Pole tower detection data optimization analysis method and system based on point cloud model

Abstract

The invention discloses a tower detection data optimization analysis method and system based on a point cloud model, and belongs to the technical field of electric power facility detection. Clustering analysis is carried out on original point cloud data of a tower by adopting a clustering algorithm, and different point cloud clusters are generated for different parts of the tower; finding out the central axis of the pole tower through a central axis fitting algorithm, carrying out the positioning and structural analysis of the pole tower according to the central axis, carrying out the sorting and splicing of the point cloud clusters of all parts of the pole tower according to the central axis, and comparing the point cloud data with ideal pole tower model data, and obtaining a target pole tower three-dimensional point cloud model; the health condition of the pole tower is evaluated through the model, integration of point cloud data of different parts of the pole tower is achieved, the construction precision of the pole tower three-dimensional point cloud model is improved, and therefore the pole tower condition detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power facility detection, and in particular to a method and system for optimizing and analyzing tower detection data based on a point cloud model. Background Art

[0002] The safe and stable operation of power towers is crucial in power systems. Traditional tower inspection methods have numerous limitations, including inefficient manual inspections, difficulty obtaining comprehensive and accurate detailed information about the towers and their associated structures, and a lack of systematic and intelligent data processing and analysis. While the development of drone technology and point cloud modeling has provided new approaches for tower inspection, there is still room for improvement in data integration, optimization, and intelligent analysis.

[0003] Chinese patent, publication number: CN113554595A, publication date: October 26, 2021, discloses a drone laser radar point cloud tower head deformation detection device and method, including a tower head standard model point cloud module, a tower measured point cloud segmentation module, a tower head point cloud model adaptation module, a tower head point cloud model matching module and a tower head deformation evaluation module. Through automatic identification and extraction of the tower head, rapid scanning of the geometric parameters of the tower head and intelligent evaluation of the tilt parameters are realized. By using advanced three-dimensional laser radar remote sensing technology, the efficiency and accuracy of the line tower head deformation inspection are significantly improved. However, it only detects the tower head and does not analyze the various parts of the tower itself. There are few data sources, the detection direction is relatively single, and the reliability of the detection results is not high. Summary of the Invention

[0004] The present invention aims to solve the problem that the existing point cloud-based tower detection method is difficult to perform comprehensive analysis of various parts of the tower, resulting in low accuracy of tower detection results. The present invention provides a tower detection data optimization analysis method and system based on a point cloud model. By using a clustering algorithm to perform cluster analysis on the original point cloud data of the tower, different point cloud clusters are generated for different parts of the tower. Then, the central axis of the tower is found by a central axis fitting algorithm, and the tower is positioned and structurally analyzed based on the central axis. Then, the point cloud clusters of various parts of the tower are sorted and spliced ​​according to the central axis, and the point cloud data are compared with the ideal tower model data to obtain a three-dimensional point cloud model of the target tower. The health status of the tower is evaluated by using the model, and the point cloud data of different parts of the tower are integrated, thereby improving the construction accuracy of the three-dimensional point cloud model of the tower, thereby improving the accuracy of tower condition detection.

[0005] In a first aspect, a technical solution provided in an embodiment of the present invention is: a method for optimizing and analyzing tower inspection data based on a point cloud model, comprising the following steps: S1. Collecting original point cloud data of the tower and preprocessing the original point cloud data of the tower to obtain target point cloud data; S2. Perform cluster analysis on the target point cloud data based on the clustering algorithm, identify the point cloud data of each part of the tower and save them to obtain point cloud clusters; S3. Analyze the point cloud data in the point cloud cluster based on the central axis fitting algorithm and fit the central axis of the tower. Determine the position and height of the tower body based on the central axis. Perform cross-sectional analysis on the tower body to determine the structural characteristics of the tower. S4. Based on the central axis and the structural characteristics of the tower, coordinate alignment and splicing of the point cloud data of each part of the tower are performed to obtain the original three-dimensional point cloud model of the tower; S5. Compare and analyze the original tower 3D point cloud model based on the ideal tower model data, obtain error point cloud data based on the comparison results, correct the error point cloud data to obtain the target tower 3D point cloud model; monitor the tower in real time based on the target tower 3D point cloud model to evaluate the health status of the tower.

[0006] In this solution, clustering algorithms (such as DBSCAN) are used to automatically identify point cloud data of various parts of the tower (such as the tower body, crossarms, lines, etc.), reducing manual intervention. This allows the point cloud data to be quickly and accurately divided into different clusters, providing a clear data basis for subsequent analysis. The central axis fitting algorithm can accurately determine the main position and height of the tower, providing a benchmark for the geometric analysis of the tower. Through cross-sectional analysis, the thickness changes, inclination, curvature and other structural characteristics of the tower can be accurately obtained, providing key data for the health assessment of the tower. In addition, due to the existence of the central axis, a calibration benchmark is provided for the subsequent construction of the three-dimensional point cloud model of the tower, so that there will be no deviation in the combination of various parts during model construction, making the three-dimensional model closer to the actual equipment, thereby improving the accuracy of model construction.

[0007] Preferably, in S1, collecting original point cloud data of the tower and preprocessing the original point cloud data of the tower to obtain target point cloud data include the following steps: A drone equipped with a laser scanning device is used to perform all-round laser scanning on the tower to obtain the original point cloud data of the tower. The original point cloud data of the tower is then subjected to environmental filtering based on a statistical filtering algorithm to obtain the target point cloud data.

[0008] In this solution, the original point cloud data of the tower is subjected to environmental filtering through a statistical filtering algorithm, such as noise points generated by environmental factors such as flying birds and swaying leaves. Therefore, when performing laser scanning on the tower, there is no need to worry about the impact of interference factors in the environment on the accuracy of tower detection, thereby improving the scanning efficiency of the tower and providing an accurate data foundation for the subsequent construction of the tower's three-dimensional point cloud model.

[0009] Preferably, in S2, cluster analysis is performed on the target point cloud data based on a clustering algorithm, point cloud data of various parts of the tower are identified and saved to obtain point cloud clusters, including the following steps: Use density-based clustering algorithm to perform cluster analysis on target point cloud data, set neighborhood radius and minimum number of points; Traverse each point in the target point cloud data and take the point to be detected as the target point. If the target point has not been visited, all points within the neighborhood radius of the query target point are recorded as the target cluster; If the number of points in the target cluster is greater than or equal to the minimum number of points, a new cluster is created. All points in the target cluster are recursively added to the new cluster, and the points within the neighborhood radius of all unvisited points in the target cluster are continued to be queried until the new cluster cannot be expanded. Then a new cluster is constructed and the above steps are repeated until all points in the target point cloud data are visited to obtain a point cloud cluster.

[0010] In this solution, the density-based clustering algorithm can automatically identify clusters in point cloud data without pre-specifying the number of each part of the tower. It can automatically traverse the point cloud data and recursively expand the clusters, and can quickly divide the point cloud data into different parts such as the tower body, crossarms, and lines, which can significantly improve data processing efficiency. The density-based clustering algorithm can identify clusters of any shape and can flexibly set parameters according to the characteristics of the tower point cloud data, so that it can process tower structures of different sizes, shapes and densities. It is very suitable for processing complex geometric structures (such as crossarms, lines, etc.) in the tower point cloud data. At the same time, it also clearly separates the different parts of the tower, providing a clear data basis for subsequent detection of the health status of the tower.

[0011] Preferably, in S2, the point cloud cluster includes a tower body point cloud cluster, a cross arm point cloud cluster and a line point cloud cluster.

[0012] In this solution, since the main parts of the tower include the tower body, crossarms, and lines, when modeling and testing the health status of the tower, only these three parts need to be tested to determine the overall condition of the tower. Therefore, corresponding point cloud clusters are generated for these three parts.

[0013] Preferably, in S3, the point cloud data in the tower body point cloud cluster is analyzed based on the central axis fitting algorithm to fit the central axis of the tower, the position and height of the tower body are determined based on the central axis, and the cross-section analysis of the tower body is performed to determine the structural characteristics of the tower, including the following steps: The point cloud data in the tower main body point cloud cluster is divided into several segments according to the height direction, and the point cloud data within each height range is marked to obtain segmented point cloud data; Project the segmented point cloud data onto a plane to obtain segmented sections and calculate the center points of the sections. Fit all the center points of the sections into a spatial curve to obtain the center axis equation of the tower. The main position of the tower is obtained based on the central axis equation, the tower height is obtained by integrating the central axis equation, and the structural characteristics of the tower are obtained based on the shape and side length of the segmented section.

[0014] In this solution, the central axis is fitted through the cross-section center point of the segmented point cloud data, which can accurately reflect the geometric shape and spatial position of the tower. It is used to deal with complex geometric features such as the bending and tilt of the tower. Based on the central axis equation, the main position of the tower (such as the lower and upper ends) can be accurately determined, so that the height of the tower can be accurately calculated, providing a benchmark for subsequent analysis. The geometric parameters of the tower are determined by the central axis, which can provide accurate tower geometric parameters (such as height and position) for power inspections, ensuring rapid analysis and evaluation of inspection data during inspections. Through cross-sectional analysis, deformation, tilt and other problems of the tower can be discovered in a timely manner, which can avoid potential safety hazards.

[0015] Preferably, in S4, coordinate alignment and splicing of point cloud data of various parts of the tower are performed based on the central axis and the structural characteristics of the tower to obtain an original three-dimensional point cloud model of the tower, including the following steps: The geometric center of the segmented point cloud data is aligned to the central axis, and the aligned segmented point cloud data is spliced ​​based on the height order to obtain the original tower 3D point cloud model.

[0016] In this solution, by aligning the geometric center of the segmented point cloud data to the central axis, the consistency of the spatial position of each segment of the point cloud data is ensured, the misalignment or deviation caused by segmentation processing is avoided, and the local error in the point cloud data is eliminated, laying the foundation for building a high-precision three-dimensional point cloud model; based on the highly sequentially spliced ​​and aligned segmented point cloud data, a complete three-dimensional point cloud model of the tower can be constructed, which fully reflects the geometric shape and structural characteristics of the tower and improves the accuracy of tower detection.

[0017] Preferably, in S5, a comparative analysis is performed on the original tower 3D point cloud model based on the ideal tower model data, error point cloud data is obtained based on the comparison result, and the error point cloud data is corrected to obtain the target tower 3D point cloud model, including the following steps: The ideal tower model data is compared with the point cloud data in the original tower 3D point cloud model. If there are error points whose deviation from the ideal tower model data exceeds the set threshold, the error points are corrected and optimized by interpolation and local fitting to obtain the target tower 3D point cloud model.

[0018] In this scheme, by comparing with the ideal tower model data, the error points in the original point cloud data that deviate greatly from the ideal model can be accurately identified. The error points are corrected by the interpolation method, which can effectively fill the missing areas in the point cloud data and improve the integrity of the data. The error points are then optimized through the local fitting method, which can smooth the irregular areas in the point cloud data and improve the geometric accuracy of the model.

[0019] Preferably, in S5, real-time monitoring of the tower and evaluating the health status of the tower based on the three-dimensional point cloud model of the target tower include the following steps: Real-time detection of the angle between the central axis of the target tower's 3D point cloud model and the vertical direction to obtain the tower's offset angle. If the tower's offset angle exceeds the tilt threshold, the tower is judged to be in a tilted state and an alarm signal is issued; The curvature of the central axis of the target tower's 3D point cloud model is detected in real time. If the curvature exceeds the bending threshold, the tower is judged to be in a bent state and an alarm signal is issued.

[0020] In this solution, based on the 3D point cloud model of the target tower, the offset angle and curvature of the central axis can be accurately calculated, and intuitive visualization results can be generated, allowing relevant personnel to quickly understand the health status of the tower. The health status of the tower can be automatically determined through the offset angle and curvature without the need for human intervention, reducing the workload of manual inspections and improving monitoring efficiency. Moreover, through the detection of both tilt and bending dimensions, the health status of the tower can be comprehensively assessed, avoiding the limitations of a single indicator.

[0021] In a second aspect, a technical solution provided in an embodiment of the present invention is: a tower detection data optimization and analysis system based on a point cloud model, comprising a data acquisition and preprocessing module, a tower feature extraction and model building module, a data optimization and intelligent analysis module, and a data storage module; The data acquisition and preprocessing module acquires the original point cloud data of the tower, and preprocesses the original point cloud data of the tower to obtain the target point cloud data; The tower feature extraction and model building module performs cluster analysis on the target point cloud data based on the clustering algorithm, identifies the point cloud data of each part of the tower and saves it to the data storage module to obtain a point cloud cluster, and analyzes the point cloud data in the point cloud cluster based on the central axis fitting algorithm and fits the central axis of the tower, determines the position and height of the tower body based on the central axis, performs cross-sectional analysis on the tower body to determine the structural characteristics of the tower, and aligns and splices the point cloud data of each part of the tower based on the central axis and the structural characteristics of the tower to obtain the original three-dimensional point cloud model of the tower; The data optimization and intelligent analysis module compares and analyzes the original tower three-dimensional point cloud model based on the ideal tower model data, obtains error point cloud data based on the comparison results, corrects the error point cloud data to obtain the target tower three-dimensional point cloud model; and monitors the tower in real time based on the target tower three-dimensional point cloud model to evaluate the health status of the tower.

[0022] In this solution, by designing a system for automatic processing of original tower point cloud data, the integration, optimization and intelligent analysis of the original point cloud data are realized. During the monitoring process of the tower, the impact of environmental factors on the monitoring results is reduced, and the status of the tower is visualized in three dimensions, which facilitates the staff to have an intuitive understanding of the tower status. In addition, the detection of the tower status is judged by the central axis without the need for human intervention, which reduces the workload of inspection personnel and improves inspection efficiency.

[0023] Preferably, the data storage module stores tower point cloud data, tower three-dimensional point cloud model data and tower health assessment data, and classifies and stores the above data respectively using a hierarchical storage structure; when storing new data of the same category, the old data is automatically updated.

[0024] This solution builds a dedicated database to store tower point cloud data, model data, and test analysis results. A hierarchical storage structure is employed to categorize and store data from different towers, facilitating data query, access, and management. A data update mechanism is also developed to automatically update the database with new test data, ensuring data timeliness and integrity. Data backup and recovery capabilities are also provided to prevent data loss and ensure stable system operation.

[0025] The beneficial effects of the present invention are as follows: (1) The present invention automatically identifies point cloud data of various parts of the tower (such as the tower body, crossarms, lines, etc.) based on a clustering algorithm (such as DBSCAN), reducing manual intervention, thereby being able to quickly and accurately divide the point cloud data into different clusters, providing a clear data basis for subsequent analysis; (2) The present invention can accurately determine the main position and height of the tower through the central axis fitting algorithm, providing a benchmark for the geometric analysis of the tower. Then, through cross-sectional analysis, it can accurately obtain the thickness change, inclination, curvature and other structural characteristics of the tower, providing key data for the health assessment of the tower; (3) Due to the existence of the central axis, the present invention provides a calibration benchmark for constructing a three-dimensional point cloud model of the tower, so that there will be no deviation in the combination of various parts during model construction, making the three-dimensional model closer to the actual equipment, thereby improving the accuracy of model construction.

[0026] The above content of the invention is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other features, objects, and advantages of the present invention will become more apparent upon reading the detailed description of the non-limiting embodiments made with reference to the following drawings. The drawings are for the purpose of illustrating preferred embodiments only and are not to be construed as limiting the present invention. Like reference characters are used throughout the drawings to designate like parts.

[0028] Figure 1 This is a flow chart of the tower detection data optimization analysis method based on the point cloud model of the present invention; Figure 2 This is a block diagram of the tower detection data optimization and analysis system based on the point cloud model of the present invention. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0030] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be performed in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operations are completed, but can also have additional steps not included in the figures; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0031] Example 1: Figure 1 As shown, in order to solve the problem that the existing point cloud-based tower detection method is difficult to perform comprehensive analysis of all parts of the tower, resulting in low accuracy of tower detection results, this embodiment provides a tower detection data optimization and analysis method based on a point cloud model, including the following steps: S1: Collect the original point cloud data of the tower and pre-process the original point cloud data of the tower to obtain the target point cloud data.

[0032] In this embodiment, collecting the original point cloud data of the tower and preprocessing the original point cloud data of the tower to obtain target point cloud data include the following steps: A drone equipped with a laser scanning device is used to perform all-round laser scanning on the tower to obtain the original point cloud data of the tower. The original point cloud data of the tower is then subjected to environmental filtering based on a statistical filtering algorithm to obtain the target point cloud data.

[0033] Specifically, the specific process of environmental filtering of the original point cloud data of the tower is as follows: Define the original point cloud data of the tower as P = {p1,p2,...,p n}, where each point p i All contain three-dimensional coordinate information (x i ,y i ,z i ), pre-set the neighborhood radius r, the number of neighborhood points k and the standard deviation multiple α based on the needs. The selection of the neighborhood radius is related to the point cloud density. If the point cloud density is high, a smaller radius can be selected. If the point cloud density is low, a larger radius can be selected.

[0034] For each point p in the point cloud i , with point p i As the center, find all points within its neighborhood radius r, or find its nearest k neighborhood points, and calculate point p i The average distance d to its neighboring points i , the specific calculation formula is as follows: where p j For point p i Neighborhood points, ||p i -p j || represents the Euclidean distance between two points.

[0035] The mean and standard deviation of the average distance from all points to neighboring points in the original point cloud data of the tower are calculated. The formula is as follows: Based on the standard deviation multiple α, set the outlier judgment threshold d threshold =μ+α·σ; for each point p i , if the average distance d i >d threshold , then the point is considered to be an outlier and removed from the point cloud data. Otherwise, the point is retained and all points in the original point cloud data of the tower are judged to obtain the target tower point cloud data P'={p1',p2',...,p m '}, where m <n。

[0036] This embodiment uses a statistical filtering algorithm to perform environmental filtering on the original point cloud data of the tower, such as noise points generated by environmental factors such as flying birds and swaying leaves. Therefore, when performing laser scanning on the tower, there is no need to worry about the impact of interference factors in the environment on the accuracy of tower detection, thereby improving the scanning efficiency of the tower and providing an accurate data foundation for the subsequent construction of a three-dimensional point cloud model of the tower.

[0037] S2: Perform cluster analysis on the target point cloud data based on the clustering algorithm, identify the point cloud data of each part of the tower and save them to obtain point cloud clusters.

[0038] In this embodiment, cluster analysis is performed on the target point cloud data based on a clustering algorithm to identify the point cloud data of each part of the tower and save them to obtain point cloud clusters, including the following steps: Using density-based clustering algorithm, the target point cloud data P'={p1',p2',...,p m '} Perform cluster analysis, set the neighborhood radius r and the minimum number of points Pts min ; Traverse each point p in the target point cloud data i ', the point to be detected p i 'As the target point, if p i 'has been visited, skip this point and continue to process the next point. If the target point p i 'If it is not visited, all points within the neighborhood radius r of the query target point are recorded as the target cluster N(p i '); If the target cluster N(p i ') is less than the minimum number of points Pts min Then p i 'As a noise point; If the target cluster N(p i ') is greater than or equal to the minimum number of points, a new cluster C is created, all points in the target cluster are recursively added to the new cluster C, and the target cluster N(p i ') until the new cluster C cannot be expanded. Otherwise, a new cluster is constructed and the above steps are repeated until all points in the target point cloud data are visited to obtain point cloud clusters C1, C2, C3...C m .

[0039] This embodiment uses a density-based clustering algorithm to automatically identify clusters in point cloud data without pre-specifying the number of each part of the tower. It can automatically traverse the point cloud data and recursively expand the clusters, and can quickly divide the point cloud data into different parts such as the tower body, crossarms, and lines, which can significantly improve data processing efficiency. The density-based clustering algorithm can also identify clusters of any shape and can flexibly set parameters according to the characteristics of the tower point cloud data, so that it can process tower structures of different sizes, shapes and densities. It is very suitable for processing complex geometric structures (such as crossarms, lines, etc.) in the tower point cloud data. At the same time, it also clearly separates different parts of the tower, providing a clear data basis for subsequent detection of the health status of the tower.

[0040] In this embodiment, the point cloud cluster includes a tower body point cloud cluster, a cross arm point cloud cluster, and a line point cloud cluster.

[0041] In this embodiment, since the main parts of the tower include the tower body, crossarms and lines, when modeling and detecting the health status of the tower, only these three parts need to be detected to determine the overall condition of the tower. Therefore, corresponding point cloud clusters are generated for these three parts.

[0042] S3: Analyze the point cloud data in the point cloud cluster based on the central axis fitting algorithm and fit the central axis of the tower. Determine the position and height of the tower body based on the central axis. Perform cross-sectional analysis on the tower body to determine the structural characteristics of the tower.

[0043] In this embodiment, the point cloud data in the tower body point cloud cluster is analyzed based on the central axis fitting algorithm to fit the central axis of the tower body, the position and height of the tower body are determined based on the central axis, and the cross-section analysis of the tower body is performed to determine the structural characteristics of the tower, including the following steps: The tower main point cloud cluster C i The point cloud data in the image is divided into several segments according to the height direction, and the point cloud data within each height range is marked to obtain segmented point cloud data, which is recorded as C i ={C i1 ,C i2 ,...,C in}; Project each segment of the segmented point cloud data onto a plane to obtain a segmented cross section and calculate its cross section center point c i =(x i ,y i ,z i ), the specific calculation formula is as follows: All the cross-section center points {c1,c2,...,c k}And use the least squares method to fit it into a spatial curve to obtain the center axis equation of the tower L(t) = (x(t), y(t), z(t)), where t is a parameter; The main position of the tower is obtained based on the central axis equation, and the tower height is obtained by integrating the central axis equation. The specific calculation formula is as follows: If the central axis is a straight line, the tower height can be directly calculated as the distance between the two end points.

[0044] The structural characteristics of the tower are obtained based on the shape and side length of the segmented section.

[0045] This embodiment fits the central axis through the cross-section center points of the segmented point cloud data, which can accurately reflect the geometric shape and spatial position of the tower, and is used to handle complex geometric features such as the bending and tilt of the tower. Based on the central axis equation, the main position of the tower (such as the lower and upper ends) can be accurately determined, thereby accurately calculating the height of the tower and providing a benchmark for subsequent analysis; the geometric parameters of the tower are determined by the central axis, so that accurate tower geometric parameters (such as height and position) can be provided for power inspections, ensuring rapid analysis and evaluation of inspection data during inspections. Through cross-section analysis, deformation, tilt and other problems of the tower can be discovered in a timely manner, and potential safety hazards can be avoided.

[0046] S4: Based on the structural characteristics of the central axis and the tower, the point cloud data of each part of the tower are aligned and spliced ​​to obtain the original tower 3D point cloud model.

[0047] In this embodiment, coordinate alignment and splicing of point cloud data of various parts of the tower are performed based on the central axis and the structural characteristics of the tower to obtain an original three-dimensional point cloud model of the tower, including the following steps: The geometric center of the segmented point cloud data is aligned to the central axis, and the aligned segmented point cloud data is spliced ​​based on the height order to obtain the original tower 3D point cloud model.

[0048] This embodiment aligns the geometric centers of the segmented point cloud data to the central axis to ensure the consistency of the spatial position of each segment of point cloud data, avoids misalignment or deviation caused by segmentation processing, eliminates local errors in the point cloud data, and lays the foundation for building a high-precision three-dimensional point cloud model. Based on the highly sequentially spliced ​​and aligned segmented point cloud data, a complete three-dimensional point cloud model of the tower can be constructed, which fully reflects the geometric shape and structural characteristics of the tower and improves the accuracy of tower detection.

[0049] S5: Compare and analyze the original tower 3D point cloud model based on the ideal tower model data, obtain error point cloud data based on the comparison results, correct the error point cloud data to obtain the target tower 3D point cloud model; monitor the tower in real time based on the target tower 3D point cloud model to evaluate the health status of the tower.

[0050] In this embodiment, a comparative analysis is performed on the original tower 3D point cloud model based on the ideal tower model data, error point cloud data is obtained based on the comparison result, and the error point cloud data is corrected to obtain the target tower 3D point cloud model, including the following steps: The ideal tower model data is compared with the point cloud data in the original tower 3D point cloud model. If there are error points whose deviation from the ideal tower model data exceeds the set threshold, the error points are corrected and optimized by interpolation and local fitting to obtain the target tower 3D point cloud model.

[0051] By comparing the data with the ideal tower model, this embodiment can accurately identify the error points in the original point cloud data that deviate greatly from the ideal model. The error points are corrected by the interpolation method, which can effectively fill the missing areas in the point cloud data and improve the integrity of the data. The error points are then optimized by the local fitting method, which can smooth the irregular areas in the point cloud data and improve the geometric accuracy of the model.

[0052] In this embodiment, the health status of the tower is evaluated by real-time monitoring of the tower based on the three-dimensional point cloud model of the target tower, including the following steps: Real-time detection of the angle between the central axis of the target tower's 3D point cloud model and the vertical direction to obtain the tower's offset angle. If the tower's offset angle exceeds the tilt threshold, the tower is judged to be in a tilted state and an alarm signal is issued; The curvature of the central axis of the target tower's 3D point cloud model is detected in real time. If the curvature exceeds the bending threshold, the tower is judged to be in a bent state and an alarm signal is issued.

[0053] This embodiment is based on the three-dimensional point cloud model of the target tower, and can accurately calculate the offset angle and curvature of the central axis, and can also generate intuitive visualization results, so that relevant personnel can quickly understand the health status of the tower. The health status of the tower can be automatically judged by the offset angle and curvature without human intervention, reducing the workload of manual inspections and improving monitoring efficiency. Moreover, by detecting the two dimensions of tilt and bending, the health status of the tower can be comprehensively evaluated, avoiding the limitations of a single indicator.

[0054] Example 2: Figure 2 As shown, this embodiment provides a tower detection data optimization and analysis system based on a point cloud model, including a data acquisition and preprocessing module, a tower feature extraction and model building module, a data optimization and intelligent analysis module, and a data storage module; The data acquisition and preprocessing module acquires the original point cloud data of the tower, and preprocesses the original point cloud data of the tower to obtain the target point cloud data; The tower feature extraction and model building module performs cluster analysis on the target point cloud data based on the clustering algorithm, identifies the point cloud data of each part of the tower and saves it to the data storage module to obtain a point cloud cluster, and analyzes the point cloud data in the point cloud cluster based on the central axis fitting algorithm and fits the central axis of the tower, determines the position and height of the tower body based on the central axis, performs cross-sectional analysis on the tower body to determine the structural characteristics of the tower, and aligns and splices the point cloud data of each part of the tower based on the central axis and the structural characteristics of the tower to obtain the original three-dimensional point cloud model of the tower; The data optimization and intelligent analysis module compares and analyzes the original tower three-dimensional point cloud model based on the ideal tower model data, obtains error point cloud data based on the comparison results, corrects the error point cloud data to obtain the target tower three-dimensional point cloud model; and monitors the tower in real time based on the target tower three-dimensional point cloud model to evaluate the health status of the tower.

[0055] This embodiment designs a system for automatically processing the original tower point cloud data, thereby realizing the integration, optimization and intelligent analysis of the original point cloud data. In the process of monitoring the tower, the influence of environmental factors on the monitoring results is reduced, and the status of the tower is visualized in three dimensions, which facilitates the staff to have an intuitive understanding of the tower status. In addition, the detection of the tower status is judged by the central axis without the need for human intervention, which reduces the workload of the inspection personnel and improves the inspection efficiency.

[0056] In this embodiment, the data storage module stores tower point cloud data, tower three-dimensional point cloud model data and tower health assessment data, and uses a hierarchical storage structure to classify and store the above data respectively; when storing new data of the same category, the old data is automatically updated.

[0057] This embodiment builds a dedicated database to store tower point cloud data, model data, and inspection and analysis results. It adopts a hierarchical storage structure, categorizing and storing data from different towers to facilitate data query, access, and management. A data update mechanism is also developed to automatically update the relevant information in the database when new inspection data is generated, ensuring data timeliness and integrity. Furthermore, data backup and recovery functions are provided to prevent data loss and ensure stable system operation.

[0058] It can be seen from the above embodiments that at least the following substantial effects are achieved: (1) The present invention automatically identifies point cloud data of various parts of the tower (such as the tower body, crossarms, lines, etc.) based on a clustering algorithm (such as DBSCAN), reducing manual intervention, thereby quickly and accurately dividing the point cloud data into different clusters, providing a clear data basis for subsequent analysis; (2) The present invention can accurately determine the main position and height of the tower through the central axis fitting algorithm, providing a benchmark for the geometric analysis of the tower. Then, through cross-sectional analysis, it can accurately obtain the thickness change, inclination, curvature and other structural characteristics of the tower, providing key data for the health assessment of the tower; (3) Due to the existence of the central axis, the present invention provides a calibration benchmark for constructing a three-dimensional point cloud model of the tower, so that there will be no deviation in the combination of various parts during model construction, making the three-dimensional model closer to the actual equipment, thereby improving the accuracy of model construction.

[0059] The specific implementation described above is a preferred implementation of the tower detection data optimization analysis method and system based on the point cloud model of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation. Any equivalent changes made in accordance with the shape and structure of the present invention are within the scope of protection of the present invention.

Claims

1. A tower inspection data optimization and analysis method based on a point cloud model, characterized by: The following steps are involved: S1. Collecting original point cloud data of the tower and preprocessing the original point cloud data of the tower to obtain target point cloud data; S2. Perform cluster analysis on the target point cloud data based on the clustering algorithm, identify the point cloud data of each part of the tower and save them to obtain point cloud clusters; S3. Analyze the point cloud data in the point cloud cluster based on the central axis fitting algorithm and fit the central axis of the tower. Determine the position and height of the tower body based on the central axis. Perform cross-sectional analysis on the tower body to determine the structural characteristics of the tower. S4. Based on the central axis and the structural characteristics of the tower, coordinate alignment and splicing of the point cloud data of each part of the tower are performed to obtain the original three-dimensional point cloud model of the tower; S5. Comparing and analyzing the original tower 3D point cloud model based on the ideal tower model data, obtaining error point cloud data based on the comparison results, and correcting the error point cloud data to obtain the target tower 3D point cloud model; The health status of the tower is evaluated by real-time monitoring of the tower based on the three-dimensional point cloud model of the target tower.

2. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 1, wherein: In S1, the original point cloud data of the tower is collected and pre-processed to obtain the target point cloud data, including the following steps: A drone equipped with a laser scanning device is used to perform all-round laser scanning on the tower to obtain the original point cloud data of the tower. The original point cloud data of the tower is then subjected to environmental filtering based on a statistical filtering algorithm to obtain the target point cloud data.

3. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 1, wherein: In S2, cluster analysis is performed on the target point cloud data based on the clustering algorithm, the point cloud data of each part of the tower is identified and saved to obtain point cloud clusters, including the following steps: Use density-based clustering algorithm to perform cluster analysis on target point cloud data, set neighborhood radius and minimum number of points; Traverse each point in the target point cloud data and take the point to be detected as the target point. If the target point has not been visited, all points within the neighborhood radius of the query target point are recorded as the target cluster; If the number of points in the target cluster is greater than or equal to the minimum number of points, a new cluster is created. All points in the target cluster are recursively added to the new cluster, and the points within the neighborhood radius of all unvisited points in the target cluster are continued to be queried until the new cluster cannot be expanded. Then a new cluster is constructed and the above steps are repeated until all points in the target point cloud data are visited to obtain a point cloud cluster.

4. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 1, wherein: In S2, the point cloud cluster includes a tower body point cloud cluster, a cross arm point cloud cluster, and a line point cloud cluster.

5. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 4, wherein: In S3, the point cloud data in the tower main body point cloud cluster is analyzed based on the central axis fitting algorithm to fit the central axis of the tower. The position and height of the tower main body are determined based on the central axis. The cross-section analysis of the tower main body is performed to determine the structural characteristics of the tower, including the following steps: The point cloud data in the tower main body point cloud cluster is divided into several segments according to the height direction, and the point cloud data within each height range is marked to obtain segmented point cloud data; Project the segmented point cloud data onto a plane to obtain segmented sections and calculate the center points of the sections. Fit all the center points of the sections into a spatial curve to obtain the center axis equation of the tower. The main position of the tower is obtained based on the central axis equation, the tower height is obtained by integrating the central axis equation, and the structural characteristics of the tower are obtained based on the shape and side length of the segmented section.

6. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 5, characterized in that: In S4, based on the central axis and the structural characteristics of the tower, the point cloud data of each part of the tower is aligned and spliced ​​to obtain the original tower 3D point cloud model, which includes the following steps: The geometric center of the segmented point cloud data is aligned to the central axis, and the aligned segmented point cloud data is spliced ​​based on the height order to obtain the original tower 3D point cloud model.

7. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 1, wherein: In S5, the original tower 3D point cloud model is compared and analyzed based on the ideal tower model data, error point cloud data is obtained based on the comparison result, and the error point cloud data is corrected to obtain the target tower 3D point cloud model, including the following steps: The ideal tower model data is compared with the point cloud data in the original tower 3D point cloud model. If there are error points whose deviation from the ideal tower model data exceeds the set threshold, the error points are corrected and optimized by interpolation and local fitting to obtain the target tower 3D point cloud model.

8. The method for optimizing and analyzing tower inspection data based on a point cloud model according to claim 1, wherein: In S5, the health status of the tower is evaluated by real-time monitoring of the tower based on the 3D point cloud model of the target tower, including the following steps: Real-time detection of the angle between the central axis of the target tower's 3D point cloud model and the vertical direction to obtain the tower's offset angle. If the tower's offset angle exceeds the tilt threshold, the tower is judged to be in a tilted state and an alarm signal is issued; The curvature of the central axis of the target tower's 3D point cloud model is detected in real time. If the curvature exceeds the bending threshold, the tower is judged to be in a bent state and an alarm signal is issued.

9. A tower inspection data optimization and analysis system based on a point cloud model, applicable to the tower inspection data optimization and analysis method based on a point cloud model according to any one of claims 1 to 8, characterized in that: It includes data acquisition and preprocessing module, tower feature extraction and model building module, data optimization and intelligent analysis module and data storage module; The data acquisition and preprocessing module acquires the original point cloud data of the tower, and preprocesses the original point cloud data of the tower to obtain the target point cloud data; The tower feature extraction and model building module performs cluster analysis on the target point cloud data based on the clustering algorithm, identifies the point cloud data of each part of the tower and saves it to the data storage module to obtain a point cloud cluster, and analyzes the point cloud data in the point cloud cluster based on the central axis fitting algorithm and fits the central axis of the tower, determines the position and height of the tower body based on the central axis, performs cross-sectional analysis on the tower body to determine the structural characteristics of the tower, and aligns and splices the point cloud data of each part of the tower based on the central axis and the structural characteristics of the tower to obtain the original three-dimensional point cloud model of the tower; The data optimization and intelligent analysis module compares and analyzes the original tower 3D point cloud model based on the ideal tower model data, obtains error point cloud data based on the comparison result, and corrects the error point cloud data to obtain the target tower 3D point cloud model; The health status of the tower is evaluated by real-time monitoring of the tower based on the three-dimensional point cloud model of the target tower.

10. The point cloud model-based tower inspection data optimization and analysis system according to claim 9, characterized in that: The data storage module stores tower point cloud data, tower three-dimensional point cloud model data and tower health status assessment data, and adopts a hierarchical storage structure to classify and store the above data respectively; when storing new data of the same category, the old data is automatically updated.

Images