Multi-view point cloud fusion splicing method and device, storage medium and computer program product

Abstract

The invention discloses a multi-view point cloud fusion splicing method and device, a storage medium and a computer program product, and relates to the technical field of point cloud processing, and the method comprises the steps: obtaining an original point cloud image corresponding to each view, carrying out the transposition transformation of each original point cloud image based on a preset transformation rule, and determining a to-be-spliced point cloud; obtaining point cloud gravity centers of the to-be-spliced point clouds, and constructing a collapsed triangle according to the point cloud gravity centers; and carrying out fusion splicing on the to-be-spliced point clouds based on the collapse triangle, and obtaining a point cloud splicing result when a preset collapse stop condition is satisfied. According to the point cloud fusion splicing method and device, fusion splicing of point clouds of all view angles is achieved by obtaining the gravity center of the point clouds to construct the collapsing triangle and setting the stopping condition, and compared with an existing mode, the point clouds which are low in overlapping rate and not concentrated in overlapping area can be effectively spliced without depending on the point cloud overlapping condition.

Description

Technical Field

[0001] The present application relates to the field of point cloud processing technology, and in particular to a multi-view point cloud fusion and splicing method, device, storage medium and computer program product. Background Art

[0002] Laser profile sensors can use laser scanning technology, based on the principle of laser triangulation strategy, to output high-precision point cloud data in real time to build a three-dimensional model of an object. Therefore, the profiler built based on the laser profile sensor can be used in scenarios with high measurement accuracy requirements, such as defect inspection of bars. Profilers are usually used for plane-like scanning, that is, the point cloud obtained by the profiler when scanning a certain surface of an object in a single scan is actually 2.5D. Therefore, it is necessary to obtain point cloud images of two or more surfaces of the object before splicing them to obtain a three-dimensional point cloud model of the object.

[0003] However, since traditional point cloud stitching algorithms such as principal component analysis (PCA) and iterative closest point (ICP) rely on the overlap of point clouds in point cloud images of each view, and the overlap of point cloud images of each view obtained by profilometer scanning is too low and only partially overlapped, the point cloud stitching results obtained by stitching point cloud images of each view of the profilometer based on traditional point cloud stitching algorithms are poor. Summary of the invention

[0004] The main purpose of this application is to provide a multi-view point cloud fusion stitching method, device, storage medium and computer program product, aiming to solve the technical problem that the traditional point cloud stitching algorithm obtains poor point cloud stitching results when stitching point cloud images of various viewpoints of the profilometer.

[0005] To achieve the above objectives, the present application proposes a multi-view point cloud fusion and stitching method, which includes:

[0006] Acquire the original point cloud images corresponding to each viewing angle, and perform transposition transformation on each of the original point cloud images based on a preset transformation rule to determine the point cloud to be spliced;

[0007] Obtaining the point cloud centroid of each of the point clouds to be spliced, and constructing a collapsed triangle according to the point cloud centroids;

[0008] The point clouds to be spliced ​​are fused and spliced ​​based on the collapsed triangles, and the point cloud splicing result is obtained when a preset collapse stop condition is met.

[0009] In one embodiment, before the step of obtaining the point cloud centroids of each of the point clouds to be spliced ​​and constructing collapsed triangles according to the point cloud centroids, the step further includes:

[0010] Aligning the point clouds to be spliced ​​according to a preset alignment method to obtain the aligned point clouds to be spliced, wherein the preset alignment method is alignment based on a Y-axis reference point and / or alignment based on an X-axis reference point;

[0011] Correspondingly, the step of obtaining the point cloud centroids of each of the point clouds to be spliced, and constructing collapsed triangles according to the point cloud centroids, includes:

[0012] The point cloud centroids of the aligned point clouds to be spliced ​​are obtained, and collapsed triangles are constructed according to the point cloud centroids.

[0013] In one embodiment, the step of aligning the point clouds to be spliced ​​according to a preset alignment method to obtain the aligned point clouds to be spliced ​​includes:

[0014] Calculate the point cloud gravity center corresponding to each of the point clouds to be spliced ​​according to the gravity center solution formula, and determine the point cloud gravity center point according to the gravity center of each of the point clouds;

[0015] The center point of the center of gravity of the point cloud is used as a Y-axis reference point, and each of the point clouds to be spliced ​​is aligned based on the Y-axis reference point to obtain the aligned point clouds to be spliced.

[0016] In one embodiment, the step of aligning the point clouds to be spliced ​​according to a preset alignment method to obtain the aligned point clouds to be spliced ​​further includes:

[0017] Determine the first edge point and the second edge point of each of the point clouds to be spliced ​​in the X-axis direction;

[0018] Determine whether the first edge point and the second edge point meet preset abnormal isolated point conditions respectively;

[0019] If not, obtaining the X-axis midpoints corresponding to the point clouds to be spliced ​​according to the first edge point and the second edge point, and determining the center point in the X-axis direction according to the X-axis midpoints;

[0020] The center point in the X-axis direction is used as an X-axis reference point, and each of the point clouds to be spliced ​​is aligned based on the X-axis reference point to obtain the aligned point clouds to be spliced.

[0021] In one embodiment, the step of respectively determining whether the first edge point and the second edge point satisfy a preset abnormal isolated point condition comprises:

[0022] Calculate the mean and standard deviation of the nearest point distances corresponding to each of the point clouds to be spliced;

[0023] Obtain the distance between the first edge point and the nearest neighbor point of the second edge point;

[0024] It is determined whether the difference between the nearest neighbor distance and the mean of the nearest neighbor distances is greater than the standard deviation.

[0025] In one embodiment, the step of fusing and splicing the point clouds to be spliced ​​based on the collapsed triangles and obtaining the point cloud splicing result when a preset collapse stop condition is met includes:

[0026] Based on each vertex of the collapsed triangle, collapse and fuse each of the point clouds to be spliced ​​along the Z-axis direction;

[0027] Determine the number of overlapping points of the current point cloud according to the collapse fusion result, and judge whether the number of overlapping points of the current point cloud is the maximum number of overlapping points of the point cloud;

[0028] If so, the collapsed fusion result corresponding to the maximum point cloud overlap number is used as the point cloud stitching result.

[0029] In one embodiment, the original point cloud images corresponding to each viewing angle include: a middle viewing angle point cloud image, a left viewing angle point cloud image, and a right viewing angle point cloud image, and the step of performing a transposition transformation on each of the original point cloud images based on a preset transformation rule to determine the point cloud to be spliced ​​includes:

[0030] Based on a preset observation perspective, display perspective conversion is performed on the middle perspective point cloud image, the left perspective point cloud image, and the right perspective point cloud image respectively;

[0031] Determine the mid-view point cloud to be spliced ​​according to the mid-view point cloud image after the display view is converted;

[0032] Performing a first rotation transformation on the left-view point cloud image after the display perspective conversion to determine the left-view point cloud to be spliced;

[0033] A second rotation transformation is performed on the right view point cloud image after the display view conversion to determine the right view point cloud to be spliced.

[0034] In addition, to achieve the above purpose, the present application also proposes a multi-view point cloud fusion and splicing device, the device comprising:

[0035] A point cloud loading module is used to obtain the original point cloud images corresponding to each viewing angle, and perform a transposition transformation on each of the original point cloud images based on a preset transformation rule to determine the point cloud to be spliced;

[0036] A splicing preparation module, used for obtaining the point cloud centroids of each of the point clouds to be spliced, and constructing collapsed triangles according to the point cloud centroids;

[0037] The result generating module is used to merge and splice the point clouds to be spliced ​​based on the collapsed triangles, and obtain the point cloud splicing result when a preset collapse stop condition is met.

[0038] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, on which a multi-view point cloud fusion and stitching program is stored. When the multi-view point cloud fusion and stitching program is executed by a processor, the steps of the multi-view point cloud fusion and stitching method described above are implemented.

[0039] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the multi-view point cloud fusion and stitching method as described above.

[0040] The present application discloses a multi-view point cloud fusion and stitching method, including: obtaining the original point cloud image corresponding to each view, and performing a transposition transformation on each of the original point cloud images based on a preset conversion rule to determine the point cloud to be stitched; obtaining the point cloud centroid of each of the point clouds to be stitched, and constructing a collapsed triangle based on the centroid of each of the point clouds; fusing and stitching each of the point clouds to be stitched based on the collapsed triangle, and obtaining the point cloud stitching result when a preset collapse stop condition is met. Since the present application first performs a transposition transformation on the original point cloud image according to a preset conversion rule, so that the original point cloud image is adaptively adjusted, and then the point cloud is fused and stitched by obtaining the point cloud centroid to construct a collapsed triangle and setting a stop condition, compared with the existing method, the relative position relationship between each point cloud image can be accurately determined independently of a large number of overlapping points, thereby achieving effective stitching between point clouds with low overlap rates and non-concentrated overlapping areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0043] Figure 1 This is a flowchart of the first embodiment of the multi-view point cloud fusion and stitching method of the present application;

[0044] Figure 2 Schematic diagram of the original point cloud image corresponding to each viewing angle;

[0045] Figure 3 It is a schematic diagram of fusion stitching based on collapsed triangles;

[0046] Figure 4 This is a flow chart of the second embodiment of the multi-view point cloud fusion and stitching method of the present application;

[0047] Figure 5 This is a flowchart of the third embodiment of the multi-view point cloud fusion and stitching method of the present application;

[0048] Figure 6 This is a schematic diagram of the entire process of the multi-view point cloud fusion and stitching method of this application;

[0049] Figure 7 This is a schematic diagram of the first operating scenario corresponding to the multi-view point cloud fusion and stitching method of this application;

[0050] Figure 8 This is a schematic diagram of a second operating scenario corresponding to the multi-view point cloud fusion and stitching method of this application;

[0051] Fig. 9 This is a schematic diagram of a third operating scenario corresponding to the multi-view point cloud fusion and stitching method of the present application method;

[0052] Fig.10 This is a schematic diagram of the module structure of the first embodiment of the multi-view point cloud fusion and stitching device of the present application.

[0053] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0054] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0055] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0056] The present application embodiment provides a multi-view point cloud fusion and splicing method, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the multi-view point cloud fusion and stitching method of the present application. In this embodiment, the method includes: Steps S10 to S30:

[0057] Step S10: obtaining original point cloud images corresponding to each viewing angle, and performing transposition transformation on each of the original point cloud images based on a preset transformation rule to determine a point cloud to be spliced.

[0058] It should be noted that this embodiment can be applied to the scene of constructing a three-dimensional point cloud model of an object, and can also be applied to the scene of fusing and stitching multi-view point cloud images. The execution subject of the method of this embodiment can be a computing electronic device with functions such as data processing, network communication, and program running, such as a mobile phone, a tablet, a computer, etc., and can also be other electronic devices that can achieve the same or similar functions. The following takes a multi-view point cloud fusion and stitching device (hereinafter referred to as a "sticking device") as an example to illustrate this embodiment and the following embodiments.

[0059] It should be understood that the stitching device may have a display interface for interacting with a user, and the original point cloud image may be obtained by importing point cloud data into the stitching device and displaying it on the display interface. The original point cloud data may be in PCD format, or in other formats such as ECD, CSV, etc.

[0060] Specifically, after acquiring the original point cloud data imported by the user, the stitching device can uniformly convert the original point cloud data into PCD format, and perform preliminary filtering on the original point cloud data to remove discrete points, thereby displaying the imported original point cloud image on the display interface.

[0061] It should be noted that the point cloud images of each view can be obtained by rotating the inspected object and scanning it from three angles by a profilometer, including: a middle view point cloud image, a left view point cloud image, and a right view point cloud image. The profilometer can use a laser line profilometer with high measurement accuracy (the measurement accuracy can reach 0.00387mm), thereby ensuring that the point cloud splicing result obtained by subsequent fusion has high accuracy on the non-spliced ​​surface.

[0062] In order to facilitate users to stitch point cloud images of various perspectives based on the display interface, the stitching angle of the point cloud of each perspective can be adjusted based on the preset conversion rules: first, the coordinate system of the point cloud images of each perspective can be transposed, and the Z axis and the Y axis can be transposed and interchanged; then the point cloud image of the middle perspective remains unchanged, the point cloud image of the left perspective is rotated at a negative angle, and the point cloud image of the right perspective is rotated at a positive angle, so as to obtain the point clouds to be stitched corresponding to each perspective.

[0063] Specifically, in order to illustrate the transposition transformation process of the original point cloud image based on the preset transformation rule, step S10 includes: steps S101 to S104:

[0064] Step S101: performing display perspective conversion on the middle perspective point cloud image, the left perspective point cloud image and the right perspective point cloud image respectively based on a preset observation perspective.

[0065] It should be noted that the profilometer is installed directly above the object to be inspected for scanning, so the original point cloud image obtained can be Figure 2 As shown, Figure 2 Schematic diagram of the original point cloud image corresponding to each perspective. Figure 2 In the figure, 2-a is the left-view point cloud image, 2-b is the middle-view point cloud image, and 2-c is the right-view point cloud image.

[0066] Depend on Figure 2 It can be seen that the point cloud coordinate system corresponding to each original point cloud image is: the horizontal direction of the original point cloud image is the X-axis (point cloud X-axis), the vertical direction is the Y-axis (point cloud Y-axis), and the direction perpendicular to the original point cloud image is the Z-axis (point cloud Z-axis). The coordinate system (world coordinate system) provided by the display interface for users to observe can maintain the same perspective as the point cloud's own coordinate system, that is, after the point cloud data is imported, the user can observe the XY plane (point cloud reference plane) of the original point cloud image on the display interface, and the coordinates of the original point cloud points in each original point cloud image can be expressed based on the world coordinate system.

[0067] In order to facilitate user observation, the original point cloud image can be visually rotated so that the original point cloud image observed by the user on the display interface is rotated from the XY plane to the XZ plane. At this time, in order to make the world coordinate system consistent with the point cloud's own coordinate system perspective, the world coordinate system can be transposed from the Y axis to the Z axis, so that the world coordinate system transposition is synchronized with the point cloud visual rotation, ensuring that the coordinate representation of the original point cloud points in each original point cloud image remains unchanged, and completing the display perspective conversion.

[0068] Step S102: determining a mid-view point cloud to be spliced ​​according to the mid-view point cloud image after display view conversion.

[0069] Step S103: performing a first rotation transformation on the left-view point cloud image after the display perspective conversion to determine the left-view point cloud to be spliced.

[0070] Step S104: performing a second rotation transformation on the right view point cloud image after the display view conversion to determine the right view point cloud to be spliced.

[0071] It should be understood that since the scanning angles of the point cloud images of each perspective differ by 120 degrees, the mid-view point cloud image can be used as a reference to directly determine the mid-view point cloud to be spliced ​​according to the mid-view point cloud image after the display perspective conversion, and for the sake of subsequent unified expression, the i-th point in the mid-view point cloud is represented as (x Ci ',y Ci ',z Ci ').

[0072] The left-view point cloud image after the display perspective conversion is rotated 120 degrees around the Y axis. The coordinates of the i-th point in the left-view point cloud before rotation are (x Li ,y Li ,z Li), the coordinates of the i-th point in the left view point cloud after rotation are (x Li ',y Li ',z Li '), then the coordinate relationship before and after the rotation, that is, the first rotation transformation relationship, can be expressed as follows:

[0073]

[0074] The right view point cloud image after the display perspective conversion is rotated 120 degrees around the Y axis. The coordinates of the i-th point in the right view point cloud before rotation are (x Ri ,y Ri ,z Ri ), the coordinates of the i-th point in the left view point cloud after rotation are (x Ri ',y Ri ',z Ri '), then the coordinate relationship before and after the rotation, that is, the second rotation transformation relationship, can be expressed as follows:

[0075]

[0076] In the specific implementation, the original point cloud images of each perspective are transposed respectively, so as to obtain the point clouds to be spliced ​​corresponding to each perspective on the display interface (the point cloud of the middle perspective to be spliced, the point cloud of the left perspective to be spliced, and the point cloud of the right perspective to be spliced), ensuring that the interlacing angle of the XY plane (point cloud reference plane) of each point cloud to be spliced ​​is 120 degrees.

[0077] Step S20: obtaining the point cloud centroids of each of the point clouds to be spliced, and constructing collapsed triangles according to the point cloud centroids.

[0078] First, the point cloud centroid of each point cloud to be spliced ​​can be obtained. The point cloud centroid can be calculated in advance from the average position of all the point cloud points in each point cloud to be spliced. The point cloud centroid is the geometric centroid of each point cloud to be spliced. Next, an equilateral triangle can be constructed, and the point cloud centroid of each point cloud to be spliced ​​can be assigned to the vertices of the equilateral triangle respectively; finally, each point cloud to be spliced ​​is moved to the three vertices of the equilateral triangle, and the equilateral triangle is the collapsed triangle used to splice each point cloud to be spliced.

[0079] Step S30: fusing and splicing the point clouds to be spliced ​​based on the collapsed triangles, and obtaining the point cloud splicing result when a preset collapse stop condition is met.

[0080] It should be noted that the stitching device can collapse and fuse each of the point clouds to be stitched along the Z-axis direction of the point cloud coordinate axis based on the vertices of the collapsed triangle; determine the number of overlapping points of the current point cloud according to the collapse and fusion result, and judge whether the current number of overlapping points of the point cloud is the maximum number of overlapping points of the point cloud; if so, use the collapse and fusion result corresponding to the maximum number of overlapping points of the point cloud as the point cloud stitching result.

[0081] You can refer to here Figure 3 The fusion and stitching process of each point cloud to be stitched based on the collapsed triangle is specifically described. Figure 3 Schematic diagram of fusion stitching based on collapsed triangles.

[0082] exist Figure 3 In the figure, the red dot is the point cloud centroid of each point cloud to be spliced, and a collapsed triangle is constructed with the point cloud centroid as the vertex. The green arc represents the point cloud to be spliced, and the orange arrow is the Z-axis direction of the point cloud coordinate axis (the direction of the point cloud normal). Collapse is performed according to the Z-axis direction of the point cloud coordinate axis corresponding to each point cloud to be spliced.

[0083] During the collapse process, the left and right edges of each point cloud to be spliced ​​will overlap each other as they approach each other. When the number of overlapping points in the point cloud reaches the maximum, if the collapse continues, the number of overlapping points in the point cloud will decrease. Therefore, the preset collapse stop condition can be set to the maximum number of overlapping points.

[0084] In addition, the unit collapse step size during the collapse process can also be set, for example, the step size is set to 0.1. Then, each time the collapse occurs, all points in each point cloud to be spliced ​​move forward one step size in the Z-axis direction of their respective point cloud coordinate axes, and the number of overlapping points M at this time is counted.

[0085] Since the points in each point cloud to be spliced ​​are discrete, they may not coincide precisely. Therefore, a criterion for judg- ing overlapping points can be pre-set: assuming that the coordinates of two points are (x1, y1, z1) and (x2, y2, z2), when the distance between the two points satisfies the coincidence point judgment formula, the two points are considered to coincide. The overlap point judgment formula is expressed as follows:

[0086]

[0087] The size of the parameter Th is related to the scanning accuracy, and the user can customize the setting. For example, the two closest points in the point cloud of the viewpoint to be stitched can be first determined, then the distance L between the two points can be calculated, and finally Th=1 / 3L can be defined.

[0088] In the specific implementation, each point cloud to be spliced ​​is fused and spliced ​​along the Z-axis direction of its respective point cloud coordinate axis based on the collapsed triangle, and the fusion and splicing is stopped when the number of overlapping points M of the point cloud reaches the maximum. The point cloud splicing result obtained at this time is determined as the final point cloud splicing result, that is, the three-dimensional point cloud model of the inspected object is obtained.

[0089] In this embodiment, the original point cloud image is transposed through a preset conversion rule, so that the original point cloud image is adaptively adjusted, and then the point cloud is fused and spliced ​​by obtaining the centroid of the point cloud to construct a collapsed triangle and setting a stop condition. Compared with the existing method, the relative position relationship between each point cloud image can be accurately determined independently of a large number of overlapping points, thereby achieving effective splicing between point clouds with low overlap rate and non-concentrated overlapping areas, and obtaining a high-density three-dimensional point cloud model with high accuracy on non-spliced ​​surfaces.

[0090] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 4 , Figure 4 This is a flow chart of the second embodiment of the multi-view point cloud fusion and stitching method of the present application.

[0091] In this embodiment, when scanning the inspected object at different viewing angles, it is difficult to avoid vibration of the inspected object during rotation. Therefore, in order to achieve high-precision registration and stitching, before step S20, the following steps are further included:

[0092] Step S01: aligning the point clouds to be spliced ​​according to a preset alignment method to obtain aligned point clouds to be spliced, wherein the preset alignment method is alignment based on a Y-axis reference point and / or alignment based on an X-axis reference point.

[0093] It should be noted that the method based on the Y-axis reference point alignment can be a process of aligning the point clouds to be spliced ​​corresponding to each perspective using the center point of the point cloud's center of gravity as the Y-axis reference point; the method based on the X-axis reference point alignment can be a process of aligning the point clouds to be spliced ​​corresponding to each perspective using the center point in the X-axis direction as the X-axis reference point.

[0094] The display interface of the stitching device can provide the user with the above two preset alignment options, so that the user can determine whether to use the point cloud Y-axis reference point alignment and / or the point cloud X-axis reference point alignment.

[0095] Accordingly, step S20 specifically includes:

[0096] Step S200: obtaining the point cloud centroids of the aligned point clouds to be spliced, and constructing collapsed triangles according to the point cloud centroids.

[0097] Specifically, after aligning the point clouds to be spliced ​​according to the preset alignment method, iterative filtering can be performed on the point clouds to be spliced ​​corresponding to each perspective, thereby reducing noise interference caused by abnormal discrete points. After iterative filtering, the center of gravity is solved and the collapsed triangle is constructed.

[0098] Furthermore, when the Y-axis reference point alignment method is adopted, step S01 specifically includes: steps S001 to S002:

[0099] Step S001: Calculate the point cloud gravity center corresponding to each of the point clouds to be spliced ​​according to the gravity center solution formula, and determine the point cloud gravity center point according to the gravity center of each of the point clouds.

[0100] It can be understood that the centroid solution formula can be a calculation formula for the average position of all points in each point cloud to be spliced. Taking the solution of the point cloud centroid of the point cloud to be spliced ​​corresponding to a certain perspective as an example, the point cloud centroid (x c ,y c ,z c ) is calculated as:

[0101]

[0102] Where N is the total number of points in the point cloud to be spliced, (x i ,y i ,z i ) is the coordinate of the i-th point in the point cloud to be spliced.

[0103] Therefore, based on the calculation formula of the point cloud center of gravity, we can use (x c_L ,y c_L ,z c_L )、(x c_C ,y c_C ,z c_C )、(x c_R ,y c_R ,z c_R ) represents the point cloud centroid of the middle view point cloud to be stitched, the left view point cloud to be stitched, and the right view point cloud to be stitched.

[0104] It should be noted that, since the object to be inspected needs to be rotated during the scanning process of the profilometer, the object to be inspected may have an error on the Y-axis of the world coordinate system during the rotation process, that is, the Y-axis origin of the point cloud of the point cloud itself corresponding to the point cloud image of each viewpoint may not be consistent. In order to achieve consistency in the Y-axis direction of each point cloud to be spliced, the center of gravity determined by the center of gravity of each point cloud to be spliced ​​can be used as the Y-axis reference point for point cloud alignment. The center of gravity of the point cloud is the center point of the projection of the center of gravity of each point cloud to be spliced ​​on the Y-axis, that is, the center point of the Y-axis projection of the center of gravity of the point cloud.

[0105] Specifically, the center of gravity solution formula can be used to obtain the center of gravity coordinates y of each point cloud to be spliced c_LCR The formula for solving the center of gravity is as follows:

[0106]

[0107] Among them, y c_C ,y c_L ,y c_R They are the Y coordinates of the center of gravity of the middle view point cloud to be stitched, the left view point cloud to be stitched, and the right view point cloud to be stitched in the world coordinate system.

[0108] Step S002: taking the center point of the point cloud's centroid as a Y-axis reference point, and aligning each of the point clouds to be spliced ​​based on the Y-axis reference point to obtain aligned point clouds to be spliced.

[0109] It should be understood that the following Y-axis reference alignment formulas can be used to align the point clouds to be spliced, respectively. The Y-axis reference alignment formulas are:

[0110] Left view point cloud to be stitched: x Li ” = x Li ',y Li ”=y Li '-y c_L +y c_LCR ,z Li ”=z Li ';

[0111] Point cloud of view to be stitched: x Ci ” = x Ci ',y Ci ”=y Ci '-y c_C +y c_LCR ,z Ci ”=z Ci ';

[0112] Right view point cloud to be stitched: x Ri ” = x Ri ',y Ri ”=y Ri '-y c_R +y c_LCR ,z Ri ”=z Ri ';

[0113] In a specific implementation, the Y-axis reference alignment formula is used to correct the point clouds to be spliced ​​at each perspective. After correction, the coordinates of the i-th point in the middle perspective point cloud to be spliced, the left perspective point cloud to be spliced, and the right perspective point cloud to be spliced ​​can be expressed as (x Li ”,yLi ”,z Li ”)、(x Ci ”,y Ci ”,z Ci ”)、(x Ri ”,y Ri ”,z Ri ”). At this time, the point cloud coordinate systems corresponding to the point clouds to be spliced ​​in each perspective have the same Y-axis reference point y c_LCR .

[0114] This embodiment aligns the point clouds to be spliced ​​based on the Y-axis reference point before constructing the collapsed triangle, which can ensure that the point clouds to be spliced ​​in the three perspectives have the same Y-axis reference point in the Y-axis direction of the world coordinate system, thereby improving the stitching accuracy and consistency when performing point cloud fusion stitching. Compared with the traditional multi-perspective point cloud stitching method that relies on obtaining point clouds without moving the object, this embodiment can independently process the point clouds to be spliced ​​in the three perspectives of left, middle, and right, and can reduce the Y-axis movement error caused by the vibration of the object when rotating the object during the scanning process, and does not have high requirements for the initial alignment accuracy.

[0115] Based on the first and second embodiments of the present application, in the third embodiment of the present application, the same or similar contents as those of the first and second embodiments can be referred to the above description, and will not be described in detail later. Figure 5 , Figure 5 This is a flow chart of the third embodiment of the multi-view point cloud fusion and stitching method of the present application.

[0116] In this embodiment, it is further considered that when the object is rotated during the scanning process, there may be an error in the X-axis of the world coordinate system of the inspected object, that is, the X-axis midpoints of the point cloud coordinate system of the point cloud corresponding to each viewpoint point cloud image may not be consistent. When the alignment method based on the X-axis reference point is adopted, step S01 specifically includes: steps S001'~S004':

[0117] Step S001 ′: determining the first edge point and the second edge point of each of the point clouds to be spliced ​​in the X-axis direction.

[0118] It should be noted that the first edge point and the second edge point may be corresponding points of the maximum value and the minimum value of each to-be-joined point cloud in the X-axis direction.

[0119] Step S002': respectively determining whether the first edge point and the second edge point satisfy a preset abnormal isolated point condition.

[0120] It is understandable that the preset abnormal isolated point condition is used to remove abnormal points in the point cloud to be spliced. The basic idea of ​​removing abnormal points is that the points in the point cloud to be spliced ​​should be continuous and relatively uniform. Even in the cross section, they should be continuous in a certain direction and no independent points will appear. Therefore, the abnormal point judgment process is as follows:

[0121] First, for any point (x i ,y i ,z i ), i = 1, 2, 3, ..., N, find the distance D between this point and the remaining N-1 points i-j (j=1,2,3...N and j is not equal to i), keep D i-j The minimum value among the , that is, the distance to the nearest neighbor of the i-th point, is expressed as D i express.

[0122] Next, the nearest point distance mean calculation formula and standard deviation calculation formula can be used to calculate the nearest point distance mean μ corresponding to the point cloud to be spliced D and standard deviation σ D .

[0123] The calculation formula for the mean distance of the nearest neighbor point is:

[0124] The formula for calculating standard deviation is:

[0125] Finally, if the nearest point of the i-th point is at a distance D i When the following preset abnormal isolated point condition judgment formula is met, the point is regarded as an abnormal point and removed from the point cloud to be spliced.

[0126] The preset abnormal isolated point condition judgment formula is: |D i -μ D |>σ D .

[0127] Accordingly, when judging the outliers of the first edge point and the second edge point, it can specifically include: first, respectively calculating the nearest point distances corresponding to each point cloud to be spliced; the mean and the standard deviation; then obtaining the nearest point distances of the first edge point and the second edge point; and finally judging whether the difference between the nearest point distance and the mean of the nearest point distance is greater than the standard deviation.

[0128] Step S003': if not, obtaining the X-axis midpoints corresponding to the point clouds to be spliced ​​according to the first edge points and the second edge points, and determining the X-axis center point according to the X-axis midpoints.

[0129] Taking the solution process of the X-axis midpoint of the point cloud to be spliced ​​corresponding to a certain perspective as an example, when the first edge point and the second edge point do not meet the preset abnormal isolated point condition, that is, the first edge point and the second edge point are not abnormal points, if the X coordinate of the first edge point is x max , the X coordinate of the second edge point is x min , then the midpoint of the X-axis can be expressed as

[0130] Therefore, we can use x mea_L 、x mea_C 、x mea_R Indicates the X-axis midpoint of the center view point cloud to be stitched, the left view point cloud to be stitched, and the right view point cloud to be stitched.

[0131] It should be noted that in order to achieve consistency in the X-axis direction for each point cloud to be spliced, that is, to ensure that the X-axis midpoints of the point clouds corresponding to the point cloud images of each viewpoint are consistent, the X-axis center point determined by the X-axis midpoints of each point cloud to be spliced ​​can be used as the X-axis reference point for point cloud alignment.

[0132] Specifically, the X-axis center point solution formula can be used to obtain the center point of the X-axis midpoint of each point cloud to be spliced ​​in the X-axis direction, that is, the X-axis center point coordinate x mea_LCR The solution formula for the center point in the X-axis direction is as follows:

[0133]

[0134] Among them, x mea_L 、x mea_C 、x mea_R They are the X coordinates of the midpoints of the center view point cloud to be stitched, the left view point cloud to be stitched, and the right view point cloud to be stitched in the world coordinate system.

[0135] Step S004 ′: taking the midpoint of the X-axis direction as an X-axis reference point, and aligning the point clouds to be spliced ​​based on the X-axis reference point to obtain aligned point clouds to be spliced.

[0136] It should be understood that the following X-axis reference alignment formulas can be used to align the point clouds to be spliced, respectively. The X-axis reference alignment formulas are:

[0137] Left view point cloud to be stitched: x Li ”'=x Li ”-x mea_L +x mea_LCR ,y Li ”'=y Li ”,z Li ”'=z Li ”;

[0138] Point cloud of view to be stitched: x Ci ”'=x Ci ”-x mea_C +x mea_LCR ,y Ci ”'=y Ci ”,z Ci ”'=z Ci ”;

[0139] Right view point cloud to be stitched: x Ri ”'=x Ri ”-x mea_R +x mea_LCR ,y Ri ”'=y Ri ”,z Ri ”'=z Ri ”;

[0140] In a specific implementation, the above X-axis reference alignment formula is used to correct the point clouds to be spliced ​​in each perspective. After correction, the coordinates of the i-th point in the middle perspective point cloud to be spliced, the left perspective point cloud to be spliced, and the right perspective point cloud to be spliced ​​can be expressed as (x Li ″′,yLi″′,z Li ″′)、(x Ci ″′,y Ci ″′,z Ci ″′)、(x Ri ″′,y Ri ″′,z Ri ″′). At this time, the point cloud coordinate systems corresponding to the point clouds to be spliced ​​in each perspective have the same X-axis reference point x mea_LCR .

[0141] It should be noted that, considering that the center point of the point cloud center of gravity determined based on the point cloud center of gravity is limited by the uneven density of the point cloud, that is, the Y-axis reference point calculated above may deviate from the geometric center of the object, this embodiment uses the X-axis reference point for point cloud alignment, and the X-axis reference point is the center point in the X-axis direction, which is the center point obtained by the widest side length of the point cloud to be spliced ​​from each perspective. Since the point clouds to be spliced ​​from each perspective have some overlapping areas, that is, the widths of the reference planes of each point cloud are the same, the use of the X-axis center point can ensure that the correspondence between the point clouds of each perspective in the subsequent splicing process is more accurate, further reducing the splicing error.

[0142] It should also be noted that, in actual applications, the splicing device may only use the alignment method based on the X-axis reference point or only use the alignment method based on the Y-axis reference point in response to the user's selection of the preset alignment method on the display interface, or may also use the alignment method based on the X-axis reference point at the same time as the alignment method based on the Y-axis reference point. When the alignment method based on the X-axis reference point and the alignment method based on the Y-axis reference point are used at the same time, the execution order of the two alignment methods is not restricted.

[0143] In this embodiment, before constructing the collapsed triangle, each point cloud to be spliced ​​is aligned based on the X-axis reference point, which can ensure that the point clouds to be spliced ​​in the three perspectives have the same X-axis reference point in the X-axis direction, thereby further improving the stitching accuracy and consistency when performing point cloud fusion stitching. Compared with the traditional multi-perspective point cloud stitching method that relies on obtaining point clouds when the object is stationary, since this embodiment can independently process the point clouds to be spliced ​​in the three perspectives of left, middle, and right, it can reduce the X-axis direction movement error caused by the vibration of the object when the object is rotated during the scanning process, and there is no high requirement for the initial alignment accuracy.

[0144] In addition, you can also refer to Figure 6 The full process of the multi-view point cloud fusion and stitching method of this application is described. Figure 6 This is a schematic diagram of the entire process of the multi-view point cloud fusion and stitching method of this application.

[0145] First, the user can import the original point cloud data of each view on the display interface of the splicing device, unify the format of the original point cloud data, and then display the corresponding original point cloud image in the display area of ​​each view in the display interface. Figure 7 As shown, Figure 7 This is a schematic diagram of the first operating scenario corresponding to the multi-view point cloud fusion and splicing method of this application. Figure 7 In the figure, from left to right, the red, green, and blue point cloud images are respectively the imported left-view point cloud image, middle-view point cloud image, and right-view point cloud image.

[0146] Next, in order to facilitate the subsequent stitching operation, the original point cloud images corresponding to each perspective can be visually rotated (display perspective conversion) based on the preset conversion rules, so that the XY plane of the original point cloud image observed by the user is rotated to the XZ plane, and the coordinate axis (world coordinate axis) provided to the user for observation by the display interface is synchronously transposed, thereby ensuring that the coordinates of the original point cloud points in each original point cloud image remain unchanged.

[0147] Then, the left-view point cloud image after the display perspective conversion is rotated 120 degrees around the Y axis of its corresponding point cloud coordinate axis at a negative angle, and the right-view point cloud image after the display perspective conversion is rotated 120 degrees around the Y axis of its corresponding point cloud coordinate axis at a positive angle, to ensure that the interlaced angle of the XY plane (reference plane) of each point cloud to be spliced ​​is 120 degrees. Figure 8 As shown, Figure 8 This is a schematic diagram of the second operating scenario corresponding to the multi-view point cloud fusion and stitching method of this application. Figure 8 The red, green, and blue point cloud images from left to right are the Figure 7 Each point cloud image in the image is converted by performing coordinate system transposition and point cloud rotation according to the above conversion rules.

[0148] Secondly, the point clouds to be spliced ​​are aligned using the center point of the point cloud gravity obtained from the center of gravity of each point cloud as the Y-axis reference point, and the point clouds to be spliced ​​are aligned using the center point of the X-axis direction obtained from the midpoint of the X-axis as the X-axis reference point. The aligned point clouds to be spliced ​​are obtained.

[0149] Finally, the aligned point clouds to be spliced ​​are iteratively filtered, and an equilateral triangle (collapsed triangle) is constructed with the three point cloud centroids as vertices. The point clouds are collapsed and fused inside according to the Z-axis direction of the point cloud coordinate axis (the direction of the point cloud normal vector) of each point cloud to be spliced. When the preset collapse stop condition is met, the point cloud splicing result (3D point cloud model) is obtained. Fig. 9 As shown, Fig. 9 This is a schematic diagram of the third operating scenario corresponding to the multi-view point cloud fusion and stitching method of the present application method. Fig. 9 The larger view on the far right shows the fused 3D point cloud model.

[0150] The multi-view point cloud fusion and stitching method of this embodiment is based on the point cloud normal vector, and presets the vertex to construct a collapsed triangle, sets the collapse stop condition, and then performs a point cloud fusion and stitching algorithm that collapses inward to the center point. It has good adaptability to situations where the point cloud overlap rate is low and the overlapping area is not concentrated. And only the center of gravity of the point cloud is required as the vertex of the triangle, and the interlaced angle of the point cloud reference plane is 120 degrees, and there is no high requirement for the initial alignment accuracy. Based on the multi-view point cloud fusion and stitching method of this embodiment, only one profilometer can be used to construct a three-dimensional point cloud model for the inspected object (such as a rod, etc.) that moves during the processing and has a large dead weight.

[0151] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the multi-view point cloud fusion and stitching method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0152] In addition, the present application also provides a multi-view point cloud fusion and splicing device, referring to Fig.10 , Fig.10 This is a structural block diagram of the first embodiment of the multi-view point cloud fusion and splicing device of the present application; Fig.10 As shown, the device comprises:

[0153] The point cloud loading module 1001 is used to obtain the original point cloud images corresponding to each viewing angle, and perform transposition transformation on each of the original point cloud images based on a preset transformation rule to determine the point cloud to be spliced;

[0154] The splicing preparation module 1002 is used to obtain the point cloud centroids of each of the point clouds to be spliced, and construct collapsed triangles according to the point cloud centroids;

[0155] The result generation module 1003 is used to merge and splice the point clouds to be spliced ​​based on the collapsed triangles, and obtain the point cloud splicing result when a preset collapse stop condition is met.

[0156] Furthermore, the stitching preparation module 1002 is also used to align each of the point clouds to be stitched according to a preset alignment method to obtain each aligned point cloud to be stitched, and the preset alignment method is based on the Y-axis reference point alignment and / or based on the X-axis reference point alignment; obtain the point cloud centroid of each of the aligned point clouds to be stitched, and construct a collapsed triangle based on the centroid of each point cloud.

[0157] Furthermore, the stitching preparation module 1002 is also used to calculate the point cloud centroid corresponding to each of the point clouds to be stitched according to the centroid solution formula, and determine the point cloud centroid center point according to the point cloud centroid; use the point cloud centroid center point as the Y-axis reference point, and align each of the point clouds to be stitched based on the Y-axis reference point to obtain the aligned point clouds to be stitched.

[0158] Furthermore, the stitching preparation module 1002 is also used to determine the first edge point and the second edge point of each of the point clouds to be stitched in the X-axis direction; respectively determine whether the first edge point and the second edge point meet the preset abnormal isolated point condition; if not, obtain the X-axis midpoint corresponding to each of the point clouds to be stitched according to the first edge point and the second edge point, and determine the X-axis center point according to each of the X-axis midpoints; use the X-axis center point as the X-axis reference point, and align each of the point clouds to be stitched based on the X-axis reference point to obtain the aligned point clouds to be stitched.

[0159] Furthermore, the stitching preparation module 1002 is also used to respectively calculate the mean and standard deviation of the nearest point distances corresponding to each of the point clouds to be stitched; obtain the nearest point distances of the first edge point and the second edge point; and determine whether the difference between the nearest point distance and the mean of the nearest point distance is greater than the standard deviation.

[0160] Furthermore, the stitching preparation module 1002 is also used to collapse and fuse each of the point clouds to be stitched along the Z-axis direction based on the vertices of the collapsed triangle; determine the number of overlapping points of the current point cloud according to the collapse and fusion result, and judge whether the current number of overlapping points of the point cloud is the maximum number of overlapping points of the point cloud; if so, use the collapse and fusion result corresponding to the maximum number of overlapping points of the point cloud as the point cloud stitching result.

[0161] The original point cloud images corresponding to each viewing angle include: a middle viewing angle point cloud image, a left viewing angle point cloud image, and a right viewing angle point cloud image;

[0162] Furthermore, the point cloud loading module 1001 is also used to perform display perspective conversion on the middle perspective point cloud image, the left perspective point cloud image and the right perspective point cloud image respectively based on a preset observation perspective; determine the middle perspective point cloud to be spliced ​​according to the middle perspective point cloud image after the display perspective conversion; perform a first rotation transformation on the left perspective point cloud image after the display perspective conversion to determine the left perspective point cloud to be spliced; perform a second rotation transformation on the right perspective point cloud image after the display perspective conversion to determine the right perspective point cloud to be spliced.

[0163] This embodiment transposes the original point cloud image through preset conversion rules, so that the original point cloud image is adaptively adjusted, and then the point cloud is fused and spliced ​​by obtaining the center of gravity of the point cloud to construct a collapsed triangle and setting a stop condition. Compared with the existing method, the relative position relationship between each point cloud image can be accurately determined independently of a large number of overlapping points, thereby achieving effective splicing of point clouds with low overlap rate and sparse overlapping area.

[0164] The present application also provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, the computer-readable program instructions being used to execute the multi-view point cloud fusion and stitching method in the above-mentioned embodiment.

[0165] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0166] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the multi-view point cloud fusion and stitching method as described above.

[0167] The computer program product provided by the present application can solve the technical problems corresponding to the multi-view point cloud fusion and stitching method. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the multi-view point cloud fusion and stitching method provided by the above embodiment, which will not be repeated here.

[0168] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other elements in the process, method, article or system including the element.

[0169] The serial numbers of the above-mentioned embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. They are only some embodiments of the present application and do not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the description and drawings of the present application under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A multi-view point cloud fusion and splicing method, characterized in that: The method comprises: Acquire the original point cloud images corresponding to each viewing angle, and perform transposition transformation on each of the original point cloud images based on a preset transformation rule to determine the point cloud to be spliced; Obtaining the point cloud centroid of each of the point clouds to be spliced, and constructing a collapsed triangle according to the point cloud centroids; The point clouds to be spliced ​​are fused and spliced ​​based on the collapsed triangles, and the point cloud splicing result is obtained when a preset collapse stop condition is met.

2. The method according to claim 1, characterized in that Before the step of obtaining the point cloud centroids of the point clouds to be spliced ​​and constructing collapsed triangles according to the point cloud centroids, the method further includes: Aligning the point clouds to be spliced ​​according to a preset alignment method to obtain the aligned point clouds to be spliced, wherein the preset alignment method is alignment based on a Y-axis reference point and / or alignment based on an X-axis reference point; Correspondingly, the step of obtaining the point cloud centroids of each of the point clouds to be spliced, and constructing collapsed triangles according to the point cloud centroids, includes: The point cloud centroids of the aligned point clouds to be spliced ​​are obtained, and collapsed triangles are constructed according to the point cloud centroids.

3. The method according to claim 2, characterized in that The step of aligning the point clouds to be spliced ​​according to a preset alignment method to obtain the aligned point clouds to be spliced ​​comprises: Calculate the point cloud gravity center corresponding to each of the point clouds to be spliced ​​according to the gravity center solution formula, and determine the point cloud gravity center point according to the gravity center of each of the point clouds; The center point of the center of gravity of the point cloud is used as a Y-axis reference point, and the point clouds to be spliced ​​are aligned based on the Y-axis reference point to obtain the aligned point clouds to be spliced.

4. The method according to claim 2, characterized in that The step of aligning the point clouds to be spliced ​​according to a preset alignment method to obtain the aligned point clouds to be spliced ​​further includes: Determine the first edge point and the second edge point of each of the point clouds to be spliced ​​in the X-axis direction; Determine whether the first edge point and the second edge point meet preset abnormal isolated point conditions respectively; If not, obtaining the X-axis midpoints corresponding to the point clouds to be spliced ​​according to the first edge point and the second edge point, and determining the center point in the X-axis direction according to the X-axis midpoints; The center point in the X-axis direction is used as an X-axis reference point, and each of the point clouds to be spliced ​​is aligned based on the X-axis reference point to obtain the aligned point clouds to be spliced.

5. The method according to claim 4, characterized in that The step of respectively judging whether the first edge point and the second edge point meet the preset abnormal isolated point condition comprises: Calculate the mean and standard deviation of the nearest point distances corresponding to each of the point clouds to be spliced; Obtain the distance between the first edge point and the nearest neighbor point of the second edge point; It is determined whether the difference between the nearest neighbor distance and the mean of the nearest neighbor distances is greater than the standard deviation.

6. The method according to claim 1, characterized in that The step of fusing and splicing the point clouds to be spliced ​​based on the collapsed triangles and obtaining the point cloud splicing result when a preset collapse stop condition is met includes: Based on each vertex of the collapsed triangle, collapse and fuse each of the point clouds to be spliced ​​along the Z-axis direction; Determine the number of overlapping points of the current point cloud according to the collapse fusion result, and judge whether the number of overlapping points of the current point cloud is the maximum number of overlapping points of the point cloud; If so, the collapsed fusion result corresponding to the maximum point cloud overlap number is used as the point cloud stitching result.

7. The method according to claim 1, characterized in that The original point cloud images corresponding to each viewing angle include: a middle viewing angle point cloud image, a left viewing angle point cloud image, and a right viewing angle point cloud image. The step of performing a transposition transformation on each of the original point cloud images based on a preset transformation rule to determine a point cloud to be spliced ​​includes: Based on a preset observation perspective, display perspective conversion is performed on the middle perspective point cloud image, the left perspective point cloud image, and the right perspective point cloud image respectively; Determine the mid-view point cloud to be spliced ​​according to the mid-view point cloud image after the display view is converted; Performing a first rotation transformation on the left-view point cloud image after the display perspective conversion to determine the left-view point cloud to be spliced; A second rotation transformation is performed on the right view point cloud image after the display view conversion to determine the right view point cloud to be spliced.

8. A multi-view point cloud fusion and splicing device, characterized in that: The device comprises: A point cloud loading module is used to obtain the original point cloud images corresponding to each viewing angle, and perform a transposition transformation on each of the original point cloud images based on a preset transformation rule to determine the point cloud to be spliced; A splicing preparation module, used for obtaining the point cloud centroids of each of the point clouds to be spliced, and constructing collapsed triangles according to the point cloud centroids; The result generating module is used to merge and splice the point clouds to be spliced ​​based on the collapsed triangles, and obtain the point cloud splicing result when a preset collapse stop condition is met.

9. A storage medium, characterized in that: The storage medium stores a multi-view point cloud fusion and stitching program, and when the multi-view point cloud fusion and stitching program is executed by the processor, the steps of the multi-view point cloud fusion and stitching method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the multi-view point cloud fusion and stitching method according to any one of claims 1 to 7 are implemented.

Images