). This involves using an airborne LiDAR system to quickly and efficiently collect high-precision 3D point cloud data of the power line corridor. Furthermore, through fine processing of the point cloud data, including filtering and segmentation, a point cloud set representing the power conductor is extracted from the massive data. Finally, a pre-defined straight-parabolic hybrid model is used to fit the point cloud set to achieve 3D reconstruction of the power conductor.

[0076] Specifically, by preprocessing the 3D point cloud data, ground and other non-power line target points are removed, while power line feature points are retained. Furthermore, a clustering algorithm suitable for the characteristics of power lines is used to identify and separate point cloud data belonging to a single power conductor. Then, based on this point cloud data, a proposed straight-parabolic hybrid model is used for fitting to achieve 3D reconstruction of the power conductor. In addition, the model parameters can be iteratively adjusted to make the reconstruction model fit the shape and size of the actual power conductor to the greatest extent. This reconstruction model is a model that combines straight line segments and parabolic segments to describe the 3D shape of the power conductor, and can better adapt to the shape changes of the conductor under different conditions.

[0077] It should be noted that the above method can realize automated and high-precision three-dimensional reconstruction of power conductors, which helps to improve the safety and efficiency of power line operation and maintenance. In practical applications, comparative tests have verified the applicability and accuracy of the method in complex terrain and long-distance transmission line scenarios.

[0078] However, both of the above methods rely on most of the data on the span between power poles when solving the problem of power conductor point cloud data coverage. They share the common feature of using algorithms to effectively fit and supplement a small number of missing point clouds. However, when faced with point cloud data on a concentrated portion of the conductor length that is missing, such as when the power conductor point cloud data only covers less than 15% of the span between power poles, these two methods still cannot accurately and completely complete the missing data, resulting in low accuracy of the 3D reconstruction of power conductors.

[0079] Understandably, traditional point cloud completion algorithms for power lines primarily focus on scenarios involving point cloud data from both ends of the power line. In these cases, the length of the missing portion in the required data is relatively low compared to the total length. However, the situation becomes more complex when dealing with LiDAR scan data mounted on a pan-tilt unit. This presents the challenge of dealing with scan data containing only point cloud information representing no more than 15% of the total power line length. This makes traditional completion algorithms unsuitable, significantly reducing accuracy in achieving automatic power line completion. Therefore, even if these scan data are used to accurately acquire key measurement data such as the sag point of the power line, the low accuracy of the fitted conductor makes it difficult to implement important business functions such as distance measurement based on this limited scan data.

[0080] To address the aforementioned problems, this application provides a method for processing power line data. By utilizing the characteristics of a data acquisition device, such as a moving lidar, mounted on a rotatable platform, a rotatable motion trajectory is designed. Based on this trajectory, comprehensive and accurate point cloud data can be acquired. Specifically, the acquisition device is positioned below at least two power lines. If the acquisition device is positioned to capture point cloud data from the furthest power line, choosing to acquire data below the power lines provides a better viewing angle, ensuring clear and complete point cloud data. Furthermore, after initial point scanning at this position, the acquisition device begins to rotate point-by-point in a first direction, then controls rotation point-by-point in a second direction (the first and second directions are opposite). This point-by-point rotation scanning method covers the entire acquisition range of the power lines, ensuring a sufficient amount of point cloud data is collected. This orderly scanning path improves data acquisition efficiency, reduces unnecessary movement and adjustments, and minimizes duplicate scanning and omissions. Furthermore, the point cloud data acquired based on this motion trajectory is used for 3D reconstruction of the power lines, significantly improving the accuracy of the 3D reconstruction.

[0081] For example, Figure 1 This is a schematic diagram of an application scenario provided in an embodiment of this application, such as... Figure 1As shown, this application scenario can be applied to the reconstruction of power conductors in power transmission tunnels. This application scenario includes: LiDAR 101 and data processing system 102.

[0082] Specifically, the lidar 101 collects high-precision point cloud data near the top of the power conductor based on a preset motion trajectory, and sends the point cloud data to the data processing system 102 for processing. The data processing system 102 uses the point cloud data collected by the lidar 101 to perform three-dimensional reconstruction of the power conductor and construct a detailed three-dimensional model of the power conductor.

[0083] The motion trajectory is a composite motion trajectory designed for the LiDAR 101 to improve the coverage of point cloud data acquisition of power lines. At the beginning of the acquisition, the LiDAR 101 is positioned to capture point cloud data of the furthest power lines. After completing the initial fixed-point scan, the LiDAR 101 begins to rotate point-by-point in a first direction based on its starting point, and then controls the rotation point-by-point in a second direction, such as... Figure 1 As shown, the lidar 101 rotates vertically upwards point by point until it reaches the top of the power pole (vertex A), and then rotates vertically downwards along the other side of the power line.

[0084] It should be noted that the embodiments of this application do not limit the specific orientation and angle corresponding to the first direction and the second direction, as long as the first direction and the second direction are opposite. In this way, by scanning along the opposite direction, it can be ensured that the acquisition device covers the entire target area, reducing blind spots and omissions, thereby obtaining comprehensive point cloud data.

[0085] Optionally, after obtaining the three-dimensional model of the power conductors, the power conductors inside the power transmission corridor can be accurately completed and reconstructed. In this way, the completion technology based on point cloud data can greatly improve the efficiency and accuracy of power facility management.

[0086] In this way, the above method can not only ensure the integrity of power conductor data, but also significantly improve the reliability and practicality of the constructed three-dimensional model in subsequent business operations such as ranging, monitoring and evaluation.

[0087] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0088] Figure 2 This is a flowchart illustrating a method for processing power conductor data according to an embodiment of this application. This method can be applied to the aforementioned data processing system, such as... Figure 2 As shown, the power conductor data processing method includes the following steps:

[0089] S201. Acquire point cloud data of at least two power conductors; the point cloud data is acquired by a data acquisition device based on a motion trajectory; the data acquisition device is positioned below the at least two power conductors; the motion trajectory is based on a point-by-point rotation in a first direction from a starting point to acquire point cloud data of a portion of the power conductors, and then point-by-point rotation in a second direction to acquire point cloud data of the remaining power conductors; the first direction and the second direction are opposite.

[0090] The lower position is determined based on the collectible range of the acquisition device.

[0091] In this embodiment of the application, in order to improve the coverage of point cloud data acquisition of power lines by the acquisition device, a composite motion trajectory is designed. The motion trajectory includes: at the beginning of the acquisition, the acquisition device is positioned below the point cloud data of the farthest power line. The position below is determined based on the acquisition range of the acquisition device. The specific position below the acquisition device is different in different application scenarios.

[0092] The location below is situated between at least two power conductors, such as... Figure 1 As shown, point C is between two power lines, and lidar 101 is on the same vertical line as point C.

[0093] Optionally, the lower position can be manually adjusted. The acquisition device is placed at a first position below the power lines. At this first position, the acquisition device collects point cloud data of the power lines and sends the collected point cloud data to the data processing system. The data processing system analyzes whether the first position is the lower position where point cloud data of the farthest power line can be collected. If not, the acquisition device is manually moved to a second position, and the above process is repeated until the corresponding lower position is found. If so, the first position is determined to be the lower position.

[0094] The data acquisition device can be a lidar, a camera, an ultrasonic sensor, etc. This application does not specifically limit the type of data acquisition device.

[0095] Optionally, the first direction is upward and the second direction is downward; the motion trajectory is based on the starting point of the position, rotating vertically upward point by point until reaching the position point above the power pole, and then rotating vertically downward point by point until returning to the starting point of the position; during the rotation process, horizontal rotation is performed at each position point so that the acquisition device can acquire point cloud data of the power conductor at the position point.

[0096] For example, taking a lidar as the data acquisition device, after the lidar completes an initial fixed-point scan at the lower position, the lidar base position starting point begins to rotate vertically upwards point by point, while simultaneously rotating horizontally at each position point to maintain continuous tracking of the power lines. Since vertical upward rotation causes the visible range of the power lines to gradually shrink, horizontal rotation is added at each position point to ensure that power line data from at least one side of the lidar can always be captured during the scanning process. That is, after the lidar vertically scans above the power pole based on the position starting point, it then rotates vertically downwards along the direction of the other acquisition device until it returns to the initial position (position starting point).

[0097] It should be noted that the embodiments of this application do not specifically limit the process of returning to the starting point of the position, such as... Figure 1 As shown, one can either rotate horizontally from point D to point B, or continue rotating from point D along the second direction and return to point B after a 180-degree rotation.

[0098] It should be noted that if the initial position of the data acquisition device is not located at the starting point, such as at... Figure 1 At point A, the initial position of the acquisition device can be rotated to the starting point, and then S201 can be executed. That is, the initial position of the acquisition device can be located at any position other than the starting point. The initial position of the acquisition device can be rotated to the starting point first, and then S201 can be executed to reduce the movement trajectory of the acquisition, save acquisition time and computing resources, and improve work efficiency.

[0099] This vertical upward and downward movement trajectory ensures that all parts of the power poles and conductors can be collected, avoiding data blind spots. Moreover, by rotating vertically and horizontally point by point, the acquisition device can collect data from multiple angles and positions, ensuring the integrity of the power poles and conductors in the data collection, which is helpful for subsequent analysis and modeling.

[0100] Optionally, the first direction is downward and the second direction is upward. That is, the initial position of the acquisition device is not located at the starting point, but at any position above the horizontal line. Then the motion trajectory can be based on the starting point, rotating vertically downward point by point. After reaching the starting point, it rotates vertically upward point by point until it reaches the position point above the power pole. Then it controls the vertical downward rotation point by point until it returns to the starting point. During the rotation, it rotates horizontally at each position point so that the acquisition device can acquire point cloud data of the power conductor at the position point.

[0101] The embodiments of this application do not limit the specific orientation and angle of the first direction and the second direction, as long as they are opposite. For example, the first direction can also be diagonally upward and the second direction can be diagonally downward.

[0102] S202. Perform three-dimensional reconstruction of the power conductors based on the point cloud data.

[0103] In this embodiment of the application, since the collected point cloud data may correspond to more than one power conductor, the point cloud data corresponding to multiple power conductors can be segmented and / or clustered, and then the three-dimensional reconstruction of the power conductors can be performed based on the segmented and / or clustered point cloud data.

[0104] Optionally, S202 includes: performing clustering processing on the point cloud data to obtain at least one set of point cloud cluster data; and performing three-dimensional reconstruction of the power conductor based on the at least one set of point cloud cluster data.

[0105] Clustering processing divides point cloud data into multiple clusters, each representing an independent part. This clustering process can be performed using clustering algorithms to determine which clusters represent power lines and to identify outlier data points that do not conform to any cluster. Clustering algorithms include DBSCAN (Density-Based Spatial Clustering of Applications with Noise), K-means clustering, Euclidean clustering, Euclidean Cluster Extraction, etc. This application does not specifically limit the clustering algorithm used for clustering processing.

[0106] Taking the Euclidean Cluster Extraction clustering algorithm as an example, it can perform in-depth analysis on the acquired point cloud data to identify and separate a group of power conductor clusters. Each cluster contains a set of point cloud data representing power conductors. Then, by comparing and filtering the length of each set of point cloud data along the X-axis, the power conductor cluster with the longest projected length can be determined.

[0107] As is understandable, the Euclidean Cluster Extraction (ECE) algorithm is a commonly used point cloud data clustering algorithm used to extract different objects or structures from 3D point cloud data. This algorithm is typically applied in fields such as robot navigation, autonomous vehicles, and terrain analysis. Below is a brief description of the Euclidean Cluster Extraction algorithm's processing flow:

[0108] Step 1): Input data, that is, the input is a set of 3D point cloud data.

[0109] Step 2): Clustering process, which includes:

[0110] Initialization: Select an unlabeled point as the starting point for clustering.

[0111] Neighborhood search: For each point, find all neighboring points that are less than a preset threshold. This distance can be the Euclidean distance. In this embodiment, the size of the preset threshold is not specifically limited. It can be determined based on the actual application scenario, such as the distribution of power lines.

[0112] Cluster expansion: Add these neighboring points to the current cluster and mark them as processed.

[0113] Recursive execution: For each newly added point, repeat the above process until no more points can be added to the current cluster.

[0114] Repeat the steps: If there are still unprocessed points, select a new unlabeled point as the starting point for the next cluster and repeat the above process.

[0115] Termination condition: The algorithm terminates when all points are labeled as part of a cluster.

[0116] Step 3): Output the result, which is a set of multiple clusters, each cluster representing a connected region.

[0117] For example, in the application scenario of power line detection, the point cloud data of power lines collected from LiDAR is processed based on the Euclidean Cluster Extraction clustering algorithm to identify the specific location of the power lines. Optionally, point cloud data near the power lines is obtained using LiDAR, and then the point cloud data is filtered or denoised. The Euclidean Cluster Extraction clustering algorithm is then used to cluster the filtered or denoised point cloud data. Based on appropriate key parameters, the characteristics of each cluster, such as location and orientation, are analyzed to determine which clusters represent power lines.

[0118] The key parameters include the distance threshold and the minimum number of points. The distance threshold refers to the maximum distance between two points that are considered to be members of the same cluster. The minimum number of points is used to specify how many points a cluster must contain to be considered a valid cluster.

[0119] Understandably, the Euclidean Cluster Extraction algorithm is a simple and effective method for clustering point cloud data, especially suitable for situations where specific structures need to be separated from complex environments. In power line inspection projects, by reasonably setting key parameters, the location and shape of power lines can be effectively identified, thereby supporting subsequent monitoring and maintenance work.

[0120] It should be noted that multiple point cloud clusters can be processed in parallel to improve data processing efficiency.

[0121] Therefore, by using clustering, massive point cloud data can be divided into at least one set of point cloud clusters. This allows the complex 3D reconstruction process to be broken down into multiple simple steps, where each step only needs to process one point cloud cluster. This simplifies the 3D reconstruction process, and clustering ensures that each point cloud cluster can be processed, avoiding any omissions and ensuring the accuracy and completeness of the power line reconstruction.

[0122] In this embodiment of the application, the three-dimensional reconstruction can be either fitting and generating at least one power conductor equation or fitting and generating point cloud data corresponding to the power conductor. This embodiment of the application does not specifically limit the result of the three-dimensional reconstruction.

[0123] Optionally, the 3D reconstruction process may include point cloud fitting, curve fitting, and model generation. This application embodiment does not specifically limit the 3D reconstruction process.

[0124] Thus, in this embodiment of the application, the location of the acquisition device is set below the power lines and determined based on the collectable range. This location selection ensures that the acquisition device can fully cover all parts of the power lines. In addition, this application also designs a point-by-point rotation trajectory to ensure high data acquisition accuracy at each location point. As a result, the generated point cloud data is more detailed and accurate. The comprehensiveness and accuracy of the data improve the quality of the basic data for 3D reconstruction, ensuring that the generated model is more accurate and reliable.

[0125] For example, a detailed and accurate point cloud dataset can be reconstructed in 3D, covering the complete conductor span between two poles. In this way, the 3D reconstruction results are not only complete and accurate, but can also be directly applied to business processes such as distance measurement, significantly improving work efficiency.

[0126] Optionally, acquire point cloud data for at least two power lines, including:

[0127] Acquire the first point cloud data of at least two power lines within a preset time period;

[0128] The first point cloud data is spliced ​​and deduplicated to obtain the second point cloud data;

[0129] Based on a preset spatial bounding box, the second point cloud data is processed to obtain point cloud data of power lines located within the spatial bounding box.

[0130] Optionally, this application may also acquire first point cloud data of the power conductor on at least one side of the acquisition device within a preset time period.

[0131] In this embodiment of the application, the preset time period refers to at least the time period corresponding to the point cloud data collected by the acquisition device based on the motion trajectory. This embodiment of the application does not specifically limit the length of the preset time period. The preset time period is designed to ensure that a more comprehensive and accurate power conductor point cloud dataset is obtained.

[0132] Since there is duplicate data in the point cloud data collected at various locations, especially the point cloud data collected from any two adjacent locations, the first point cloud data of the power conductors on at least one side collected by the acquisition device can be spliced ​​to remove duplicates and obtain the second point cloud data. This can remove redundant data, significantly reduce the total amount of point cloud data, and reduce the complexity of data processing.

[0133] Furthermore, through stitching processing, point cloud data collected from multiple locations can be integrated into a complete point cloud dataset, ensuring the integrity of subsequent 3D reconstruction.

[0134] Optionally, in order to distinguish the point cloud data of power lines collected by the acquisition device based on the motion trajectory, tags can be added to the point cloud data within a preset time period during the data processing stage, thereby achieving data labeling.

[0135] In this embodiment of the application, in order to filter out the point cloud data of power lines in a specific direction from the field of view and avoid mixing in interference data from other directions, and considering that the installation position of a single pan-tilt device and the preset scanning angle position are preset, the range of point cloud data obtained each time is relatively fixed. Therefore, by pre-setting a bounding box size that only contains the point cloud data of the target power line, the point cloud data of the power line collected at the preset position is filtered out.

[0136] Specifically, based on the labels set in the point cloud data of the power conductors and the spatial bounding box size (box1) of the power conductors, key information is extracted from the complete LAS format second point cloud data acquired by the LiDAR, such as the three-dimensional spatial coordinates (X, Y, Z) of each point, the acquisition timestamp, and radar reflectivity information. Furthermore, the CloudCompare tool is used to perform visualization analysis of the three-dimensional spatial coordinates (X, Y, Z) to obtain the point cloud data of the power conductors within the spatial bounding box. The spatial bounding box not only limits the spatial range of the power conductor point cloud data but also provides the maximum and minimum values ​​of the three-dimensional spatial coordinates, providing an important basis for subsequent data processing and analysis.

[0137] In this way, irrelevant data can be effectively eliminated based on the above steps, ensuring the accuracy and completeness of the point cloud data of the power conductors, and laying a solid foundation for subsequent data analysis and processing.

[0138] It should be noted that this application only requires point cloud data of power lines collected from preset locations, without involving complex calculations of large amounts of data. Therefore, the method of this application can be easily embedded into embedded devices for data processing. Thus, after the Simultaneous Localization and Mapping (SLAM) operation is performed at the front end of the embedded device, the SLAM algorithm can be directly used on the front end to perform power line completion. This approach avoids the cumbersome process of transmitting data to the back end for additional optimization, saving data transmission resources and significantly reducing time costs, thereby achieving real-time and efficient data processing and application.

[0139] Optionally, before performing three-dimensional reconstruction of the power conductors based on the point cloud data, the method further includes:

[0140] A statistical filtering algorithm is used to filter the point cloud data to remove abnormal data.

[0141] In this embodiment of the application, in order to reduce the impact of noise values ​​and abnormal point cloud data on the results during the acquisition process, an efficient statistical filtering algorithm can be used to remove outliers in the point cloud data.

[0142] Among them, the statistical filtering algorithm is based on the principle of statistical analysis. It performs detailed statistical analysis on each point cloud data, calculates the statistical characteristics in its neighborhood, such as the mean and standard deviation, and then accurately determines which points are noise or outliers based on these statistical characteristics, and effectively filters out the noise or outliers.

[0143] In this way, by using statistical filtering algorithms to process the point cloud data, a clean set of point cloud data was obtained, which only contains point cloud data of power lines. Through filtering, not only was the accuracy and reliability of the data improved, but also the efficiency and precision of data processing were enhanced, providing a high-quality data foundation for subsequent analysis and applications.

[0144] Optionally, three-dimensional reconstruction of power conductors is performed based on the at least one set of point cloud cluster data, including:

[0145] For each set of point cloud clusters, calculate the rotation angle required for the orthographic projection of the power conductor;

[0146] Based on the rotation angle, a spatial transformation matrix is ​​constructed, and the point cloud cluster data is transformed by projection coordinates based on the spatial transformation matrix to obtain the target point cloud data;

[0147] The target point cloud data is segmented based on the number of power lines to obtain at least one set of point cloud clusters; the target point cloud data in each set of point cloud clusters corresponds to the same power line.

[0148] The first equation and the second equation of the at least one set of point cloud clusters are calculated on the first projection plane and the second equation on the second projection plane, respectively, and the three-dimensional reconstruction of the power conductor is performed based on the first equation and the second equation.

[0149] The first projection plane refers to the XOY plane, and the second plane refers to the XOZ plane. The XOY plane is a plane formed by the X-axis and the Y-axis, and the XOZ plane is a plane formed by the X-axis and the Z-axis.

[0150] For example, after analyzing point cloud data using the Euclidean Cluster Extraction clustering algorithm to identify and separate at least one group of point cloud cluster data for power lines, the least squares method is used to solve the equations of the projected cross-section lines on the XOY and XOZ planes for the three-dimensional spatial coordinate information of the point cloud cluster data corresponding to the power lines. The precise geometric parameters of the projected cross-section line equations are then obtained. Furthermore, the horizontal (yaw) and pitch (pitch) rotation angles required for the pan-tilt unit to face the power pole are calculated using the arctangent function based on the geometric parameters. Based on these two key rotation angle values, a spatial transformation matrix can be constructed to achieve orthographic projection of the point cloud data, thereby guiding the pan-tilt unit to perform precise spatial positioning and orientation adjustment.

[0151] It should be noted that the method for obtaining the spatial transformation matrix is ​​not specifically limited in the embodiments of this application. The above is only an example. The gimbal device includes a data acquisition device and a gimbal. The data acquisition device is deployed on the gimbal, and the gimbal controls the data acquisition device to move based on the motion trajectory.

[0152] Based on the spatial transformation matrix required for the orthographic projection of the power conductor obtained from the aforementioned steps, the point cloud data of the power conductor can be accurately projected. The purpose of this projection transformation operation is to simplify the representation of the point cloud data in three-dimensional spatial coordinates, making its projection on the XYZ axes clearer.

[0153] Figure 3 This is a schematic diagram of a projection onto the XOY plane provided in an embodiment of this application. After projection transformation, as shown... Figure 3 As shown, the data for a line segment of the power conductor pointing towards the power poles on both sides can be obtained on the XOY plane. Figure 4 A schematic diagram of a projection onto the XOZ plane is provided as an embodiment of this application, such as... Figure 4 As shown, on the XOZ plane, a catenary-like electric conductor data is presented. This projection data not only facilitates the observation and analysis of the shape and position of the electric conductor, but also provides a more intuitive and concise data foundation for subsequent data processing and applications.

[0154] Furthermore, after transforming the point cloud cluster data into projected coordinates based on the spatial transformation matrix to obtain the target point cloud data, the target point cloud data can be segmented based on the number of power conductors, wherein the target point cloud data in each group of point cloud clusters corresponds to the same power conductor; then, at least one group of point cloud clusters is calculated in terms of the first equation on the first projection plane and the second equation on the second projection plane, so as to perform three-dimensional reconstruction of the power conductor based on the first equation and the second equation.

[0155] In this way, by calculating the rotation angle and constructing the spatial transformation matrix, the point cloud data can be accurately projected onto the appropriate coordinate system to obtain accurate target point cloud data for subsequent processing. Furthermore, by segmenting the target point cloud data based on the number of power lines, different lines can be processed separately. This segmentation method helps to more clearly identify and analyze the state of each line. Furthermore, by calculating equations on the first and second projection planes respectively, the power lines can be analyzed from different perspectives. Moreover, multi-perspective analysis helps to comprehensively understand the three-dimensional structure and spatial position of the power lines, improving the accuracy of reconstruction. Therefore, in this application, by using rotation and projection transformation, power lines under different terrains and environments can be transformed into planes to obtain suitable target point cloud data, ensuring that the three-dimensional reconstruction of power lines can be accurately performed under various conditions.

[0156] Optionally, the target point cloud data is segmented based on the number of power lines to obtain at least one set of point cloud clusters, including:

[0157] The target point cloud data is segmented based on a preset step size to obtain at least one segment, and the maximum value of the point cloud data in each segment is obtained.

[0158] Based on the maximum value of the point cloud data, a clustering algorithm is used to perform cluster analysis on the target point cloud data in each segment to obtain multiple initial point cloud clusters;

[0159] Calculate the three-dimensional bounding box of each initial point cloud cluster, and use the three-dimensional bounding box to determine the coordinates of the plane center point;

[0160] Obtain the distance threshold between power lines, and classify the multiple initial point cloud clusters based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold to obtain at least one point cloud cluster group.

[0161] In this embodiment, the preset step size is the length of the wire corresponding to the random segmentation of the target point cloud data, which is defined in advance. This embodiment does not specifically limit the preset step size, which can be determined based on the application scenario requirements.

[0162] For example, Figure 5This application provides a schematic diagram of the process for segmenting point cloud data of power lines, as illustrated in the embodiments of this application. Figure 5 As shown, the process includes the following steps:

[0163] Step A: Load the target point cloud data and sort it along the X-axis to facilitate the removal of point cloud data of power poles interspersed in the nearby target point cloud data. For example, a descending order can be used, that is, first traverse the distant target point cloud data, and then gradually load the nearby target point cloud data so that the target point cloud data can be filtered sequentially along the X-axis direction during subsequent target point cloud data traversal. Further, load the coordinates of each point in the target point cloud data in sequence to obtain the three-dimensional spatial coordinate value (x, y, z) of each point, and then start the data traversal process, proceeding to Step B.

[0164] Step B: Load a point coordinate X, and determine whether X is greater than minX. minX represents the maximum value of the X-axis of the point cloud coordinate in the current segment, which is used to determine the segmented loading of point cloud data. If yes, proceed to step C; otherwise, the traversal of the point coordinate ends. The segment is pre-divided based on a preset step size.

[0165] Step C: Use a clustering algorithm to perform point cloud cluster analysis to obtain multiple initial point cloud clusters.

[0166] Optionally, point cloud clustering analysis can be performed based on the Euclidean Cluster Extraction algorithm to obtain multiple initial point cloud clusters. The Euclidean Cluster Extraction algorithm is a clustering method for point cloud data processing, especially when processing data from LiDAR or depth cameras. It can identify and extract clusters or groups in the point cloud based on Euclidean distance. Its basic idea is to group points that are close to each other in space into one class, considering them to belong to the same object or surface.

[0167] Specifically, the workflow of the Euclidean Cluster Extraction algorithm includes the following steps:

[0168] Step 1) Before starting clustering, define two key parameters: the search radius (or neighborhood radius) and the minimum number of points. The search radius determines which points are considered to be close to each other, while the minimum number of points is used to ensure that a cluster contains at least that many points.

[0169] Step 2) Traverse each point in the target point cloud data, and use each point as the center to find other points within the search radius.

[0170] Step 3) If a point has enough points around it within the search radius (at least the minimum number of points), these points are considered a cluster, the point is marked as part of the cluster, and is removed from the list of unprocessed points.

[0171] Step 4) For a point that has already been marked as part of a cluster, it is necessary to check whether other points in its neighborhood have also been marked. If not, these points are added to the cluster and removed from the list of unprocessed points. The above process is repeated iteratively until no more points can be added to the cluster.

[0172] Step 5) Continue processing the remaining unmarked points, repeating steps 2) to 4) until all points have been processed.

[0173] Step 6) Output a set of clusters, each containing a set of spatially close points.

[0174] Step D: Obtain the coordinates of the center point of each YOZ plane cluster. That is, based on the initial point cloud cluster output in Step C, traverse the initial point cloud cluster, calculate the size of the three-dimensional bounding box of each cluster, and obtain the center point coordinates in the YOZ plane based on the three-dimensional bounding box, i.e., the center point coordinates of the plane.

[0175] Step E: Since one or more of the initial point cloud clusters may include point cloud data belonging to the same power line, it is necessary to classify the initial point cloud clusters to obtain at least one set of point cloud cluster data. Optionally, the initial point cloud clusters can be classified based on the distance threshold between power lines.

[0176] In this embodiment, the target point cloud data is divided into multiple segments by a preset step size, allowing for more refined data processing. The maximum value of the point cloud data in each segment is obtained, which helps identify key points and feature points, providing a foundation for subsequent clustering analysis. Furthermore, by using clustering algorithms to analyze the target point cloud data in each segment, different point cloud clusters can be more accurately identified and separated. Calculating the 3D bounding box of each initial point cloud cluster accurately determines the spatial range and location of the cluster. Using the 3D bounding box to determine the coordinates of the plane's center point helps in further analysis and classification of the clusters. Obtaining the distance threshold between power lines and classifying them based on the plane's center point coordinates and distance thresholds effectively distinguishes and identifies different power lines. Classifying multiple initial point cloud clusters yields at least one cluster group, facilitating accurate identification and analysis of each power line. Therefore, through refined segmentation, clustering, and classification, different point cloud clusters can be more accurately identified and separated, providing a high-quality data foundation for 3D reconstruction.

[0177] Optionally, based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold, the multiple initial point cloud clusters are classified to obtain at least one point cloud cluster group, including:

[0178] For each initial point cloud cluster, determine whether the coordinates of the plane center point meet a preset condition; the preset condition is determined based on the distance threshold.

[0179] If so, the initial point cloud cluster corresponding to the coordinates of the plane center point is added to the first point cloud cluster;

[0180] If not, then create a second point cloud cluster for the initial point cloud cluster corresponding to the coordinates of the center point of the plane, and add the initial point cloud cluster to the second point cloud cluster;

[0181] The first point cloud cluster and the second point cloud cluster are summarized and filtered to obtain at least one set of point cloud clusters.

[0182] For example, step E further includes the following steps, such as Figure 5 As shown, after executing step D, step F is executed.

[0183] Step F: Based on the distance value between the plane center point coordinates obtained in Step D and the center point coordinates of the existing power conductor point cloud clusters, compare it with the set farthest distance threshold. If the distance value is less than the farthest distance threshold, it means that the point cloud cluster belongs to the power conductor group, so add the point cloud cluster to the group to obtain the first point cloud cluster; otherwise, create a new group of point cloud clusters, indicating that the point cloud cluster does not belong to any existing power conductor group, so create and add it to a new group of power conductors to obtain the second point cloud cluster.

[0184] The farthest distance threshold is the preset distance threshold between power lines. In this application embodiment, the size of the threshold is not specifically limited, and it can be determined based on the actual application scenario.

[0185] Step G: Assign minX+Xspan to the new minX, that is, obtain the next data point for judgment. Xspan is the interval length of the X-axis of the point cloud coordinates in the segment. After the initial point cloud cluster is traversed, execute step B until all initial point cloud clusters are traversed. Further, summarize the first point cloud cluster and the second point cloud cluster to obtain at least one set of point cloud clusters of power conductors.

[0186] Step H: Traverse each generated power conductor point cloud cluster and compare the size of the point cloud data of each power conductor in the X-axis direction of the 3D bounding box with the preset shortest length threshold. If it is less than the shortest length threshold, it means that the length of the power conductor point cloud data is insufficient, and the point cloud cluster can be filtered out. Thus, through step H, at least one set of point cloud data containing multiple sets of power conductors can be obtained, where each set of data uses the same power conductor.

[0187] In this way, by judging whether the coordinates of the center point of the plane meet the preset conditions (based on the distance threshold), the initial point cloud clusters can be accurately classified. For the initial point cloud clusters that do not meet the preset conditions, a second point cloud cluster can be created and added. This dynamic adjustment mechanism can flexibly cope with different point cloud data characteristics, summarize and filter the first and second point cloud clusters, integrate the point cloud data of different power lines, and filter out abnormal data to reduce the impact of noise and abnormal data, thereby improving the accuracy of three-dimensional reconstruction.

[0188] Optionally, calculating the first equation and the second equation of the at least one set of point cloud clusters in the first projection plane and the second equation in the second projection plane respectively includes:

[0189] For each point cloud cluster, calculate the cross-sectional area of ​​the point cloud cluster on the third projection plane;

[0190] The target point cloud data in the point cloud cluster is filtered based on the cross-sectional area to obtain the third point cloud data.

[0191] Using the third point cloud data, the first equation of the first projection plane and the second equation of the second projection plane are obtained by fitting.

[0192] The third projection plane refers to the YOZ plane, which is the plane formed by the Y-axis and the Z-axis. The first equation refers to the equation of the straight line of the projection cross section, and the second equation refers to the equation of the quadratic curve of the projection cross section. The specific forms of the first and second equations are not limited in the embodiments of this application.

[0193] For example, by Figure 5The processing flow of the illustrated embodiment can obtain point cloud data of at least one set of point cloud clusters. However, sometimes other types of point cloud data, such as insulators, may be mixed in with the point cloud clusters. In order to filter out pure power conductor point cloud data and avoid this type of data from affecting the calculation accuracy of the fitting equation parameters of the power conductors, the data in the point cloud clusters can be traversed and the cross-sectional area of ​​the data in each point cloud cluster projected onto the YOZ plane can be calculated. If the cross-sectional area of ​​a certain target point cloud data exceeds a preset threshold, it means that this part of the data may not belong to the power conductor with a small cross-section, but to objects such as insulators. At this time, this part of the target point cloud data can be removed to obtain the third point cloud data. Furthermore, the third point cloud data is used to fit the linear equation of the projected cross-section of the XOY plane and the quadratic curve equation of the projected cross-section of the XOZ plane.

[0194] Thus, the present application embodiment filters the target point cloud data based on the cross-sectional area, which can remove irrelevant or noisy data. Fitting based on the filtered third point cloud data helps to improve the accuracy of the fitting equation.

[0195] Optionally, a three-dimensional reconstruction of the power conductor is performed based on the first equation and the second equation, including:

[0196] At preset intervals, the fourth point cloud data of a single-strand power conductor is determined based on the first equation and the second equation, respectively;

[0197] The fourth point cloud data is used for fitting to generate the power conductor equation.

[0198] In this embodiment, the preset distance is a pre-set distance value used to obtain point cloud data on a single power conductor. For example, every 0.05 meters, a value is taken on the first equation of the first projection plane and the second equation of the second projection plane to obtain the point coordinate data of the target point cloud data on the Y-axis and the point coordinate data on the Z-axis. This embodiment does not specifically limit the size of the preset distance, which can be determined based on the actual application scenario.

[0199] For example, the target point cloud data in each point cloud cluster can be traversed. Based on the target point cloud data of the cluster, the least squares method is used to obtain the linear equation of the projected cross section of the XOY plane (the first equation of the first projection plane) and the quadratic curve equation of the projected cross section of the XOZ plane (the second equation of the second projection plane). Furthermore, an X-axis point coordinate data is generated every 0.05 meters. Then, the Y-axis point coordinate data and Z-axis point coordinate data are determined based on the X-axis point coordinate data. In this way, the fourth point cloud data (X, Y, Z) of the single-strand power conductor can be determined. Then, the power conductor equation can be generated by fitting based on the fourth point cloud data.

[0200] The Least Squares Method (LSM) is a mathematical optimization technique used to find the best function fit for data by minimizing the sum of squared errors. Least Squares is widely used in regression analysis, especially linear regression. The core idea of ​​Least Squares is to find a function (usually a linear function) that minimizes the sum of squared differences between the function and the given data points.

[0201] Specifically, if there are n data points (x1, y1), (x2, y2), ..., (xn, yn), then we can find a function y = f(x, a) (where a is the parameter of the function) such that the sum of the squares of the differences between all data points and the function is minimized.

[0202] It is understandable that this application uses the least squares method to derive the first equation of the first projection plane and the second equation of the second projection plane. A major advantage is that it is easy to compute and, in most cases, yields good results. However, the least squares method also has some limitations. For example, it assumes that the errors are independent and identically distributed and follow a normal distribution. If these assumptions do not hold, then the least squares method may not be the optimal choice. Therefore, the embodiments of this application are not limited to deriving the first equation of the first projection plane and the second equation of the second projection plane based on the least squares method. Other fitting algorithms can also be used to derive the first equation of the first projection plane and the second equation of the second projection plane. The embodiments of this application do not specifically limit this.

[0203] In summary, the least squares method is a powerful and commonly used mathematical tool that can be used to find a good fit model from data.

[0204] Therefore, by sampling at preset intervals, this embodiment of the application can ensure the uniform spatial distribution of point cloud data, which helps to accurately locate the spatial position of the power conductor. Then, the point cloud data of a single power conductor can be determined using the first and second equations, and a more accurate power conductor equation can be generated by fitting the point cloud data. The accurate fitting equation can more accurately describe the geometry and spatial position of the power conductor, providing a reliable mathematical model for subsequent analysis and application.

[0205] Optionally, the method further includes:

[0206] Acquire edge point cloud data and 3D reconstruction results of power poles on both sides of any two sides;

[0207] Based on the edge point cloud data, the point cloud data, and the results of the 3D reconstruction, supplementary point cloud data is determined;

[0208] The supplementary point cloud data and the point cloud data are labeled and stored respectively.

[0209] For example, taking the result of 3D reconstruction as at least one power conductor equation as an example, for the reconstruction of power conductors between any two power poles, after generating supplementary point cloud data using at least one power conductor equation, this supplementary point cloud data can be stored in a LAS format file for subsequent processing and application. To clearly distinguish the fitted supplementary point cloud data from the original collected data, a special tag can be added when saving this data. This tag, as a unique identifier, can clearly identify which data was generated by fitting the power conductor equation and which data was originally collected.

[0210] Optionally, the result of this 3D reconstruction can also be a complete point cloud dataset spanning both sides of the power poles and containing multiple power conductors. That is, after fitting the power conductor equation, a point cloud dataset for at least one power conductor can be determined based on the power conductor equation. Furthermore, based on the edge point cloud data, the point cloud data, and the point cloud dataset, supplementary point cloud data is determined, and the supplementary point cloud data and the point cloud data are labeled and stored respectively.

[0211] In this way, the labeling method not only ensures the accuracy and integrity of the data, but also facilitates subsequent data processing and analysis, ensuring that all types of data are correctly applied and processed.

[0212] It should be noted that if only one set of data is obtained during the processing of the above embodiments, such as obtaining a set of initial point cloud clusters, the execution process of the above embodiments can also be executed. The embodiments of this application do not specifically limit the amount of data required for the specific execution process.

[0213] In the foregoing embodiments, the power conductor data processing method provided by the embodiments of this application has been described. To implement the functions of the methods provided by the embodiments of this application, the electronic device serving as the execution subject may include hardware structures and / or software modules, implementing the above functions in the form of hardware structures, software modules, or a combination of hardware structures and software modules. Whether a particular function is executed in the form of hardware structures, software modules, or a combination of hardware structures and software modules depends on the specific application and design constraints of the technical solution.

[0214] For example, Figure 6 This is a schematic diagram of the structure of a power conductor data processing device provided in an embodiment of this application, as shown below. Figure 6As shown, the device 600 includes: an acquisition module 601, used to acquire point cloud data of at least two power conductors; the point cloud data is acquired by an acquisition device based on a motion trajectory; the acquisition device is positioned below the at least two power conductors; the motion trajectory is based on a position starting point, rotating point by point in a first direction to acquire point cloud data of a portion of the power conductors, and then rotating point by point in a second direction to acquire point cloud data of the remaining power conductors; the first direction and the second direction are opposite.

[0215] The reconstruction module 602 is used to perform three-dimensional reconstruction of power conductors based on the point cloud data.

[0216] Optionally, the first direction is upward and the second direction is downward; the motion trajectory is based on the starting point of the position, rotating vertically upward point by point until reaching the position point above the power pole, and then rotating vertically downward point by point until returning to the starting point of the position; during the rotation process, horizontal rotation is performed at each position point so that the acquisition device can acquire point cloud data of the power conductor at the position point.

[0217] Optionally, module 601 is used for:

[0218] Acquire the first point cloud data of at least two power lines within a preset time period;

[0219] The first point cloud data is spliced ​​and deduplicated to obtain the second point cloud data;

[0220] Based on a preset spatial bounding box, the second point cloud data is processed to obtain point cloud data of power lines located within the spatial bounding box.

[0221] Optionally, before performing three-dimensional reconstruction of the power lines based on the point cloud data, the device 600 further includes a filtering module, which is used for:

[0222] A statistical filtering algorithm is used to filter the point cloud data to remove abnormal data.

[0223] Optionally, the reconstruction module 602 includes clustering units and reconstruction units;

[0224] Clustering unit, used to perform clustering processing on the point cloud data to obtain at least one set of point cloud cluster data;

[0225] A reconstruction unit is used to perform three-dimensional reconstruction of power conductors based on the at least one set of point cloud cluster data.

[0226] Optionally, the reconstruction unit includes a computation unit, a transformation unit, a segmentation unit, and a 3D reconstruction unit;

[0227] The computing unit is used to calculate the rotation angle required for the orthographic projection of the power conductor for each set of point cloud cluster data.

[0228] The transformation unit is used to construct a spatial transformation matrix based on the rotation angle, and to perform projection coordinate transformation on the point cloud cluster data based on the spatial transformation matrix to obtain target point cloud data;

[0229] A segmentation unit is used to segment the target point cloud data based on the number of power conductors to obtain at least one set of point cloud clusters; the target point cloud data in each set of point cloud clusters corresponds to the same power conductor.

[0230] A three-dimensional reconstruction unit is used to calculate the first equation and the second equation of the at least one set of point cloud clusters on the first projection plane and the second equation on the second projection plane, respectively, and to perform three-dimensional reconstruction of the power conductor based on the first equation and the second equation.

[0231] Optional, segmentation unit, specifically used for:

[0232] The target point cloud data is segmented based on a preset step size to obtain at least one segment, and the maximum value of the point cloud data in each segment is obtained.

[0233] Based on the maximum value of the point cloud data, a clustering algorithm is used to perform cluster analysis on the target point cloud data in each segment to obtain multiple initial point cloud clusters;

[0234] Calculate the three-dimensional bounding box of each initial point cloud cluster, and use the three-dimensional bounding box to determine the coordinates of the plane center point;

[0235] Obtain the distance threshold between power lines, and classify the multiple initial point cloud clusters based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold to obtain at least one point cloud cluster group.

[0236] Optionally, the segmentation unit includes a classification unit, which is used for:

[0237] For each initial point cloud cluster, determine whether the coordinates of the plane center point meet a preset condition; the preset condition is determined based on the distance threshold.

[0238] If so, the initial point cloud cluster corresponding to the coordinates of the plane center point is added to the first point cloud cluster;

[0239] If not, then create a second point cloud cluster for the initial point cloud cluster corresponding to the coordinates of the center point of the plane, and add the initial point cloud cluster to the second point cloud cluster;

[0240] The first point cloud cluster and the second point cloud cluster are summarized and filtered to obtain at least one set of point cloud clusters.

[0241] Optional, 3D reconstruction unit, specifically used for:

[0242] For each point cloud cluster, calculate the cross-sectional area of ​​the point cloud cluster on the third projection plane;

[0243] The target point cloud data in the point cloud cluster is filtered based on the cross-sectional area to obtain the third point cloud data.

[0244] Using the third point cloud data, the first equation of the first projection plane and the second equation of the second projection plane are obtained by fitting.

[0245] Optional, 3D reconstruction unit, specifically used for:

[0246] At preset intervals, the fourth point cloud data of a single-strand power conductor is determined based on the first equation and the second equation, respectively;

[0247] The fourth point cloud data is used for fitting to generate the power conductor equation.

[0248] Optionally, the device 600 further includes a storage module, which is used for:

[0249] Acquire edge point cloud data and 3D reconstruction results of power poles on both sides of any two sides;

[0250] Based on the edge point cloud data, the point cloud data, and the results of the 3D reconstruction, supplementary point cloud data is determined;

[0251] The supplementary point cloud data and the point cloud data are labeled and stored respectively.

[0252] It should be noted that the specific implementation principle and effects of the above-mentioned power conductor data processing device can be found in the relevant descriptions and effects of the above embodiments, and will not be elaborated further here.

[0253] This application also provides a schematic diagram of the structure of an electronic device. Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 7 As shown, the electronic device may include: a processor 701 and a memory 702 communicatively connected to the processor; the memory 702 stores a computer program; the processor 701 executes the computer program stored in the memory 702, causing the processor 701 to perform the method described in any of the above embodiments.

[0254] The memory 702 and the processor 701 can be connected via bus 703.

[0255] This application also provides a computer-readable storage medium storing computer program execution instructions, which, when executed by a processor, are used to implement the methods described in any of the foregoing embodiments of this application.

[0256] This application also provides a chip for executing instructions, which is used to perform the methods described in any of the foregoing embodiments executed by an electronic device as described in any of the foregoing embodiments of this application.

[0257] This application also provides a computer program product, which includes a computer program that, when executed by a processor, can implement the methods described in any of the foregoing embodiments executed by an electronic device as described in any of the foregoing embodiments of this application.

[0258] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0259] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to implement the solution of this embodiment according to actual needs.

[0260] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The unit composed of the above modules can be implemented in hardware or in the form of hardware plus software functional units.

[0261] The integrated modules implemented as software functional modules described above can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application.

[0262] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the application can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0263] The memory may include high-speed random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0264] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0265] The aforementioned storage media can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The storage media can be any available medium accessible to general-purpose or special-purpose computers.

[0266] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. Both the processor and the storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components in an electronic device or host device.

[0267] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0268] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0269] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0270] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0271] The above description is merely a specific implementation of the embodiments of this application, but the protection scope of the embodiments of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the embodiments of this application should be covered within the protection scope of the embodiments of this application. Therefore, the protection scope of the embodiments of this application should be determined by the protection scope of the claims.

Claims

1. A method for processing power conductor data, characterized in that, The method includes: The point cloud data of at least two power lines is acquired; the point cloud data is acquired by the acquisition device based on a motion trajectory; the acquisition device is positioned below the at least two power lines; the motion trajectory is based on a point-by-point rotation in a first direction to acquire point cloud data of a portion of the power lines, and then point-by-point rotation in a second direction to acquire point cloud data of the remaining power lines; the first direction and the second direction are opposite. Three-dimensional reconstruction of power lines is performed based on the point cloud data.

2. The method according to claim 1, characterized in that, The first direction is upward, and the second direction is downward; the movement trajectory is based on the starting point of the position, rotating vertically upward point by point until reaching the position point above the power pole, and then rotating vertically downward point by point until returning to the starting point of the position; During the rotation, the device rotates horizontally at each location point so that it can acquire point cloud data of the power line at that location.

3. The method according to claim 1, characterized in that, Obtain point cloud data for at least two power lines, including: Acquire the first point cloud data of at least two power lines within a preset time period; The first point cloud data is spliced ​​and deduplicated to obtain the second point cloud data; Based on a preset spatial bounding box, the second point cloud data is processed to obtain point cloud data of power lines located within the spatial bounding box.

4. The method according to claim 1, characterized in that, Before performing three-dimensional reconstruction of the power conductors based on the point cloud data, the method further includes: A statistical filtering algorithm is used to filter the point cloud data to remove abnormal data.

5. The method according to claim 1, characterized in that, The three-dimensional reconstruction of power lines based on the point cloud data includes: The point cloud data is clustered to obtain at least one set of point cloud cluster data; Three-dimensional reconstruction of power conductors is performed based on at least one set of point cloud cluster data.

6. The method according to claim 5, characterized in that, Three-dimensional reconstruction of power conductors based on the at least one set of point cloud cluster data includes: For each set of point cloud clusters, calculate the rotation angle required for the orthographic projection of the power conductor; Based on the rotation angle, a spatial transformation matrix is ​​constructed, and the point cloud cluster data is transformed by projection coordinates based on the spatial transformation matrix to obtain the target point cloud data; The target point cloud data is segmented based on the number of power lines to obtain at least one set of point cloud clusters; the target point cloud data in each set of point cloud clusters corresponds to the same power line. The first equation and the second equation of the at least one set of point cloud clusters are calculated on the first projection plane and the second equation on the second projection plane, respectively, and the three-dimensional reconstruction of the power conductor is performed based on the first equation and the second equation.

7. The method according to claim 6, characterized in that, The target point cloud data is segmented based on the number of power lines to obtain at least one set of point cloud clusters, including: The target point cloud data is segmented based on a preset step size to obtain at least one segment, and the maximum value of the point cloud data in each segment is obtained. Based on the maximum value of the point cloud data, a clustering algorithm is used to perform cluster analysis on the target point cloud data in each segment to obtain multiple initial point cloud clusters; Calculate the three-dimensional bounding box of each initial point cloud cluster, and use the three-dimensional bounding box to determine the coordinates of the plane center point; Obtain the distance threshold between power lines, and classify the multiple initial point cloud clusters based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold to obtain at least one point cloud cluster group.

8. The method according to claim 7, characterized in that, Based on the coordinates of the plane center point corresponding to each initial point cloud cluster and the distance threshold, the multiple initial point cloud clusters are classified to obtain at least one point cloud cluster group, including: For each initial point cloud cluster, determine whether the coordinates of the plane center point meet a preset condition; the preset condition is determined based on the distance threshold. If so, the initial point cloud cluster corresponding to the coordinates of the plane center point is added to the first point cloud cluster; If not, then create a second point cloud cluster for the initial point cloud cluster corresponding to the coordinates of the center point of the plane, and add the initial point cloud cluster to the second point cloud cluster; The first point cloud cluster and the second point cloud cluster are summarized and filtered to obtain at least one set of point cloud clusters.

9. The method according to claim 6, characterized in that, Calculating the first equation and the second equation of the at least one group of point cloud clusters in the first projection plane and the second equation in the second projection plane, respectively, includes: For each point cloud cluster, calculate the cross-sectional area of ​​the point cloud cluster on the third projection plane; The target point cloud data in the point cloud cluster is filtered based on the cross-sectional area to obtain the third point cloud data. Using the third point cloud data, the first equation of the first projection plane and the second equation of the second projection plane are obtained by fitting.

10. The method according to claim 6, characterized in that, Three-dimensional reconstruction of power conductors based on the first equation and the second equation includes: At preset intervals, the fourth point cloud data of a single-strand power conductor is determined based on the first equation and the second equation, respectively; The fourth point cloud data is used for fitting to generate the power conductor equation.

11. The method according to any one of claims 1-10, characterized in that, The method further includes: Acquire edge point cloud data and 3D reconstruction results of power poles on both sides of any two sides; Based on the edge point cloud data, the point cloud data, and the results of the 3D reconstruction, supplementary point cloud data is determined; The supplementary point cloud data and the point cloud data are labeled and stored respectively.

12. A power conductor data processing device, characterized in that, The device includes: An acquisition module is used to acquire point cloud data of at least two power conductors; the point cloud data is acquired by an acquisition device based on a motion trajectory; the acquisition device is positioned below the at least two power conductors; the motion trajectory is based on a point-by-point rotation in a first direction from a starting point to acquire point cloud data of a portion of the power conductors, and then point-by-point rotation in a second direction to acquire point cloud data of the remaining power conductors; the first direction and the second direction are opposite. The reconstruction module is used to perform three-dimensional reconstruction of power lines based on the point cloud data.

13. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-11.

15. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1-11.

Images