Substation equipment point cloud replacement method based on principal component analysis method

Abstract

The invention discloses a substation equipment point cloud replacement method based on a principal component analysis method. The method comprises the following specific steps: S1, preprocessing a target point cloud; s2, principal component analysis and calculation; s3, posture adjustment based on principal component analysis; s4, performing point cloud transformation; s5, matching the point cloud in the library with the attitude of the target point cloud so as to directly replace the target point cloud with the point cloud model in the library; according to the method, only a point cloud model of one view angle of the target point cloud is needed, attitude alignment is carried out on the point cloud model and a complete point cloud model in a self-built library, the target point cloud can be directly replaced, the complete point cloud model is complemented, convenience and rapidness are achieved, use is facilitated, a principal component analysis method is adopted, the calculation process is simple and convenient, and then a rotation matrix and a translation vector are calculated. The method is simple and convenient, and a large amount of computing resources and time are reduced.

Description

Technical Field

[0001] The invention relates to the technical field of point cloud processing, and in particular to a method for replacing point cloud of substation equipment based on principal component analysis. Background Art

[0002] Point cloud completion refers to the process of processing incomplete point cloud data to restore it to a more complete state. When there are point cloud data from multiple perspectives, each perspective may only obtain partial information about the object, resulting in incomplete point clouds. First, the point clouds from different perspectives need to be adjusted to make their postures consistent in space. Then, the information of the point clouds in the overlapping area is used to complete the missing parts. During the fusion process, by comparing the position and attributes of the points in the overlapping area, it is determined how to merge the point clouds from different perspectives to fill the missing areas. It is widely used in the field of 3D reconstruction. For example, when scanning a large electrical equipment, point cloud data from multiple perspectives is obtained by scanning from different directions and heights. Through the point cloud completion method based on multi-perspective fusion, these incomplete point cloud data can be completed to construct a more complete 3D model of the electrical equipment.

[0003] However, the existing point cloud completion requires multiple suitable perspectives to complete the completion better. If the number of perspectives is insufficient or the perspective selection is unreasonable, it may not be possible to obtain sufficient overlapping area information to complete the missing part, resulting in poor completion effect. In addition, the posture adjustment of point clouds from multiple different perspectives is a key step, but in actual operation, it is difficult to adjust the posture accurately. When it comes to operations such as matching and fusion of multi-perspective point clouds, the calculation process is relatively complicated, especially for large-scale point cloud data, which may require a lot of computing resources and time.

[0004] Therefore, it is necessary to further solve the above problems. The applicant has proposed a point cloud replacement method for substation equipment based on principal component analysis. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method for substation equipment point cloud replacement based on principal component analysis, which solves the problems raised by the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: A method for replacing substation equipment point cloud based on principal component analysis, the specific steps of which include: S1, preprocessing the target point cloud; S2, principal component analysis calculation; S3, posture adjustment based on principal component analysis; S4, performing point cloud transformation; S5, matching the pose of the library point cloud with the target point cloud, thereby directly replacing the target point cloud with the library point cloud model; The specific steps of step S1 are as follows: S11. For a given point cloud data , for any point , , , first remove outliers; S12, outliers are detected by statistical analysis method, the specific steps are as follows: Calculate the mean of a data set and standard deviation ; Determine a threshold, k=3; Will satisfy Data points Determine as outlier and remove, where For each data point in the dataset; Then the point cloud data is centralized and the center of mass of the point cloud is calculated. , and each point Convert to .

[0007] As a further solution of the present invention: the specific steps of step S2 are as follows: S21. Covariance matrix calculation: Calculate the covariance matrix of the point cloud , for a 3D point cloud, the covariance matrix is a The symmetric matrix of the covariance matrix Indicates and The covariance between the coordinate dimensions; S22, eigenvalue and eigenvector calculation: Covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue , arranged in order from largest to smallest, and the corresponding eigenvector , the eigenvector defines the principal component direction of the point cloud, and the maximum eigenvalue The corresponding eigenvector Indicates that the point cloud changes the most in this direction, which is the main direction; The corresponding eigenvector is the secondary direction and is related to Orthogonal; The corresponding eigenvector and and are all orthogonal, and are the third principal directions, such as Figure 2 and Figure 3 shown.

[0008] As a further solution of the present invention: the specific steps of step S3 are as follows: S31. Perform principal component analysis on the target point cloud and the complete point cloud of Curry respectively: Perform the above principal component analysis calculation on point cloud A and point cloud B respectively to obtain the centroid of point cloud A , eigenvalue and the eigenvector , and the centroid of point cloud B , eigenvalue and the eigenvector ; S32. Calculate the rotation matrix: The goal is to find a rotation matrix , so that , align the three eigenvectors of the target point cloud and the complete point cloud of Curry respectively: First, consider Align to , calculate the vector , the rotation axis , calculate the rotation angle , then according to the Rodriguez rotation formula Calculate the Align to The rotation matrix ,in is the identity matrix, Is The antisymmetric matrix formed by Based on Align to The rotation matrix , and Align to The rotation matrix , the final rotation matrix ; S33. Calculate the translation vector .

[0009] As a further solution of the present invention: the specific steps of step S4 are: For each point in point cloud A , transforming it into , thus realizing point cloud Pose and point cloud same.

[0010] Compared with the prior art, the present invention has the following beneficial effects: (1) This method only requires a point cloud model of one perspective of the target point cloud. After aligning the posture with the complete point cloud model in the self-built library and determining that it is the same type of power equipment, the target point cloud can be directly replaced to complete the complete point cloud model. It does not require multiple suitable perspectives to complete the completion well. At the same time, it will not fail to obtain sufficient overlapping area information to complete the missing part due to insufficient number of perspectives or unreasonable perspective selection, resulting in poor completion effect. It solves the problem that the actual substation scene can only obtain an incomplete point cloud model of a single perspective, which makes it impossible to complete the complete point cloud model. It is convenient and fast, and easy to use; (2) This method adopts the principal component analysis method, and the calculation process is simple and convenient. The principal component analysis is performed on the target point cloud and the complete point cloud in the library, and then the rotation matrix and translation vector are calculated to complete the posture alignment, thereby performing replacement. It is simple and convenient, and reduces a lot of computing resources and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a specific operation flow chart of the present invention; Figure 2 This is a schematic diagram of the target residual defect cloud model of the present invention; Figure 3 It is a schematic diagram of a complete point cloud model in the self-built point cloud library of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] Example See also Figure 1-3 The present invention provides a method for replacing substation equipment point cloud based on principal component analysis, and the specific steps include: Step 1: Preprocess the target point cloud (1) For a given point cloud data , for any point , , , first remove outliers; (2) Outliers are detected by statistical analysis method. The specific steps are as follows: Calculate the mean of a data set and standard deviation ; Determine a threshold, k=3; Will satisfy Data points Determine as outlier and remove, where For each data point in the dataset; Then the point cloud data is centralized and the center of mass of the point cloud is calculated. , and each point Convert to .

[0014] Step 2: Principal component analysis calculation (1) Covariance matrix calculation: Calculate the covariance matrix of the point cloud , for a 3D point cloud, the covariance matrix is a The symmetric matrix of the covariance matrix Indicates and The covariance between the coordinate dimensions; (2) Calculation of eigenvalues ​​and eigenvectors: Covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue , arranged in order from largest to smallest, and the corresponding eigenvector , the eigenvector defines the principal component direction of the point cloud, and the maximum eigenvalue The corresponding eigenvector Indicates that the point cloud changes the most in this direction, which is the main direction; The corresponding eigenvector is the secondary direction and is related to Orthogonal; The corresponding eigenvector and and are all orthogonal, which is the third main direction, as shown in Figure 2 and Figure 3 As shown in the figure, the three arrows represent the directions of the principal components, and the length represents the magnitude of the change in the point cloud in that direction.

[0015] Step 3: Posture adjustment based on principal component analysis (1) Perform principal component analysis on the target point cloud and the complete point cloud of Curry respectively: Perform the above principal component analysis calculation on point cloud A and point cloud B respectively to obtain the centroid of point cloud A , eigenvalue and the eigenvector , and the centroid of point cloud B , eigenvalue and the eigenvector ; (2) Calculate the rotation matrix: The goal is to find a rotation matrix , so that , align the three eigenvectors of the target point cloud and the complete point cloud of Curry respectively: First, consider Align to , calculate the vector , the rotation axis , calculate the rotation angle , then according to the Rodriguez rotation formula Calculate the Align to The rotation matrix ,in is the identity matrix, Is The antisymmetric matrix formed by Based on Align to The rotation matrix , and Align to The rotation matrix , the final rotation matrix ; (3) Calculate the translation vector .

[0016] Step 4: Transform the point cloud For each point in point cloud A , transforming it into , thus realizing point cloud Pose and point cloud same.

[0017] Step 5: Through the above steps, the pose of the library point cloud can be matched with that of the target point cloud, so that the library point cloud model can directly replace the target point cloud.

[0018] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for replacing substation equipment point cloud based on principal component analysis, characterized in that: The specific steps include: S1, preprocessing the target point cloud; S2, principal component analysis calculation; S3, posture adjustment based on principal component analysis; S4, performing point cloud transformation; S5, matching the pose of the library point cloud with the target point cloud, thereby directly replacing the target point cloud with the library point cloud model; The specific steps of step S2 are as follows: S21. Covariance matrix calculation: Calculate the covariance matrix of the point cloud , for a 3D point cloud, the covariance matrix is a The symmetric matrix of the covariance matrix Indicates and The covariance between the coordinate dimensions; S22, eigenvalue and eigenvector calculation: Covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue , arranged in order from largest to smallest, and the corresponding eigenvector , the eigenvector defines the principal component direction of the point cloud, and the maximum eigenvalue The corresponding eigenvector Indicates that the point cloud changes the most in this direction, which is the main direction; The corresponding eigenvector is the secondary direction and is related to Orthogonal; The corresponding eigenvector and and All are orthogonal and are the third principal direction.

2. The method for replacing substation equipment point cloud based on principal component analysis according to claim 1, characterized in that: The specific steps of step S1 are as follows: S11. For a given point cloud data , for any point , , , first remove outliers; S12, outliers are detected by statistical analysis method, the specific steps are as follows: Calculate the mean of a data set and standard deviation ; Determine a threshold, k=3; Will satisfy Data points Determine as outlier and remove, where For each data point in the dataset; Then the point cloud data is centralized and the center of mass of the point cloud is calculated. , and each point Convert to .

3. The method for replacing substation equipment point cloud based on principal component analysis according to claim 2 is characterized in that: The specific steps of step S3 are as follows: S31. Perform principal component analysis on the target point cloud and the complete point cloud of Curry respectively: Perform the above principal component analysis calculation on point cloud A and point cloud B respectively to obtain the centroid of point cloud A , eigenvalue and the eigenvector , and the centroid of point cloud B , eigenvalue and the eigenvector ; S32. Calculate the rotation matrix: The goal is to find a rotation matrix , so that , align the three eigenvectors of the target point cloud and the complete point cloud of Curry respectively: First, consider Align to , calculate the vector , the rotation axis , calculate the rotation angle , then according to the Rodriguez rotation formula Calculate the Align to The rotation matrix ,in is the identity matrix, Is The antisymmetric matrix formed by Based on Align to The rotation matrix , and Align to The rotation matrix , the final rotation matrix ; S33. Calculate the translation vector .

4. The method for replacing substation equipment point cloud based on principal component analysis according to claim 3 is characterized in that: The specific steps of step S4 are: For each point in point cloud A , transforming it into , thus realizing point cloud Posture and point cloud same.

Images