Three-dimensional point cloud feature extraction based on random sample consistency

Abstract

The invention discloses a method for extracting three-dimensional point cloud data features based on random sample consistency and classifying and labeling extracted targets. The method comprises the following steps: S1, data preprocessing; s2, fine treatment; s3, feature extraction; S4, secondary retrieval; s5, removing the wall body; and S6, classifying labels. According to the method, the extraction precision of the target object is greatly improved while the feature extraction of the target object is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of point cloud feature extraction, and more specifically, relates to a method for extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling extracted targets. Background Art

[0002] Point cloud data is a data representation method that describes the surface of an object in three-dimensional space. It is widely used in computer vision, geographic information science, remote sensing technology and other fields. Point cloud segmentation is one of the key technologies for processing point cloud data, which aims to divide point cloud data into multiple parts with different attributes or semantic information. In recent years, point cloud component segmentation has played a vital role in various fields, such as autonomous driving, machine vision and equipment maintenance. Especially in the field of equipment maintenance, extracting the various components of the equipment from large-scale disordered point cloud data has brought significant advantages for tasks such as fault diagnosis and defect identification. However, due to the complex spatial structure and noise interference of point cloud data, effective component segmentation poses a great challenge. For point cloud data in complex environments, how to accurately extract the various components of the equipment from large-scale disordered point cloud data is an important basis for subsequent power equipment fault diagnosis and defect identification.

[0003] In recent years, many deep learning models have emerged specifically for processing point cloud data, such as deep learning models based on RGB-D images, voxels, and other representation elements. In terms of voxel-based deep learning models, there are 3D ShapeNets and VoxNet, which first convert point cloud data into voxel grids and then use 3D convolutional networks for processing, thereby guiding voxel-based three-dimensional semantic segmentation. This method may lose some information in the process of extracting spatial features, but it is more efficient for processing large-scale uniformly distributed point cloud data. In addition, there is a method that convolves the point cloud by defining a learnable geometric kernel, but its computational complexity is high and it is difficult to meet real-time requirements.

[0004] The application of random sample consistency algorithm in processing distribution site cloud data is particularly important. Random sample consistency is an iterative method, which is mainly used to fit mathematical models in data containing a lot of noise. Its advantages include strong robustness, wide applicability and simple implementation. Random sample consistency can process data containing a lot of noise and outliers, automatically ignore outliers, and ensure the accuracy of model fitting; it is suitable for various types of model fitting, not limited to specific geometric shapes; and its implementation is relatively simple and computationally efficient. Especially in the processing of distribution site cloud data, the random sample consistency algorithm can effectively extract the geometric features of key equipment such as transformers and circuit breakers from complex and noisy point cloud data, thereby realizing high-precision component segmentation. Compared with the traditional random sample consistency algorithm, by adding pass-through filtering and point cloud cropping, the accuracy of equipment fault diagnosis and defect identification is further improved, and the scheme of the present invention is proposed. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the present invention provides a method for extracting features of three-dimensional point cloud data based on random sample consistency and classifying and labeling the extracted targets. While realizing the feature extraction of the target object, the accuracy of target object extraction is greatly improved, which is beneficial to the accuracy of equipment fault diagnosis and defect identification.

[0006] The technical solution to achieve the purpose of the present invention is to provide a method for extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling the extracted targets, comprising the following steps:

[0007] S1. Data preprocessing: Determine the xyz coordinates of the extracted target in the point cloud processing software using the initial point cloud data, use the straight-through filtering method, and set the corresponding xyz filtering threshold;

[0008] S2. Refined processing: the filtered point cloud data is further filtered through the normal vector to filter out the noise points at the edge of the target object;

[0009] S3. Feature extraction: Set the maximum number of iterations and the maximum distance from the point to the model, select the cylindrical model according to the characteristics of the target object, use the random sample consistency algorithm to fit the model, and perform feature extraction;

[0010] S4. Secondary retrieval: comparing the extracted point cloud data with the original point cloud data, and performing secondary retrieval;

[0011] S5. Eliminate the wall; after extracting the cylindrical features,

[0012] Use KD-Tree clustering index to remove objects other than the target object;

[0013] S6. Classification label; according to the type of point cloud in the point cloud data, simply distinguish it into target objects and non-target objects, and give the right classification label; train the global model through the point cloud data and classification labels until convergence, and obtain the pre-built global extraction model.

[0014] Furthermore, the specific steps of the method for determining the through filtering threshold in step S1 are as follows:

[0015] S11. Preliminary analysis and determination of target area: Use point cloud processing software to conduct preliminary visual analysis of point cloud data to determine the approximate location and range of the target object;

[0016] Determine the approximate coordinate range of the target object in the three directions of xyz through observation and interactive operation;

[0017] S12. Perform statistical analysis on the point cloud data of the target area and calculate the point distribution density of the target object in the three directions of xyz;

[0018] According to the statistical results and combined with experience, set the preliminary filtering threshold;

[0019] S13. Iterative adjustment: Apply the initially set threshold to perform straight-through filtering on the point cloud data, and observe whether the filtered point cloud data contains the target object and does not have too much interference data;

[0020] If the filtering result is not ideal, adjust the threshold value for multiple iterations to gradually approach the optimal threshold value; after each adjustment, recheck the filtered point cloud data until the expected effect is achieved;

[0021] S14. Optimization based on feature extraction; when extracting the features of the target object, if some features are found to be weak or unclear, further adjust the filter threshold to enhance the performance of the features.

[0022] Furthermore, S1 also includes the following steps:

[0023] S15.Automation optimization;

[0024] When batch processing a large amount of point cloud data, consider using an automated optimization algorithm to automatically adjust the threshold to obtain the best filtering effect.

[0025] Furthermore, S2 specifically includes the following steps:

[0026] Normal vector estimation: Normal vector estimation based on local surface fitting method: For each scan point p in the point cloud, search for the K neighboring points closest to the scan point p, and calculate the local plane P in the least squares sense of the neighboring points. The local plane P is expressed as:

[0027]

[0028] In the formula, is the normal vector of plane P, d is the distance from P to the origin of the coordinate system;

[0029] The normal of the plane fitted by K adjacent points is the normal vector of the current scanning point; the normal vector of plane P is obtained by principal component analysis. It is known from the calculation that P passes through the centroid P0 of K adjacent points, and the normal vector satisfy First, perform eigenvalue decomposition on the covariance matrix M in formula (5) to obtain the eigenvalues ​​of M and the normal vector of the eigenvector P corresponding to the minimum eigenvalue of M.

[0030]

[0031] Normal vector orientation;

[0032] Assuming that the point cloud is dense enough and the sampling plane is smooth everywhere, the normal vectors of two adjacent points will be close to parallel; let For two adjacent points x i , the normal vector of xj, if the direction of the normal vector is consistent, then If the inner product is negative, it means that the normal vector of one of the points needs to be flipped; therefore, first set a normal vector direction for a point in the point cloud, and then traverse all other points. If the normal vector of the current point is set to is the next point to be traversed. If Then flip nj, otherwise it remains unchanged;

[0033] Surface curvature;

[0034] Perform eigenvalue decomposition on the covariance matrix M of formula (5) to obtain the eigenvalues ​​of M. If the eigenvalues ​​satisfy λ0≤λ1≤λ2, the surface curvature of point P is:

[0035]

[0036] The smaller δ is, the flatter the area is, while the larger δ is, the more fluctuations in the neighborhood are.

[0037] Furthermore, the specific steps of S4 are:

[0038] S41. Direct filtering;

[0039] Create a pass filter object and set the point cloud data to be filtered;

[0040] Specify the filtering dimensions (X-axis, Y-axis, Z-axis), set the value range of each dimension by filtering threshold, and use 3D point cloud processing software to perform preliminary visual analysis on the point cloud data to determine the approximate location and range of the target object;

[0041] Through observation and interactive operation, determine the approximate coordinate range of the target object in the three directions of x, y and z. This can be done manually or by guessing based on the known size and position range of the object;

[0042] Traverse each point in the point cloud, determine whether the value of the point in the specified dimension is within the value range, and delete the points whose values ​​are not within the value range;

[0043] After the traversal is completed, the remaining points constitute the filtered point cloud;

[0044] S42. Point cloud clipping;

[0045] According to the extracted point cloud area, i.e., ROI area (use point cloud processing software to visualize the initial point cloud data and preliminarily identify the approximate location of the target object. Through interactive operations such as rotation, translation, and scaling, observe the point cloud from different angles to find the boundary or feature points of the target object). Separate the point cloud into regions; crop the point cloud according to the known ROI area; select the ROI area to be retained;

[0046] First, extract feature data to satisfy the following relationship:

[0047] x min ≤x≤x max

[0048] y min ≤y≤y max

[0049] z min ≤z≤z max

[0050] In the formula, x min , x max ,y min ,y max , z min , z max are the six boundary points of the rectangle;

[0051] S43. Point cloud combination;

[0052] Combine multiple point cloud data sets together and directly add the point cloud data that meets the features to another point cloud to form a more accurate point cloud data.

[0053] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of the method of extracting features of three-dimensional point cloud data based on random sample consistency and classifying and labeling the extracted targets of the present invention are implemented.

[0054] According to another aspect of the present invention, there is provided a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method of extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling extracted targets according to the present invention are implemented.

[0055] Compared with the prior art, the present invention has at least the following beneficial effects:

[0056] While realizing the feature extraction of the target object, the present invention greatly improves the accuracy of target object extraction, which is beneficial to the accuracy of equipment fault diagnosis and defect identification. By adopting the three-dimensional laser point cloud data processing technology based on voxel grid and cleverly constructing the KD-Tree structure, efficient organization and retrieval of scattered three-dimensional laser point cloud data are achieved. More importantly, the present invention incorporates a sophisticated threshold determination method into the data processing flow, especially in the secondary retrieval link, by accurately setting the threshold, the screening criteria for target object features are further refined, thereby significantly improving the accuracy of target object extraction. This innovation not only optimizes the data processing process, but also greatly enhances the accuracy of equipment fault diagnosis and defect identification, providing strong support for technical applications in related fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present invention, but are not intended to limit the present invention.

[0058] Figure 1 is a method flow chart of a preferred embodiment of the present invention;

[0059] Figure 2 is a point cloud schematic diagram of raw data in a preferred embodiment of the present invention;

[0060] Figure 3 is a schematic diagram of a point cloud after through-filter preprocessing in a preferred embodiment of the present invention;

[0061] Figure 4 is a schematic diagram of a point cloud after normal vector filtering in a preferred embodiment of the present invention;

[0062] Figure 5 It is a schematic diagram of a point cloud after feature extraction using a random sample consistency algorithm in a preferred embodiment of the present invention;

[0063] Figure 6 It is a schematic diagram of a point cloud after a secondary search with the original data in a preferred embodiment of the present invention;

[0064] Figure 7It is a schematic diagram of a point cloud after KD-tree clustering indexing in a preferred embodiment of the present invention;

[0065] Figure 8 It is a schematic diagram of a point cloud after classification labeling in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0066] To make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the embodiments of the present invention will be described clearly and completely in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0067] Unless otherwise defined, technical or scientific terms used herein shall have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs.

[0068] The following is combined with Figure 1-8 The specific embodiments of the present invention are described in detail with examples.

[0069] Embodiment 1: The present invention provides the following technical solution, a method for extracting features of three-dimensional point cloud data based on random sample consistency and classifying and labeling the extracted targets, which greatly improves the accuracy of target object extraction while realizing feature extraction of the target object, is beneficial to the accuracy of equipment fault diagnosis and defect identification, and has strong practical application significance.

[0070] like Figure 1 , which is a flow chart of the method of the present invention, comprising the following steps:

[0071] S1. Data preprocessing.

[0072] Before extracting features from the data, the point cloud data needs to be preprocessed in order to improve the accuracy of the model. In the point cloud processing software, the upper and lower limits of the xyz of the point cloud number extraction target are determined, and the corresponding threshold is set to remove points beyond this range, thereby retaining the area of ​​interest.

[0073] like Figure 2 The figure shows the point cloud 3D model after the original point cloud data is subjected to straight-through filtering and cropping. It can be clearly seen that the point cloud data after straight-through filtering and cropping significantly reduces unnecessary points and only retains points within the specified range.

[0074] S2. Refined processing

[0075] like Figure 3As shown in the figure, the normal vector can be used to further filter out unnecessary points. By calculating the normal vector of each point, the local geometric features of the point cloud surface can be determined. If the normal vector of some points deviates greatly from the expected surface normal vector direction, these points may be noise points or points that do not belong to the target structure. Filtering out these points with large deviations can significantly improve the accuracy and quality of the point cloud data.

[0076] Normal vector estimation

[0077] Normal vector estimation based on local surface fitting method: When the sampling surface of the point cloud is smooth and clear, the local shower of any point can be well fitted by a plane; for this reason, for each scan point p in the point cloud, search for the K adjacent points in its nearest area, and then calculate the local plane P of these points in the sense of least squares. For this reason, the plane can be expressed as:

[0078]

[0079] Where n is the normal vector of plane P, and d is the distance from P to the origin of the coordinate system.

[0080] The normal of the plane fitted by the K nearest points can be considered as the normal vector of the current scanning point. The normal vector of plane P can be obtained by principal component analysis (PCA). It is known from the calculation that P passes through the centroid P0 of its K domain points, and the normal vector satisfy First, perform eigenvalue decomposition on the covariance matrix M in formula (2) to obtain the eigenvalues ​​of M and the normal vector of the eigenvector P corresponding to the minimum eigenvalue of M.

[0081]

[0082] Normal vector orientation

[0083] The normal vector calculated above is ambiguous, that is, it only gets the line where the normal vector is located, but does not determine the direction of the line as the final direction of the normal vector. Use the following method to redirect the normal vector:

[0084] Assuming that the point cloud is dense enough and the sampling plane is smooth everywhere, the normal vectors of two adjacent points will be close to parallel. For two adjacent points x i , the normal vector of xj, if the direction of the normal vector is consistent, then If the inner product is negative, it means that the normal vector of one of the points needs to be flipped. Therefore, first set a normal vector direction for a point in the point cloud, and then traverse all other points. If the normal vector of the current point is set to is the next point to be traversed. If Then Flip, otherwise remain unchanged.

[0085] Surface curvature

[0086] Perform eigenvalue decomposition on the covariance matrix M in formula (2) to obtain the eigenvalues ​​of M. If the eigenvalues ​​satisfy λ0≤λ1≤λ2, the surface curvature of point P is:

[0087]

[0088] The smaller δ is, the flatter the area is, while the larger δ is, the more fluctuations in the neighborhood are.

[0089] S3, feature extraction;

[0090] like Figure 4 As shown in the figure, the random sample consensus algorithm is a powerful tool for feature extraction of point cloud data. By randomly selecting a subset from a large amount of data and fitting a model, random sample consensus can effectively identify and extract geometric features while ignoring noise and outliers. When processing point cloud data, random sample consensus is often used to extract planes, cylinders, or other geometric shapes. Through continuous iteration, the algorithm evaluates the fit of each model to the data and selects the optimal model to extract a subset of point clouds that meets the characteristics. It helps to obtain clear geometric features from complex point cloud data.

[0091] A random sampling consistency algorithm is used to find an optimal homography matrix H, with a matrix size of 3×3. The purpose of random sampling S is to find the optimal parameter matrix so that the number of data points that satisfy the matrix is ​​the largest. Usually, h33=1 is set to normalize the matrix. Since the homography matrix has 8 unknown parameters, at least 8 linear equations need to be solved. Corresponding to the point position information, a set of point pairs can list two equations, which contains at least 4 sets of matching point pairs.

[0092]

[0093] The random sample consistency algorithm randomly selects 4 samples from the matching data set and ensures that these 4 samples are not collinear, calculates the homography matrix, and then uses this model to test all data, and calculates the number of data points that satisfy this model and the projection error (ie, the cost function). If this model is the optimal model, the corresponding cost function is the minimum.

[0094] Loss function:

[0095]

[0096] Random sample consensus algorithm steps:

[0097] 1. Randomly extract 4 sample data from the data set (these 4 samples cannot be collinear), calculate the transformation matrix H, and record it as model M;

[0098] 2. Calculate the projection error between all data in the data set and the model M. If the error is less than the threshold, add the internal point set I;

[0099] 3. If the number of elements in the current interior point I is greater than the optimal interior point set I_best, update I_best = I and update the iteration number k at the same time;

[0100] 4. If the number of iterations is greater than k, then exit; otherwise, increase the number of iterations by 1 and repeat the above steps;

[0101] In addition to selecting appropriate data and models, the random sample consistency algorithm also needs to select appropriate four parameters n, k, t, and d. Among them, n, t, and d can be obtained based on experience, and k can be calculated according to the following formula:

[0102]

[0103] Among them, p represents the probability that the result of the random sample consistency algorithm is useful, and w is the probability that the data is in the inliers set. Then for the n data required for model fitting once, the probability that they are all in the inliers set is w n (replacement sampling probability), the probability of not being in the inliers set is 1-w n , so the result of k iterations satisfies:

[0104] 1-p=(1-w n ) k (10)

[0105] Thus, the calculation formula of k is obtained.

[0106] S4, secondary search;

[0107] like Figure 5 As shown in the figure, in order to ensure the accuracy of the target object extracted from the point cloud, the point cloud data after feature extraction can be compared with the original data for secondary retrieval, further optimizing the extraction results and correcting the deviations and errors in the initial extraction process.

[0108] a. Pass-through filtering

[0109] Create a pass filter object and set the point cloud data to be filtered;

[0110] Specify the filtering dimensions (X-axis, Y-axis, Z-axis), set the value range of each dimension by filtering threshold, and use 3D point cloud processing software to perform preliminary visual analysis of the point cloud data to determine the approximate location and range of the target object;

[0111] Through observation and interactive operation, determine the approximate coordinate range of the target object in the three directions of x, y and z. This can be done manually or by guessing based on the known size and position range of the object;

[0112] Traverse each point in the point cloud, determine whether the value of the point in the specified dimension is within the value range, and delete the points whose values ​​are not within the value range

[0113] At the end of the traversal, the remaining points constitute the filtered point cloud

[0114] b. Point cloud clipping

[0115] According to the extracted point cloud area, i.e., ROI area (use point cloud processing software to visualize the initial point cloud data and preliminarily identify the approximate location of the target object. Through interactive operations such as rotation, translation, and scaling, observe the point cloud from different angles to find the boundary or feature points of the target object). Separate the point cloud into regions; crop the point cloud according to the known ROI area; select the ROI area to be retained;

[0116] First, extract feature data to satisfy the following relationship:

[0117] x min ≤x≤x max

[0118] y min ≤y≤y max

[0119] z min ≤z≤z max

[0120] In the formula, x min , x max ,y min ,y max , z min , z max are the six boundary points of the rectangle.

[0121] c. Point cloud combination

[0122] Combine multiple point cloud data sets together without strict alignment. Simply add point cloud data that meets the features to another point cloud to form a more accurate point cloud data.

[0123] S5, remove the wall;

[0124] like Figure 6As shown in the figure, constructing a KD-tree can quickly perform nearest neighbor search and range search. Searching for data in a KD-tree is also an important part of feature matching. Its purpose is to retrieve the data point closest to the query point in the KD-tree. The distance between two points in an N-dimensional Cartesian space is determined by the following formula:

[0125]

[0126] Through cluster analysis, different objects in the point cloud can be identified, such as walls and electric poles. Once the point cloud of the wall is identified, these points can be quickly located and removed through the KD-tree index, thereby retaining the point cloud data of other target objects.

[0127] Embodiment 2:

[0128] The computer-readable storage medium of this embodiment stores a computer program thereon, which, when executed by a processor, implements the steps of the method for extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling the extracted targets in Example 1.

[0129] The computer-readable storage medium of this embodiment may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal; the computer-readable storage medium of this embodiment may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card, a secure digital card, a flash memory card, etc. equipped on the terminal; further, the computer-readable storage medium may also include both an internal storage unit of the terminal and an external storage device.

[0130] The computer-readable storage medium of this embodiment is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium can also be used to temporarily store data that has been output or is to be output.

[0131] Embodiment 3:

[0132] The computer device of this embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the method for extracting features of three-dimensional point cloud data based on random sample consistency and classifying and labeling extracted targets in Example 1 are implemented.

[0133] In this embodiment, the processor may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, readily available programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0134] Those skilled in the art will appreciate that the disclosed content of the embodiments may be provided as methods, systems, or computer program products. Therefore, the present solution may adopt the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Moreover, the present solution may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program codes.

[0135] The present solution is described with reference to the method according to the embodiment of the present solution and the flowchart and / or block diagram of the computer program product. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions; these computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 one or more processes and / or methods Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 one or more processes and / or methods Figure 1 A function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 one or more processes and / or methods Figure 1The steps for the functions specified in one or more boxes.

[0138] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0139] The examples described in the present invention are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the concept and scope of the present invention. Without departing from the design concept of the present invention, various modifications and improvements made to the technical solutions of the present invention by engineers and technicians in this field should all fall within the protection scope of the present invention.

Claims

1. A method for extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling the extracted targets, characterized in that: The following steps are involved: S1. Data preprocessing; The initial point cloud data is used to determine the xyz coordinates of the extracted target in the point cloud processing software, and the straight-through filtering method is used to set the corresponding xyz filtering threshold; S2. Refined processing; The filtered point cloud data is further filtered through the normal vector to filter out the noise points at the edge of the target object; S3. Feature extraction; Set the maximum number of iterations and the maximum distance from the point to the model, select the cylindrical model according to the characteristics of the target object, use the random sample consistency algorithm to fit the model, and perform feature extraction; S4. Secondary retrieval: comparing the extracted point cloud data with the original point cloud data, and performing secondary retrieval; S5. Eliminate the wall; after extracting the cylindrical features, Use KD-Tree clustering index to remove objects other than the target object; S6. Classification label; according to the type of point cloud in the point cloud data, simply distinguish it into target objects and non-target objects, and give the right classification label; train the global model through the point cloud data and classification labels until convergence, and obtain the pre-built global extraction model.

2. The method according to claim 1, characterized in that The specific steps of the method for determining the through-filter threshold in step S1 are as follows: S11. Preliminary analysis and determination of target area: Use point cloud processing software to conduct preliminary visual analysis of point cloud data to determine the approximate location and range of the target object; Determine the approximate coordinate range of the target object in the three directions of xyz through observation and interactive operation; S12. Perform statistical analysis on the point cloud data of the target area and calculate the point distribution density of the target object in the three directions of xyz; According to the statistical results and combined with experience, set the preliminary filtering threshold; S13. Iterative adjustment: Apply the initially set threshold to perform straight-through filtering on the point cloud data, and observe whether the filtered point cloud data contains the target object and does not have too much interference data; If the filtering result is not ideal, adjust the threshold value for multiple iterations to gradually approach the optimal threshold value; after each adjustment, recheck the filtered point cloud data until the expected effect is achieved; S14. Optimization based on feature extraction; when extracting the features of the target object, if some features are found to be weak or unclear, further adjust the filter threshold to enhance the performance of the features.

3. The method according to claim 2, characterized in that S1 also includes the following steps: S15.Automation optimization; When batch processing a large amount of point cloud data, consider using an automated optimization algorithm to automatically adjust the threshold to obtain the best filtering effect.

4. The method according to claim 1, characterized in that ,S2 specifically includes the following steps: Normal vector estimation: Normal vector estimation based on local surface fitting method: For each scan point p in the point cloud, search for the K neighboring points closest to the scan point p, and calculate the local plane P in the least squares sense of the neighboring points. The local plane P is expressed as: In the formula, is the normal vector of plane P, d is the distance from P to the origin of the coordinate system; The normal of the plane fitted by K adjacent points is the normal vector of the current scanning point; the normal vector of plane P is obtained by principal component analysis. It is known from the calculation that P passes through the centroid P0 of K adjacent points, and the normal vector satisfy First, perform eigenvalue decomposition on the covariance matrix M of formula (5) to obtain the eigenvalues ​​of M and the normal vector of the eigenvector P corresponding to the minimum eigenvalue of M. Normal vector orientation; Assuming that the point cloud is dense enough and the sampling plane is smooth everywhere, the normal vectors of two adjacent points will be close to parallel; make For two adjacent points x i , the normal vector of xj, if the direction of the normal vector is consistent, then If this inner product is negative, it means that the normal vector of one of the points needs to be flipped; therefore, first set a normal vector direction for a point in the point cloud, and then traverse all other points. If the normal vector of the current point is set to is the next point to be traversed. If Then Flip, otherwise remain unchanged; Surface curvature; Perform eigenvalue decomposition on the covariance matrix M of formula (5) to obtain the eigenvalues ​​of M. If the eigenvalues ​​satisfy λ0≤λ1≤λ2, the surface curvature of point P is: The smaller δ is, the flatter the area is, while the larger δ is, the more fluctuations in the neighborhood are.

5. The method according to claim 1, characterized in that The specific steps of S4 are: S41. Direct filtering; Create a pass filter object and set the point cloud data to be filtered; Use 3D point cloud processing software to conduct preliminary visual analysis of point cloud data to determine the approximate location and range of the target object; Determine the approximate coordinate range of the target object in the three directions of x, y and z through observation and interactive operation; manually mark or infer based on the known size and position range of the object; Specify the filtering dimension (X-axis, Y-axis, Z-axis), and set the value range of each dimension by filtering threshold. Traverse each point in the point cloud, determine whether the value of the point in the specified dimension is within the value range, and delete the points whose values ​​are not within the value range; After the traversal is completed, the remaining points constitute the filtered point cloud; S42. Point cloud clipping; Separate the point cloud region according to the extracted point cloud region, i.e., the ROI region; crop the point cloud according to the known ROI region; and select the ROI region to be retained; First, extract feature data to satisfy the following relationship: In the formula, x min , x max ,y min ,y max , z min , z max are the six boundary points of the rectangle; S43. Point cloud combination; Combine multiple point cloud data sets together and directly add the point cloud data that meets the features to another point cloud to form a more accurate point cloud data.

6. The method according to claim 5, characterized in that In S42, the ROI area is a polygon.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling extracted targets as described in any one of claims 1 to 6 are implemented.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for extracting features from three-dimensional point cloud data based on random sample consistency and classifying and labeling extracted targets are implemented as described in any one of claims 1 to 6.

Images