A point cloud registration method and computer equipment for stereo image pairs

Abstract

This invention discloses a point cloud registration method and computer equipment for stereo image pairs. The method includes: S1, using a 3D laser scanner equipped with a high-precision camera to acquire point clouds while simultaneously capturing multiple photographs with overlapping areas; S2, using the multiple photographs with overlapping areas to construct a stereo image pair model for feature points in the common areas, transforming the stereo image pair model from the image space auxiliary coordinate system to the point cloud space coordinate system where the 3D laser scanner is located; S3, selecting a first optimal matching feature point set and a second optimal matching feature point set; S4, selecting a final optimal matching feature point set, and using the final optimal matching feature point set for point cloud data matching and alignment. This invention enables more efficient and faster matching based on feature points, improving the accuracy and reliability of point cloud matching.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional laser scanning technology, and more specifically to a point cloud registration method and computer equipment for stereo image pairs. Background Technology

[0002] 3D point cloud data contains richer object, scene, and application information than 2D images, and point clouds can also efficiently store information data with less storage space. Currently, point cloud data collected by a single sensor can no longer meet development needs; point cloud matching from multiple sensors can more accurately describe the real 3D world. However, cross-source data from multiple sensors suffers from issues such as missing points, inconsistent point cloud density distributions, and different local data patterns, making it unsuitable for direct point cloud matching. Therefore, point cloud preprocessing is a necessary step.

[0003] In existing technologies, commonly used point cloud matching algorithms can be broadly categorized into four types: ICP algorithm, feature-based methods, deep learning-based methods, and local feature descriptor matching methods. The ICP algorithm is prone to getting trapped in local optima, is sensitive to initial estimates, and is susceptible to noise and outliers, requiring a good initial estimate. Feature-based methods have high computational complexity and demand significant computational resources, potentially requiring substantial memory and processing time. Deep learning-based methods require large amounts of labeled data for training, also placing high demands on computational resources and training time. Local feature descriptor matching methods may have low matching efficiency for large-scale datasets. Summary of the Invention

[0004] The present invention provides a point cloud registration method and computer device for stereo image pairs that effectively improves the matching efficiency of large-scale datasets, which can at least solve one of the above-mentioned technical problems.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A point cloud registration method for stereo image pairs includes the following steps:

[0007] S1. Using a 3D laser scanner with a high-precision camera, multiple photos with overlapping areas are taken while acquiring point clouds.

[0008] S2. Using multiple photos with overlapping areas, construct a stereo image pair model for the feature points of the common area. Assume an image space auxiliary coordinate system based on the positional relationship of each photo. The stereo image pair model is located in this image space auxiliary coordinate system. Then, through the raster positional relationship between the high-precision camera and the 3D laser scanner, transform the stereo image pair model in the image space auxiliary coordinate system to the point cloud space coordinate system where the 3D laser scanner is located.

[0009] S3. Perform bidirectional constraint matching between the feature point sets in the spatial coordinate system of the point cloud obtained by scanning the two adjacent grating emitters in the 3D laser scanner and the feature point sets in the spatial coordinate system of the point cloud calculated by the stereo image pair model, and select the first optimal matching feature point set and the second optimal matching feature point set.

[0010] S4. Perform bidirectional constraint matching again on the first optimal matching feature point set and the second optimal matching feature point set to select the final optimal matching feature point set, and use the final optimal matching feature point set to perform point cloud data matching and alignment.

[0011] Furthermore, in step S2, during the point cloud acquisition process, the 3D laser scanner uses a high-precision camera to capture photos with a specific overlap rate. By capturing multiple images with overlapping areas, the same feature point set P is obtained, and the feature point set P is projected onto the first or second photo to obtain feature point set projection p1 or feature point set projection p2. The first image space coordinate system of the first photo is different from the second image space coordinate system of the second photo. The second image space coordinate system is assumed to be the image space auxiliary coordinate system. Using the positional relationship between the image center pairs of the first and second photos and the photographic baseline length B, the first image space coordinate system and the second image space coordinate system are uniformly converted to the image space auxiliary coordinate system.

[0012] Furthermore, both the first and second photos are taken by a high-precision camera, and both the first and second photos are located between the first grating emitter and the second grating emitter. The first grating emitter is located on the left side, and the second grating emitter is located on the right side. When constructing the stereo image pair model, the first photo is taken by the high-precision camera located on the right side of the first grating emitter, and the second photo is taken by the high-precision camera located on the left side of the second grating emitter.

[0013] Furthermore, in S2, the transformation parameters between the image space auxiliary coordinate system and the point cloud space coordinate system are determined by the actual relative positional relationship between the first grating emitter and the high-precision camera located to the right of the first grating emitter, and the actual relative positional relationship between the second grating emitter and the high-precision camera located to the left of the second grating emitter, and the point cloud space coordinates of the feature point set P under the three-dimensional laser scanner are calculated.

[0014] Furthermore, the coordinates of the feature point set projection p1 and the feature point set projection p2 in the image space auxiliary coordinate system are {(x1, y1, z1) | x1∈first image space coordinate system, y1∈first image space coordinate system, z1∈first image space coordinate system} and {(x2, y2, z2) | x2∈second image space coordinate system, y2∈second image space coordinate system, z2∈second image space coordinate system}, respectively. The first image space coordinate system and the second image space coordinate system take the first photography center S1 and the second photography center S2 as the origin of the coordinate system. The x-axis and y-axis of the first image space coordinate system and the second image space coordinate system are parallel to the x-axis and y-axis of the image plane coordinate system, respectively. The image plane coordinate system is the positional relationship of the image point in the photograph plane, that is, the two-dimensional plane coordinate system where the photograph is located. The z-axis of both the first image space coordinate system and the second image space coordinate system coincides with the principal optical axis of the high-precision camera.

[0015] Based on the relative positional relationships between the first photography center S1 and the second photography center S2 and the first grating emitter and the second grating emitter, the transformation parameters between the image space auxiliary coordinate system and the point cloud spatial coordinate system under the 3D laser scanner are calculated. Using these transformation parameters, the feature point set coordinates {(x1, y1, z1) | x1∈Image space auxiliary coordinate system, y1∈Image space auxiliary coordinate system, z1∈Image space auxiliary coordinate system} and {(x2, y2, z2) | x2∈Image space auxiliary coordinate system, y2∈Image space auxiliary coordinate system, z2∈Image space auxiliary coordinate system} in the image space auxiliary coordinate system are respectively transformed into point cloud spatial coordinates {(x1, y1, z1) | x1∈Image space auxiliary coordinate system, y1∈Image space auxiliary coordinate system, z1∈Image space auxiliary coordinate system}. K1 Y K1 Z K1 )|X K1 ∈ the point cloud spatial coordinate system where the first grating emitter is located, Y K1 ∈ the point cloud spatial coordinate system where the first grating emitter is located, Z K1 ∈ point cloud spatial coordinate system where the first grating emitter is located} and {(X K2 Y K2 Z K2 )|X K2 ∈ the point cloud spatial coordinate system where the second grating emitter is located, Y K2 ∈ the point cloud spatial coordinate system where the second grating emitter is located, Z K2 ∈ point cloud spatial coordinate system where the second grating emitter is located}.

[0016] Furthermore, in S3, the 3D laser scanner with a high-precision camera has multiple grating emitters. Each grating emitter can obtain a set of feature points in the point cloud spatial coordinate system under that grating emitter by scanning. The set of feature points is the point cloud data obtained by that grating emitter.

[0017] The feature point set in the point cloud spatial coordinate system obtained by the first grating emitter in the adjacent grating emitters is matched with the feature point set in the point cloud spatial coordinate system calculated by the stereo image pair model through bidirectional constraint matching to select the first optimal matching feature point set. The feature point set in the point cloud spatial coordinate system obtained by the second grating emitter in the adjacent grating emitters is matched with the feature point set in the point cloud spatial coordinate system calculated by the stereo image pair model through bidirectional constraint matching to select the second optimal matching feature point set.

[0018] The point cloud spatial coordinates obtained by the first grating emitter and the second grating emitter are converted into the point cloud spatial coordinates obtained by the second grating emitter by using the positional relationship between two adjacent grating emitters, namely the vertical angle α, the rotation angle ω, and the distance R. The coordinate conversion formula is as follows:

[0019]

[0020]

[0021] .

[0022] Furthermore, in S4, the two point cloud data obtained by two adjacent grating emitters have common feature points at adjacent edges. The two point cloud data with common feature points are the point cloud data that need to be stitched and matched. The two point cloud data that need to be stitched and matched are compared with the stereo image pair model by bidirectional constraint coordinate difference. The first optimal matching feature point set and the second optimal matching feature point set are obtained by bidirectional constraint matching through the constraint coordinate difference.

[0023] The first optimal matching feature set and the second optimal matching feature set are respectively located in the point cloud spatial coordinate system where the first grating emitter and the second grating emitter are located. By rotating, translating and scaling the point cloud spatial coordinate system where the first grating emitter is located in turn through the positional relationship of the two adjacent grating emitters, the system is transformed into the point cloud spatial coordinate system where the second grating emitter is located. After unifying the coordinate system, the first optimal matching feature point set and the second optimal matching feature point set are compared again by bidirectional constraint coordinate difference to finally obtain the final optimal matching feature point set used for point cloud matching.

[0024] Furthermore, the feature point set P includes at least high curvature points, edge points, and corner points.

[0025] Furthermore, the first grating emitter and the second grating emitter project structured light stripes containing coded information onto the feature point set P.

[0026] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of a point cloud registration method for stereo image pairs.

[0027] The beneficial effects of this invention are reflected in:

[0028] 1. In this invention, the optimal matching feature point set is obtained by bidirectionally constraining the feature points in the spatial coordinates of the point cloud calculated by the stereo image pair model with the feature points in the point cloud data. This feature point set enables point cloud matching to be more efficient and faster based on the feature points.

[0029] 2. In this invention, the matching of feature points is improved and made more accurate through two-way constraints. Point cloud matching will reduce the problem of point cloud data offset based on these feature points, thus making the accuracy and reliability of point cloud matching higher. Attached Figure Description

[0030] The accompanying drawings, which are provided to further illustrate this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application.

[0031] Figure 1 This is a schematic diagram of the overall process of the point cloud registration method according to an embodiment of the present invention.

[0032] Figure 2 This is a schematic diagram of the construction structure of the stereo image pair model according to an embodiment of the present invention.

[0033] Figure 3 This is a schematic diagram of the relative position structure of the grating emitter and the photograph in an embodiment of the present invention.

[0034] Figure 4 This is a schematic diagram showing the positional relationship transformation of adjacent grating emitters in an embodiment of the present invention.

[0035] Figure 5 This is a structural block diagram of a computer device according to an embodiment of the present invention.

[0036] The components in the attached diagram are labeled as follows: 1. Feature point set P; 2. Feature point set projection p1; 3. Feature point set projection p2; 4. First image space coordinate system; 5. Second image space coordinate system; 6. First photography center S1; 7. Second photography center S2; 8. Photography baseline length B; 9. First grating emitter; 10. Second grating emitter; 11. First photograph; 12. Second photograph; 13. Vertical angle α; 14. Rotation angle ω; 15. Distance R. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] It should be noted that if the embodiments of the present invention involve descriptions such as "first" and "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" and "second" may explicitly or implicitly include at least one of those features. Furthermore, "multiple" refers to two or more.

[0039] See Figures 1-2 This invention provides a point cloud registration method for stereo image pairs, comprising the following steps:

[0040] S1. Using a 3D laser scanner with a high-precision camera, multiple photos with overlapping areas are taken while acquiring point clouds.

[0041] S2. Using multiple photos with overlapping areas, construct a stereo image pair model for the feature points of the common area. Assume an image space auxiliary coordinate system based on the positional relationship of each photo. The stereo image pair model is located in this image space auxiliary coordinate system. Then, through the raster positional relationship between the high-precision camera and the 3D laser scanner, transform the stereo image pair model in the image space auxiliary coordinate system to the point cloud space coordinate system where the 3D laser scanner is located.

[0042] S3. Perform bidirectional constraint matching between the feature point sets in the spatial coordinate system of the point cloud obtained by scanning the two adjacent grating emitters in the 3D laser scanner and the feature point sets in the spatial coordinate system of the point cloud calculated by the stereo image pair model, and select the first optimal matching feature point set and the second optimal matching feature point set.

[0043] S4. Perform bidirectional constraint matching again on the first optimal matching feature point set and the second optimal matching feature point set to select the final optimal matching feature point set, and use the final optimal matching feature point set to perform point cloud data matching and alignment.

[0044] See Figures 2-3In this embodiment, during point cloud acquisition, the 3D laser scanner uses a high-precision camera to capture images with a specific overlap rate. Multiple images with overlapping areas capture the same feature point set P1, and the feature point set P1 is projected onto the first image 11 or the second image 12 to obtain feature point set projection p12 or feature point set projection p23. The first image space coordinate system 4 of the first image 11 is different from the second image space coordinate system 5 of the second image 12. The second image space coordinate system 5 is assumed to be the image space auxiliary coordinate system. Using the positional relationship of the image center pairs of the first image 11 and the second image 12 and the image baseline length B8, the first image space coordinate system 4 and the second image space coordinate system 5 are uniformly converted to the image space auxiliary coordinate system.

[0045] See Figures 2-3 In this embodiment, both the first photo 11 and the second photo 12 are photos taken by a high-precision camera, and both the first photo 11 and the second photo 12 are located between the first grating emitter 9 and the second grating emitter 10. The first grating emitter 9 is located on the left side, and the second grating emitter 10 is located on the right side. When constructing the stereo image pair model, the first photo 11 is taken by the high-precision camera located to the right of the first grating emitter 9, and the second photo 12 is taken by the high-precision camera located to the left of the second grating emitter 10.

[0046] See Figures 2-3 In this embodiment, in step S2, the transformation parameters between the image space auxiliary coordinate system and the point cloud spatial coordinate system are determined based on the actual relative positional relationship between the first grating emitter 9 and the high-precision camera located to the right of the first grating emitter 9, and the actual relative positional relationship between the second grating emitter 10 and the high-precision camera located to the left of the second grating emitter 10. The point cloud spatial coordinates of the feature point set P1 under the 3D laser scanner are then calculated. The aforementioned actual relative positional relationships include the geometric relationships of optical imaging and the spatial relative relationships between the high-precision camera, the 3D laser scanner, and the grating emitter.

[0047] See Figures 2-3In this embodiment, the coordinates of the feature point set projection p12 and the feature point set projection p23 in the image space auxiliary coordinate system are {(x1, y1, z1) | x1∈First image space coordinate system 4, y1∈First image space coordinate system 4, z1∈First image space coordinate system 4} and {(x2, y2, z2) | x2∈Second image space coordinate system 5, y2∈Second image space coordinate system 5, z2∈Second image space coordinate system 5}, respectively. The first image space coordinate system 4 and the second image space coordinate system 5 take the first photography center S16 and the second photography center S27 as the origin of the coordinate system. The x-axis and y-axis of the first image space coordinate system 4 and the second image space coordinate system 5 are parallel to the x-axis and y-axis of the image plane coordinate system, respectively. The image plane coordinate system is the positional relationship of the image point in the photograph plane, that is, the two-dimensional plane coordinate system where the photograph is located. The z-axis of the first image space coordinate system 4 and the second image space coordinate system 5 coincide with the principal optical axis of the high-precision camera.

[0048] Based on the relative positional relationship between the first photography center S16 and the second photography center S27 and the first grating emitter 9 and the second grating emitter 10, the transformation parameters between the image space auxiliary coordinate system and the point cloud spatial coordinate system under the 3D laser scanner are calculated. Using these transformation parameters, the feature point set coordinates {(x1, y1, z1) | x1∈Image space auxiliary coordinate system, y1∈Image space auxiliary coordinate system, z1∈Image space auxiliary coordinate system} and {(x2, y2, z2) | x2∈Image space auxiliary coordinate system, y2∈Image space auxiliary coordinate system, z2∈Image space auxiliary coordinate system} in the image space auxiliary coordinate system are respectively transformed into point cloud spatial coordinates {(x1, y1, z1) | x1∈Image space auxiliary coordinate system, y1∈Image space auxiliary coordinate system, z1∈Image space auxiliary coordinate system}. K1 Y K1 Z K1 )|X K1 ∈ The point cloud spatial coordinate system where the first grating emitter 9 is located, Y K1 ∈ The point cloud spatial coordinate system where the first grating emitter 9 is located, Z K1 ∈ point cloud spatial coordinate system where the first grating emitter 9 is located} and {(X K2 Y K2 Z K2 )|X K2 ∈ the point cloud spatial coordinate system where the second grating emitter 10 is located, Y K2 ∈ The point cloud spatial coordinate system where the second grating emitter 10 is located, Z K2 ∈ point cloud spatial coordinate system where the second grating emitter 10 is located}.

[0049] Typically, the X, Y, and Z axes of the auxiliary image space coordinate system are parallel to the axes corresponding to the image space coordinate system of the first image of the flight path, with the origin at the first photography center S16 and the second photography center S27. This method uses the image space coordinate system of the first image of the flight path or the left-side image to replace the auxiliary image space coordinate system.

[0050] See Figures 2-4 In this embodiment, in step S3, the three-dimensional laser scanner with a high-precision camera has multiple grating emitters. Each grating emitter can obtain a set of feature points in the point cloud spatial coordinate system under that grating emitter by scanning. The set of feature points is the point cloud data obtained by that grating emitter.

[0051] The feature point set in the point cloud spatial coordinate system obtained by the first grating emitter 9 in the adjacent grating emitters is matched with the feature point set in the point cloud spatial coordinate system calculated by the stereo image pair model through bidirectional constraint matching to select the first optimal matching feature point set. The feature point set in the point cloud spatial coordinate system obtained by the second grating emitter 10 in the adjacent grating emitters is matched with the feature point set in the point cloud spatial coordinate system calculated by the stereo image pair model through bidirectional constraint matching to select the second optimal matching feature point set.

[0052] The point cloud spatial coordinates obtained by the first grating emitter 9 and the second grating emitter 10 are converted into the point cloud spatial coordinates obtained by the second grating emitter 10 through the positional relationship between two adjacent grating emitters, namely the vertical angle α13, the rotation angle ω14, and the distance R15. The coordinate conversion formula is as follows:

[0053]

[0054]

[0055] .

[0056] See Figures 1-3 In this embodiment, in step S4, the two point cloud data obtained by two adjacent grating emitters have common feature points at adjacent edges. The two point cloud data with common feature points are the point cloud data that need to be stitched and matched. The two point cloud data that need to be stitched and matched are compared with the stereo image pair model by bidirectional constraint coordinate difference. The first optimal matching feature point set and the second optimal matching feature point set are obtained by bidirectional constraint matching through the constraint coordinate difference.

[0057] The first optimal matching feature set and the second optimal matching feature set are respectively located in the point cloud spatial coordinate system where the first grating emitter 9 and the second grating emitter 10 are located. By rotating, translating and scaling the point cloud spatial coordinate system where the first grating emitter 9 is located in turn through the positional relationship of the two adjacent grating emitters, the system is transformed into the point cloud spatial coordinate system where the second grating emitter 10 is located. After unifying the coordinate system, the first optimal matching feature point set and the second optimal matching feature point set are compared again by bidirectional constraint coordinate difference to finally obtain the final optimal matching feature point set used for point cloud matching.

[0058] When comparing the spatial coordinates of the point cloud calculated from the point cloud data with the feature points of the point cloud spatial coordinates calculated from the stereo image pair model, the feature points on the spatial coordinates of the point cloud calculated from the stereo image pair model are used as the basis. Then, feature points whose coordinate differences are less than the limit are searched one by one in the feature set of the point cloud data acquired by the first grating emitter (9) on the left side. This yields a feature point matching set of the left point cloud data relative to the stereo image pair, and similarly, a feature point matching set of the stereo image pair relative to the left point cloud data. If some feature points can simultaneously satisfy both matching sets, this is the optimal matching feature point set P. 左 ={feature point 1, feature point 2, feature point 3...}.

[0059] Similarly, based on the feature points in the spatial coordinates of the point cloud calculated by the stereo image pair model, feature points whose coordinate differences are less than the limit requirement are searched one by one in the feature set of the point cloud data acquired by the second grating emitter (10) on the right side. This yields the feature point matching set of the right point cloud data relative to the stereo image pair, and similarly, the feature point matching set of the stereo image pair relative to the right point cloud data. If some feature points simultaneously satisfy both matching sets, this is the optimal matching feature point set P. 右 ={feature point 1, feature point 2, feature point 3...}.

[0060] Again to P 左 and P 右 The two sets are compared by bidirectional constraint coordinate difference, and the optimal matching feature point set P = {feature point 1, feature point 2, feature point 3...} is obtained. This set P is used for point cloud matching and alignment.

[0061] In this embodiment, the feature point set P1 includes at least high curvature points, edge points, and corner points. These points possess important feature information, which can be used to describe the local shape and structure of the point cloud, and can be stored and transmitted as representative features of the point cloud data.

[0062] In this embodiment, the first grating emitter 9 and the second grating emitter 10 project structured light stripes containing coded information onto the feature point set P1. When the emitted light beam illuminates the surface of the object being scanned, it is modulated into deformed light stripes. These modulated gratings contain the three-dimensional information of the surface of the object being scanned.

[0063] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the point cloud registration method for stereo image pairs described above.

[0064] See Figure 5 The present invention also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the point cloud registration method for stereo image pairs as described above.

[0065] This invention also provides a computer program product containing instructions that, when run on a computer, causes the computer to perform the steps of the point cloud registration method for stereo image pairs described above.

[0066] It is understood that the system, device and storage medium provided in the embodiments of the present invention correspond to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above-described point cloud registration method for stereo image pairs.

[0067] It should be noted that those skilled in the art will understand that all or part of the steps implemented in the embodiments of the present invention can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented in hardware, it can be implemented entirely or partially by purchasing standard parts or modifications. When implemented in software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0068] In summary, this application provides a novel point cloud matching method that integrates different point cloud data into a globally consistent point cloud model in both same-source and cross-source point cloud matching. The method primarily involves constructing a stereo image pair model using multiple images with overlapping regions, determining the coordinates of feature points in the point cloud spatial coordinate system, performing a corresponding search and transformation estimation with the actual scanned point cloud coordinates to select the optimal edge feature point, and then performing a second least-squares iterative matching to determine the final optimal edge feature point that meets the coordinate difference requirement. This feature point is then used for high-precision point cloud matching.

[0069] It should be understood that the examples and embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art can make various modifications or changes based on them. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A point cloud registration method for stereo image pairs, characterized in that, Includes the following steps: S1. Using a 3D laser scanner with a high-precision camera, multiple photos with overlapping areas are taken while acquiring point clouds. S2. Using multiple photos with overlapping areas, construct a stereo image pair model for the feature points of the common area. Assume an image space auxiliary coordinate system based on the positional relationship of each photo. The stereo image pair model is located in this image space auxiliary coordinate system. Then, through the raster positional relationship between the high-precision camera and the 3D laser scanner, transform the stereo image pair model in the image space auxiliary coordinate system to the point cloud space coordinate system where the 3D laser scanner is located. S3. Perform bidirectional constraint matching between the feature point sets in the spatial coordinate system of the point cloud obtained by scanning the two adjacent grating emitters in the 3D laser scanner and the feature point sets in the spatial coordinate system of the point cloud calculated by the stereo image pair model, and select the first optimal matching feature point set and the second optimal matching feature point set. S4. Perform bidirectional constraint matching again on the first optimal matching feature point set and the second optimal matching feature point set to select the final optimal matching feature point set, and use the final optimal matching feature point set to perform point cloud data matching and alignment.

2. The point cloud registration method for stereo image pairs as described in claim 1, characterized in that, In S2, during the point cloud acquisition process, the three-dimensional laser scanner uses a high-precision camera to take photos with a specific overlap rate. It captures the same feature point set P (1) through multiple images with overlapping areas and projects the feature point set P (1) onto the first photo (11) or the second photo (12) to obtain the feature point set projection p1 (2) or the feature point set projection p2 (3). The first image space coordinate system (4) of the first photo (11) is different from the second image space coordinate system (5) of the second photo (12). The second image space coordinate system (5) is assumed to be the image space auxiliary coordinate system. Using the positional relationship of the image center pairs of the first photo (11) and the second photo (12) and the photography baseline length B (8), the first image space coordinate system (4) and the second image space coordinate system (5) are uniformly converted to the image space auxiliary coordinate system.

3. The point cloud registration method for stereo image pairs as described in claim 2, characterized in that, The first photo (11) and the second photo (12) are both photos taken by a high-precision camera. The first photo (11) and the second photo (12) are both located between the first grating emitter (9) and the second grating emitter (10). The first grating emitter (9) is located on the left side and the second grating emitter (10) is located on the right side. When constructing a stereo image pair model, the first photo (11) is taken by the high-precision camera located to the right of the first grating emitter (9), and the second photo (12) is taken by the high-precision camera located to the left of the second grating emitter (10).

4. The point cloud registration method for stereo image pairs as described in claim 3, characterized in that, In S2, the transformation parameters between the image space auxiliary coordinate system and the point cloud space coordinate system are determined by the actual relative positional relationship between the first grating emitter (9) and the high-precision camera located to the right of the first grating emitter (9), and the actual relative positional relationship between the second grating emitter (10) and the high-precision camera located to the left of the second grating emitter (10), and the point cloud space coordinates of the feature point set P (1) under the three-dimensional laser scanner are calculated.

5. The point cloud registration method for stereo image pairs as described in claim 4, characterized in that, The coordinates of the feature point set projection p1 (2) and the feature point set projection p2 (3) in the auxiliary coordinate system of image space are {(x1, y1, z1) | x1∈first image space coordinate system (4), y1∈first image space coordinate system (4), z1∈first image space coordinate system (4)} and {(x2, y2, z2) | x2∈second image space coordinate system (5), y2∈second image space coordinate system (5), z2∈second image space coordinate system (5)}, respectively. The first image space coordinate system (4) and the second image space coordinate system (5) are respectively. The image space coordinate system (5) takes the first photography center S1 (6) and the second photography center S2 (7) as the origin of the coordinate system. The x-axis and y-axis of the first image space coordinate system (4) and the second image space coordinate system (5) are parallel to the x-axis and y-axis of the image plane coordinate system, respectively. The image plane coordinate system is the positional relationship of the image point in the plane of the photograph, that is, the two-dimensional plane coordinate system where the photograph is located. The z-axis of the first image space coordinate system (4) and the second image space coordinate system (5) are both coincident with the main optical axis of the high-precision camera. Based on the relative positional relationship between the first photography center S1 (6) and the second photography center S2 (7) and the first grating emitter (9) and the second grating emitter (10), the transformation parameters between the image space auxiliary coordinate system and the point cloud spatial coordinate system under the 3D laser scanner are calculated. Through the transformation parameters, the feature point set coordinates {(x1, y1, z1)|x1∈image space auxiliary coordinate system, y1∈image space auxiliary coordinate system, z1∈image space auxiliary coordinate system} and {(x2, y2, z2)|x2∈image space auxiliary coordinate system, y2∈image space auxiliary coordinate system, z2∈image space auxiliary coordinate system} in the image space auxiliary coordinate system are respectively converted into point cloud spatial coordinates {(x1, y1, z1)|x1∈image space auxiliary coordinate system, y1∈image space auxiliary coordinate system, z1∈image space auxiliary coordinate system} and {(x2, y2, z2)|x2∈image space auxiliary coordinate system, y2∈image space auxiliary coordinate system, z2∈image space auxiliary coordinate system}. K1 Y K1 Z K1 )|X K1 ∈ The point cloud spatial coordinate system where the first grating emitter (9) is located, Y K1 ∈ The point cloud spatial coordinate system where the first grating emitter (9) is located, Z K1 ∈ The point cloud spatial coordinate system where the first grating emitter (9) is located} and {(X K2 Y K2 Z K2 )|X K2 ∈ The point cloud spatial coordinate system where the second grating emitter (10) is located, Y K2 ∈ The point cloud spatial coordinate system where the second grating emitter (10) is located, Z K2 ∈ the point cloud spatial coordinate system where the second grating emitter (10) is located}.

6. The point cloud registration method for stereo image pairs as described in claim 5, characterized in that, In S3, the three-dimensional laser scanner with a high-precision camera has multiple grating emitters. Each grating emitter can obtain a set of feature points in the point cloud spatial coordinate system under that grating emitter by scanning. The set of feature points is the point cloud data obtained by that grating emitter. The feature point set in the spatial coordinate system of the point cloud obtained by the first grating emitter (9) in the adjacent grating emitter is matched with the feature point set in the spatial coordinate system of the point cloud calculated by the stereo image pair model in a two-way constraint, and the first optimal matching feature point set is selected. The feature point set in the spatial coordinate system of the point cloud obtained by the second grating emitter (10) in the adjacent grating emitter is matched with the feature point set in the spatial coordinate system of the point cloud calculated by the stereo image pair model in a two-way constraint, and the second optimal matching feature point set is selected. The point cloud spatial coordinates obtained by the first grating emitter (9) and the second grating emitter (10) are converted into the point cloud spatial coordinates obtained by the second grating emitter (10) through the positional relationship between the two adjacent grating emitters, namely the vertical angle α (13), the rotation angle ω (14), and the distance R (15). The coordinate conversion formula is as follows: 。 7. The point cloud registration method for stereo image pairs as described in claim 6, characterized in that, In S4, the two point cloud data obtained by two adjacent grating emitters have common feature points at adjacent edges. The two point cloud data with common feature points are the point cloud data that need to be stitched and matched. The two point cloud data that need to be stitched and matched are compared with the stereo image pair model by bidirectional constraint coordinate difference. The first optimal matching feature point set and the second optimal matching feature point set are obtained by bidirectional constraint matching through the constraint coordinate difference. The first optimal matching feature set and the second optimal matching feature set are respectively located in the point cloud space coordinate system where the first grating emitter (9) and the second grating emitter (10) are located. By rotating, translating and scaling the point cloud space coordinate system where the first grating emitter (9) is located in turn through the positional relationship of the two adjacent grating emitters, the system is transformed to the point cloud space coordinate system where the second grating emitter (10) is located. After unifying the coordinate system, the first optimal matching feature set and the second optimal matching feature set are compared again by bidirectional constraint coordinate difference to finally obtain the final optimal matching feature set used for point cloud matching.

8. The point cloud registration method for stereo image pairs as described in claim 2, characterized in that, The feature point set P(1) includes at least high curvature points, edge points, and corner points.

9. The point cloud registration method for stereo image pairs as described in claim 3, characterized in that, The first grating emitter (9) and the second grating emitter (10) project structured light stripes containing coded information onto the feature point set P (1).

10. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the point cloud registration method for stereo image pairs as described in any one of claims 1-9.

Images