A Key Feature Reconstruction Method for Curved Antennas with Multimodal Visual Data Fusion

Abstract

The present invention provides a method for reconstructing key features of a curved surface antenna with multi-modal visual data fusion, which relates to the field of vision-based feature extraction. The method includes the following steps: designing an artificial enhanced feature pattern and projecting it onto the surface of the curved surface antenna to be measured; using a structured light camera to collect surface images and three-dimensional point clouds of the curved surface antenna to be measured containing the artificial enhanced feature pattern at different viewpoints; establishing a feature vector of all the center points of the assembly holes in the viewpoints based on the surface images; obtaining the matching relationship of the assembly holes in the three-dimensional point cloud, mapping the matching results of the assembly holes to the three-dimensional point cloud, and stitching and fusing them at multiple different viewpoints to obtain the complete point cloud data of the curved surface antenna to be measured; performing semantic segmentation and feature fitting on the complete point cloud data to output the set of center points of the assembly holes and the set of center points of the pads. Through non-contact measurement with multi-modal visual data fusion, the present invention realizes high-precision measurement of the size of the curved surface antenna and feature reconstruction without reducing the resolution.

Description

Technical Field

[0001] The present invention relates to the field of vision-based feature extraction, and particularly to a method for reconstructing key features of a curved surface antenna by fusing multi-modal visual data. Background Art

[0002] Existing mechanical part measurement methods are mainly implemented by two methods: contact type and non-contact type. The contact type measurement method uses a high-precision probe to contact the surface of the workpiece to obtain the coordinate data of the workpiece, and realizes the measurement of three-dimensional dimensions, shapes, and positions. This method usually has a slow measurement speed, is not suitable for measuring features with small scales, and the contact of the probe may cause surface damage to some precision workpieces, and is not suitable for measuring the curved surface conformal antenna parts studied in the present invention. The non-contact error measurement method is mainly realized by photoelectric imaging technology, such as structured light scanning, visible light or infrared ultraviolet camera photography, etc. This method does not need to contact the object to be measured, is suitable for measuring fine or objects with relatively complex surfaces, and at the same time has a fast data acquisition speed and high flexibility in use, and is suitable for the measurement object of the present invention. However, there is a principle contradiction between the field of view range and the resolution size in the photoelectric imaging technology, which makes it difficult to perform high-precision measurement on large-scale parts.

[0003] There is an existing technology: clamping a turbine blade on a fixture in the shape of a cylinder-like body, and using a secondary search method for marker points to improve the stability of rough point cloud registration, realizing high-quality multi-point cloud stitching. However, this method has a large dependence on marker points, and the number of target marker points in the overlapping area between two adjacent images needs to reach about 20, which has certain limitations for three-dimensional reconstruction applications with a small overlapping area.

[0004] The Chinese invention patent with the publication number CN118298110A discloses a three-dimensional point cloud image reconstruction method and system for the part manufacturing process. This method uses a quantitatively measurable rotary fixture to clamp the object to be measured, and fuses the rotation matrix of the rotary fixture into the fast ICP process to realize the reconstruction of the object to be measured. This method needs to rely on the geometric information provided by an external motion mechanism and does not fully utilize the information of the image itself.

[0005] The above methods have realized multi-viewpoint three-dimensional point cloud stitching, but have put forward higher requirements for the self-characteristics of the object to be measured. When the self-characteristics of the object to be measured are insufficient, additional features such as marker points are added or an external motion mechanism is used to obtain the conversion relationship between different viewpoint images, which increases the complexity of the measurement process. Summary of the Invention

[0006] Object of the Invention: To propose a method for reconstructing key features of a curved surface antenna by fusing multi-modal visual data to solve the above problems existing in the prior art.

[0007] The present invention proposes a key feature reconstruction method for a curved surface antenna with multimodal visual data fusion, including the following steps:

[0008] Design an artificial enhanced feature pattern and project it onto the surface of the curved surface antenna to be measured;

[0009] Use a structured light camera to capture surface images of the curved surface antenna to be measured containing the artificial enhanced feature pattern as two-dimensional images at different viewpoints, and synchronously collect the three-dimensional point cloud of the curved surface antenna to be measured;

[0010] Based on the two-dimensional images, establish feature vectors for all the center points of the assembly holes in the viewpoints;

[0011] Obtain the matching relationship of the assembly holes in the three-dimensional point cloud, map the matching results of the assembly holes to the three-dimensional point cloud, and splice and fuse them at multiple different viewpoints to obtain the complete point cloud data of the curved surface antenna to be measured;

[0012] Perform semantic segmentation and feature fitting on the complete point cloud data, and output the set of center points of the assembly holes and the set of center points of the pads.

[0013] In a further embodiment, the projection area A of the artificial enhanced feature pattern on the surface of the curved surface antenna to be measured is a rectangular area, and the length of the projection area A is and the width is ;

[0014] The aspect ratio of the length and width of the projection area A is the same as the aspect ratio of the length and width of the minimum bounding rectangle of the curved surface antenna in the two-dimensional camera imaging plane;

[0015] The length x and width y of the artificial enhanced feature pattern satisfy , , where k is a safety factor.

[0016] In a further embodiment, projecting the artificial enhanced feature pattern onto the surface of the curved surface antenna to be measured specifically includes:

[0017] Mesh the projection area A into , divide it into rectangular cells, and the number of cells satisfies , , where d is the diameter of the end face circle of the self-owned assembly hole of the curved surface antenna to be measured;

[0018] Generate random points in each of the obtained cells ;

[0019] Taking the random point as the center, draw a filled circular colored area, and the diameter of this circular colored area satisfies , where MP is the pixel size of the camera used to capture the two-dimensional image.

[0020] In a further embodiment, the random points generated Coordinates satisfy:

[0021]

[0022]

[0023] In the formula, i and j are the horizontal and vertical indexes of the grid respectively; , is the minimum value of the horizontal and vertical coordinates of the projection area A in the coordinate system; , is the length and width of the cell, which satisfies , ; is a random coefficient.

[0024] In a further embodiment, when the structured light camera captures the surface image of the curved antenna to be tested containing the artificially enhanced characteristic pattern at different viewpoints:

[0025] The overlapping area of ​​two adjacent viewpoint images contains at least two assembly hole features.

[0026] In a further embodiment, based on the two-dimensional image, a feature vector of the center points of all assembly holes in the viewpoint is established, specifically including:

[0027] Fit the circle in the two-dimensional image under all viewpoints and obtain its center to obtain the center point set of the assembly hole and the center point set of artificial feature patterns ;

[0028] The points in the assembly hole center point set P As the target point, search for the point in the center point set Q of the artificial feature pattern The a points with the closest Euclidean distance are used to establish an a-dimensional distance vector in ascending order of Euclidean distance. The feature vectors of the mounting holes of the curved antenna.

[0029] In a further embodiment, obtaining the matching relationship of the assembly holes in the three-dimensional point cloud specifically includes:

[0030] Fit the center of the assembly hole in the 3D point cloud and project it on the XY plane to obtain the projection point set of the center of the assembly hole , projection point set of the center of the assembly hole The corresponding relationship between the feature centers is obtained by aligning with the center point set P of the assembly hole, so as to establish the matching relationship between the assembly holes in the overlapping area of ​​the 3D point cloud of adjacent viewpoints and obtain the center point pair set of the assembly hole ,in is the center point of an assembly hole in the 3D point cloud of a certain viewing point, and is the center point of the assembly hole with the same name in the 3D point cloud of its adjacent viewing point in physical space.

[0031] In a further embodiment, the matching result of the assembly hole is mapped to the 3D point cloud and stitched and fused under multiple different viewing points, specifically including:

[0032] According to the matching relationship of the assembly hole in the 3D point cloud, filter the point cloud data within the cube region with the center of the assembly hole as the geometric center and the edge length of a in the 3D point cloud of a certain viewing point and its adjacent viewing point 3D point cloud, and use it as the point cloud to be registered and ;

[0033] Taking the point cloud to be registered as the source point cloud and the point cloud to be registered as the target point cloud, search for the nearest neighbor point in the target point cloud for each point in the source point cloud to obtain a paired point set ;

[0034] Define the weight in the form of a compound Gaussian for each point pair in the paired point set :

[0035]

[0036] In the formula, and are respectively the minimum values of the Euclidean distances between the points and and all the assembly hole centers in their respective point clouds; is a parameter to control the weight decay rate;

[0037] Calculate the weighted centroid:

[0038]

[0039]

[0040] In the formula, N is the number of elements in the paired point set , and M is the number of elements in the point pair set ; is the weight of the assembly hole center point pair set;

[0041] Calculate the weighted covariance matrix H:

[0042]

[0043] Perform SVD decomposition on the covariance matrix H, calculate the rotation matrix R and the translation vector t, and obtain the registration matrix T according to the kinematic principle;

[0044] Iteratively update the transformation matrix T until the convergence condition is satisfied;

[0045] Perform pairwise registration on the point clouds of all viewpoints in sequence to obtain a number of transformation matrices;

[0046] Using the point cloud data of a certain viewpoint as the overall target point cloud, perform a rigid transformation on the remaining point clouds:

[0047]

[0048] In the formula, X represents the number of pairwise registrations required to transform to the overall target point cloud coordinate system, and the size of X depends on the number of pairwise registrations required to transform the point cloud C to the overall target point cloud coordinate system; C represents the point cloud before transformation, represents the point cloud after transformation; is the transformation matrix under the current viewpoint;

[0049] Fuse the overall target point cloud with all the point clouds after rigid transformation to obtain the complete point cloud data of the surface antenna to be measured.

[0050] In a further embodiment, perform semantic segmentation and feature fitting on the complete point cloud data, specifically including:

[0051] Set the voxel grid size according to the desired accuracy requirement, and perform pass-through filtering, voxelization, statistical filtering, and smoothing processing on the complete point cloud data of the surface antenna to be measured to obtain the pre-processed point cloud;

[0052] Calculate the normal vector of the pre-processed point cloud;

[0053] Segment the boundary points of the point cloud according to the normal vector difference, and perform secondary segmentation on the boundary points to obtain the boundary points of the assembly holes;

[0054] Calculate the curvature of the point cloud according to the normal vector, and identify the pad plane;

[0055] Verify and eliminate the misidentified results, and finally obtain the set of center points of the assembly holes and the set of center points of the pads.

[0056] Compared with the prior art, the remarkable advantages of the present invention are: non-contact measurement through multi-modal visual data fusion can overcome the contradiction between the field of view and the resolution, increase the measurement range without reducing the resolution, and thus achieve high-precision measurement of large-size parts represented by surface antennas and reconstruction of micro-features such as assembly holes and pads, providing methods and technical support for the measurement of manufacturing errors and the adaptive assembly process. Brief Description of the Drawings

[0057] Figure 1 It is a schematic flow chart of the present invention.

[0058] Figure 2 It is a schematic flow chart of point cloud semantic segmentation and fitting of assembly holes and pad features of the present invention. Detailed implementation manners

[0059] In the following description, numerous specific details are given to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details. In other instances, in order to avoid confusion with the present invention, some well-known technical features are not described.

[0060] An embodiment of the present invention provides a method for reconstructing key features of a curved surface antenna by fusing multi-modal visual data. This method uses a structured light camera based on a monocular telecentric lens to capture images of the surface of a to-be-tested curved surface antenna part containing an artificial enhancement feature pattern to obtain two-dimensional images, and based on the principle of multi-light source coding and phase resolution to obtain the corresponding surface three-dimensional point cloud. Through the fusion of two-dimensional images and three-dimensional point clouds, the stitching of multi-viewpoint three-dimensional point clouds is realized, and then the reconstruction of key features of the curved surface antenna is realized. See Figure 1 , and its technical solution is divided into the following four major steps:

[0061] Step S1: Design of artificial enhancement feature pattern:

[0062] Step S11. In order to make the present invention have better environmental robustness, the present invention uses a laser projector with adjustable brightness and focal length to project the artificial enhancement feature pattern on the surface of the to-be-tested curved surface antenna.

[0063] Step S12. The projection area A of the artificial enhancement feature pattern is a rectangular area, and its aspect ratio is the same as the aspect ratio of the minimum bounding rectangle of the to-be-tested curved surface antenna in the two-dimensional camera imaging plane. The length x and width y of the artificial enhancement feature pattern should satisfy , , where k is a safety factor, which is determined according to engineering experience and is usually greater than 1. x and y are the length and width of the minimum bounding rectangle of the to-be-tested curved surface antenna in the two-dimensional camera imaging plane respectively.

[0064] Step S13. Mesh the projection area A into , and divide out rectangular cells. The number of cells should satisfy , , where d is the diameter of the end face circle of the self-assembly hole of the to-be-tested curved surface antenna.

[0065] Step S14. In the Randomly generate points in each cell , whose coordinates satisfy:

[0066]

[0067]

[0068] where and are grid indices;

[0069] , is the minimum value of the projection area A in a certain coordinate system;

[0070] , are the length and width of the rectangular cell, which satisfy , ;

[0071] is a random coefficient.

[0072] Step S15: Using the random points generated in step S14 as the center, draw a filled circular colored area, and the diameter of this area should satisfy , where MP is the pixel size of the camera used to collect the two-dimensional image.

[0073] Step S16: To ensure that there is no overlap in the circular areas of the artificial enhanced feature pattern, the random coefficient should satisfy:

[0074]

[0075]

[0076] Step S2: Unique coding of the assembly holes of the surface antenna based on the artificial enhanced feature pattern:

[0077] To achieve the unique coding of the self-owned features of the part, fit the circular light spot and the center of the hole feature of the surface antenna itself in the two-dimensional image, and construct a feature vector according to their spatial relative position relationship. Step S2 specifically includes:

[0078] Step S21: Use a monocular structured light camera based on a telecentric lens to obtain two-dimensional images and three-dimensional point clouds from different viewpoints, ensuring that the spatial position of the telecentric lens does not change when obtaining the two-dimensional image and three-dimensional point cloud of the same viewpoint. Through reasonable viewpoint planning, ensure that the overlapping area of two adjacent viewpoint images contains at least two assembly hole features.

[0079] Step S22: Fit the circles in the two-dimensional images of all viewpoints using the Hough transform method and obtain the circle centers, thus obtaining the set of assembly hole center points and the set of artificial feature pattern center points .

[0080] Step S23: Take the points in P as target points, search for a points in set Q that are the closest to the points in terms of Euclidean distance, and establish an a-dimensional distance vector in ascending order of Euclidean distance as the feature vector of the assembly hole of the surface antenna.

[0081] Step S24: Repeat Step S23 to establish the feature vectors of the assembly hole center points in all viewpoints, design a fuzzy threshold, regard the feature vector elements with a difference not greater than the fuzzy threshold as the same element, and perform fuzzy matching on the hole feature vectors in adjacent viewpoints to achieve the matching of the assembly holes in the overlapping regions of adjacent viewpoints.

[0082] Step S25: Fit the center of the assembly hole feature in the three-dimensional point cloud using the RANSAC algorithm, project it onto the XY plane to obtain the set of projected points of the assembly hole center , use the ICP algorithm to register the point set and the point set P to obtain the corresponding relationship of the feature centers, thereby establishing the matching relationship of the assembly holes in the overlapping regions of the three-dimensional point clouds of adjacent viewpoints, and obtaining the set of paired points of the assembly hole centers , where is a certain assembly hole center point in the three-dimensional point cloud of a certain viewpoint, is the assembly hole center point in the three-dimensional point cloud of its adjacent viewpoint that has the same name as in the physical space.

[0083] Step S3: Multi-viewpoint three-dimensional point cloud registration based on the improved ICP (Iterative Closest Point) algorithm

[0084] To achieve the stitching of multi-viewpoint three-dimensional point clouds, design an improved ICP algorithm based on the spatial geometric information weighted algorithm, and add the center of the hole feature of the surface antenna itself to the objective function of the improved ICP algorithm to achieve the stitching of three-dimensional point clouds from different viewpoints. Step S3 specifically includes:

[0085] Step S31: Based on the assembly hole matching relationship established in Step S25, screen the point cloud data within a certain cube region with an edge length of a centered on the assembly hole center in the three-dimensional point cloud of a certain viewpoint and its adjacent viewpoint three-dimensional point cloud, and use it as the point cloud to be registered and .

[0086] Step S32: Take the point cloud as the source point cloud, the point cloud Taking the target point cloud, using KD-tree to accelerate the nearest neighbor search, finding the nearest neighbor points in the target point cloud for each point in the source point cloud, and obtaining the paired point set .

[0087] Step S33: Define the weight in the form of a compound Gaussian for each point pair :

[0088]

[0089] wherein and are respectively the minimum values of the Euclidean distances between the points and and the centers of all assembly holes in their respective point clouds; is a parameter for controlling the weight decay rate, and its matrix range is determined by engineering experience, usually 0.1 - 0.3;

[0090] Step S34: Calculate the weighted centroid:

[0091]

[0092]

[0093] wherein, N is the number of elements in the paired set described in step S32, and M is the number of elements in the assembly hole center point pair set described in step S25. is the weight of the assembly hole center point pair set, and its specific value is determined by engineering experience and is related to N.

[0094] Step S35: Calculate the weighted covariance matrix:

[0095]

[0096] Perform SVD decomposition on the covariance matrix H, calculate the rotation matrix R and the translation vector t, and obtain the registration matrix T according to the kinematic principle.

[0097] Step S36: Repeat steps S32 to S35, iteratively update the transformation matrix T until the convergence condition is satisfied. The convergence condition is usually the number of iterations or the error change threshold.

[0098] Step S37: Perform pairwise registration on the point clouds of all viewpoints in sequence to obtain several transformation matrices. Taking the point cloud data of a certain viewpoint as the overall target point cloud, perform rigid transformation on the remaining point clouds . The size of X depends on the number of pairwise registrations required to transform the point cloud C into the overall target point cloud coordinate system.

[0099] ​Step S38: Fuse the overall target point cloud with all the point clouds after performing rigid transformations to obtain the complete point cloud data of the surface antenna to be measured.

[0100] Step S4: 3D point cloud semantic segmentation and feature fitting:

[0101] To realize the reconstruction of the key features of the surface antenna, an algorithm for fitting the positions of the assembly holes and pads of the surface antenna based on attribute clustering is designed to realize the reconstruction of the key features of the surface antenna. See Figure 2 , Step S4 specifically includes:

[0102] Step S41: Point cloud preprocessing: Set the voxel grid size according to the accuracy requirements, and perform pass-through filtering, voxelization, statistical filtering, and smoothing on the obtained complete point cloud data of the surface antenna. The pass-through filtering, voxelization, statistical filtering, and smoothing processes in this process are known methods, and those skilled in the art can directly use them, so they will not be elaborated here.

[0103] Step S42: Calculate the normal vectors of the preprocessed point cloud based on PCA (Principal Component Analysis). Segment the boundary points of the point cloud according to the normal vector differences, and use the region growing method to perform secondary segmentation on the boundary points to obtain the assembly hole boundary points; calculate the point cloud curvature according to the normal vector to realize the recognition of the pad plane, use the RANSAC method for shape fitting, and finally perform feature verification and pseudo-feature removal to obtain the set of assembly hole center points and the set of pad center points. In this process, PCA (Principal Component Analysis), the region growing method, and the RANSAC method are existing algorithms without specific improvements. Those skilled in the art can directly use them or use other similar technologies to replace them. This patent does not make special limitations on this and does not need to be elaborated either.

[0104] Use a coordinate measuring machine to measure as the true value to measure the measurement accuracy of the present invention. The measurement results are shown in Table 1 below:

[0105] Table 1: Comparison of coordinate measuring values and the measurement values of the present invention

[0106] Assembly hole serial number X Y Z 1 0.000158 0.006166 0.003988 2 0.010763 0.014723 0.019025 3 0.003842 0.000247 0.011533 4 0.009104 0.014592 0.014971 5 0.011877 0.010277 0.004582 6 0.012532 0.012617 0.002821 7 0.005480 0.000533 0.011419 8 0.004652 0.017415 0.000250 9 0.018239 0.009327 0.012900 Average error 0.008516 0.009544 0.009054 Maximum error 0.018239 0.017415 0.019025

[0107] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as a limitation of the present invention itself. Various changes can be made in its form and details without departing from the spirit and scope of the present invention defined by the appended claims.

Claims

1. A method for reconstructing key features of curved antennas based on multimodal visual data fusion, characterized in that: The steps include: Design an artificially enhanced feature pattern and project it onto the surface of the curved antenna to be tested; Using a structured light camera to capture the surface image of the curved antenna to be tested containing the artificially enhanced characteristic pattern at different viewpoints as a two-dimensional image, and simultaneously collecting a three-dimensional point cloud of the curved antenna to be tested; Based on the two-dimensional image, establishing feature vectors of the center points of all assembly holes in the viewpoint; Obtain the matching relationship of the assembly holes in the 3D point cloud, map the matching results of the assembly holes to the 3D point cloud, and merge them under multiple different viewpoints to obtain the complete point cloud data of the curved antenna to be tested; Performing semantic segmentation and feature fitting on the complete point cloud data, and outputting an assembly hole center point set and a pad center point set; The matching results of the assembly holes are mapped to a three-dimensional point cloud and then stitched and fused at multiple different viewpoints, specifically including: According to the matching relationship of the assembly holes in the 3D point cloud, select the point cloud data in the cube area with the center of the assembly hole as the geometric center and the edge length a in the 3D point cloud of a certain viewpoint and its adjacent viewpoints, and use it as the point cloud to be registered and ; Point cloud to be registered is the source point cloud, the point cloud to be registered For the target point cloud, for each point in the source point cloud, search for the nearest neighbor point in the target point cloud to obtain a pairing point set; For each pair of points in the paired point set Define the weights of the composite Gaussian form : ； In the formula, and Points and The minimum value of the Euclidean distance to the center of all assembly holes in the respective point cloud; Parameters that control the speed of weight decay; Compute the weighted centroid: ； ； Where N is the pairing point set The number of elements, M is the collocation pair set The number of elements in ; Assign weights to the point pairs of the assembly hole centers; Calculate the weighted covariance matrix H: ； Perform SVD decomposition on the covariance matrix H, calculate the rotation matrix R and translation vector t, and obtain the registration matrix T according to the kinematic principle; The transformation matrix T is updated iteratively until the convergence condition is met.

2. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 1, characterized in that: The projection area A of the artificially enhanced characteristic pattern on the surface of the curved antenna to be tested is a rectangular area, and the length of the projection area A is , width ; The aspect ratio of the projection area A is the same as the aspect ratio of the minimum envelope rectangle of the curved antenna to be measured on the imaging plane of the two-dimensional camera; The length x and width y of the artificially enhanced characteristic pattern satisfy , , k is the safety factor.

3. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 2, characterized in that: Projecting the artificial enhanced characteristic pattern onto the surface of the curved antenna to be tested includes: The projection area A is meshed as , divide rectangular cells, the number of cells satisfies , , d is the diameter of the end face of the assembly hole of the curved antenna to be tested; Generate a random point in each cell obtained ; Random Points Draw a filled circular colored area with the center as the circle. The diameter of the circular colored area is satisfy , MP is the pixel size of the camera used to acquire the two-dimensional image.

4. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 3, characterized in that: The random points generated Coordinates satisfy: ； ； In the formula, i and j are the horizontal and vertical indexes of the grid respectively; , is the minimum value of the horizontal and vertical coordinates of the projection area A in the coordinate system; , is the length and width of the cell, which satisfies , ; is a random coefficient.

5. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 1, characterized in that: When the structured light camera captures the surface image of the curved antenna to be tested containing the artificially enhanced characteristic pattern at different viewpoints: The overlapping area of ​​two adjacent viewpoint images contains at least two assembly hole features.

6. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 1, characterized in that: Based on the two-dimensional image, the feature vectors of the center points of all assembly holes in the viewpoint are established, specifically including: Fit the circle in the two-dimensional image under all viewpoints and obtain its center to obtain the center point set of the assembly hole and the center point set of artificial feature patterns ; The points in the assembly hole center point set P As the target point, search for the point in the center point set Q of the artificial feature pattern The a points with the closest Euclidean distance are used to establish an a-dimensional distance vector in ascending order of Euclidean distance. The feature vectors of the mounting holes of the curved antenna.

7. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 6, characterized in that: The obtaining of the matching relationship of the assembly holes in the three-dimensional point cloud specifically includes: Fit the center of the assembly hole in the 3D point cloud and project it on the XY plane to obtain the projection point set of the center of the assembly hole , projection point set of the center of the assembly hole The corresponding relationship between the feature centers is obtained by aligning with the center point set P of the assembly hole, so as to establish the matching relationship between the assembly holes in the overlapping area of ​​the 3D point cloud of adjacent viewpoints and obtain the center point pair set of the assembly hole ,in is the center point of a certain assembly hole in the 3D point cloud of a certain viewpoint, The 3D point cloud of its adjacent viewpoints is The center points of the mounting holes with the same name in physical space.

8. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 1, characterized in that: Map the matching results of the assembly holes to the 3D point cloud and merge them from multiple viewpoints. It also includes: The point clouds of all viewpoints are registered one by one in turn to obtain several transformation matrices; Take the point cloud data of a certain viewpoint as the overall target point cloud and perform rigid transformation on the remaining point clouds: ； Where X represents the number of pairwise registrations required to transform to the overall target point cloud coordinate system; C represents the point cloud before transformation, Represents the transformed point cloud; is the transformation matrix under the current viewpoint; The overall target point cloud is fused with all the point clouds after rigid transformation to obtain the complete point cloud data of the curved antenna to be tested.

9. The method for reconstructing key features of curved antennas by multimodal visual data fusion according to claim 8, characterized in that: The complete point cloud data is subjected to semantic segmentation and feature fitting, specifically including: The voxel grid size is set according to the desired accuracy requirement, and the complete point cloud data of the curved antenna to be measured is subjected to through-filtering, voxelization, statistical filtering and smoothing to obtain a pre-processed point cloud; Calculate the normal vector of the pre-processed point cloud; The boundary points of the point cloud are segmented according to the normal vector difference, and the boundary points are segmented again to obtain the boundary points of the assembly holes; Calculate the point cloud curvature based on the normal vector and identify the pad plane; Verify and eliminate the incorrect recognition results, and finally obtain the assembly hole center point set and the pad center point set.

Images