Projection-based 3D point cloud splicing method and computer equipment
Through the projection-based 3D point cloud splicing method, the AABB bounding box and kd-tree search tree fusion overlapping vertices is solved, and the redundancy problem of point cloud data in traditional methods is achieved efficient point cloud splicing and deduplication are significantly improved. Splicing speed and efficiency.
Patent Information
- Application Number
- CN202410981171.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-22
AI Technical Summary
Due to the limitation of viewing angle, traditional three-dimensional scanning devices can only obtain a limited range of point cloud data at a time, which makes it impossible for test objects exceeding their range to obtain complete point cloud data at once. The existing 3D point cloud splicing method leads to excessive point cloud data. redundancy.
Using the projection-based 3D point cloud splicing method, by obtaining the source point cloud and target point cloud, performing direct-pass filtering and overlapping area detection, using the AABB bounding box and the kd-tree search tree to construct the overlapping point pair index set, fuse the overlapping vertices, and reduce the Redundant data.
The splicing speed of point clouds is improved, the residual overlapping vertices in the spliced point cloud are reduced, and the deduplication in the overlapping area of point clouds is achieved, which significantly improves the splicing efficiency.
Smart Images

Figure CN118941444A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of image data processing, and more specifically, to a projection-based 3D point cloud stitching method, a computer device, and a computer-readable storage medium. Background Art
[0002] With the widespread application of machine vision in various industries, there is an increasing demand for large-scale, high-precision positioning, measurement, and scene stitching. Traditional 3D scanning equipment can only obtain point clouds within a certain range at a time due to its viewing angle limitation. For test objects that exceed its range, it is often impossible to obtain complete point cloud data at one time.
[0003] To address this problem, there are currently two main methods for 3D point cloud stitching: one is to use a multi-camera vision system to obtain multiple sets of point cloud data, and then rotate and translate the multiple sets of point cloud data to a unified coordinate system to form complete target point cloud data; the other is to use a scanning device to take pictures at multiple angles, and then rotate and translate the multiple sets of point cloud data to a unified coordinate system.
[0004] Both methods can obtain complete target point clouds, but due to the overlapping of camera fields of view, the obtained point cloud data is too redundant. Summary of the invention
[0005] In order to solve the above problems, the present invention provides a projection-based 3D point cloud stitching method, computer equipment, computer-readable storage medium and computer program product, which will increase the speed of point cloud stitching and greatly reduce residual overlapping vertices after stitching.
[0006] To achieve the above object, according to a first aspect of the present invention, a projection-based 3D point cloud stitching method is provided, the method comprising:
[0007] Get source point cloud and target point cloud;
[0008] Performing through filtering on the source point cloud and the target point cloud to obtain filtered source point cloud and target point cloud;
[0009] Detect the overlapping area of the filtered source point cloud and the target point cloud, project the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the Xoy plane, and construct an overlapping point pair index set;
[0010] According to the overlapping point pair index set, the overlapping vertices in the filtered source point cloud and the target point cloud are merged;
[0011] The vertex-fused source point cloud is spliced onto the vertex-fused target point cloud.
[0012] Furthermore, the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area are projected to the Xoy plane, and an overlapping point pair index set is constructed, including calculating the bounding box of the overlapping area according to the minimum point coordinates and the maximum point coordinates of the bounding box of the filtered source point cloud and the minimum point coordinates and the maximum point coordinates of the bounding box of the filtered target point cloud; the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area are projected to the Xoy plane, and an overlapping point pair index set is constructed.
[0013] Furthermore, the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area are projected to the Xoy plane, and an overlapping point pair index set is constructed, including respectively detecting the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area, and recording the sets of vertex serial numbers as the source vertex set and the target vertex set respectively; projecting the vertices of the filtered target point cloud that fall into the bounding box of the overlapping area to the Xoy plane, constructing the target overlapping area point cloud, and constructing a kd-tree search tree based on the target overlapping area point cloud; obtaining a set of overlapping point pairs through the source vertex set and the kd-tree search tree; constructing an overlapping point pair index set according to the correspondence between the vertex serial numbers in the target overlapping area point cloud and the vertex serial numbers in the target vertex set, as well as the overlapping point pairs.
[0014] Furthermore, the vertices of the filtered target point cloud that fall within the bounding box of the overlapping area are projected to the Xoy plane to construct the target overlapping area point cloud, and a kd-tree search tree is constructed based on the target overlapping area point cloud, including projecting the vertices of the filtered target point cloud that fall within the bounding box of the overlapping area to the Xoy plane to construct the target overlapping area point cloud, recording the correspondence between the vertex serial numbers in the target overlapping area point cloud and the vertex serial numbers in the target vertex set; and constructing a kd-tree search tree based on the target overlapping area point cloud.
[0015] Furthermore, a set of overlapping point pairs is obtained through the source vertex set and the kd-tree search tree, including traversing the vertices belonging to the source vertex set in the filtered source point cloud, and projecting the vertices belonging to the source vertex set in the filtered source point cloud to the xoy plane; searching the point in the target overlapping area point cloud that is closest to the projection point of the vertices belonging to the source vertex set in the filtered source point cloud on the xoy plane through the kd-tree search tree, and recording it as the closest point; if the distance is less than the set distance threshold, then the serial number of the closest point and the corresponding vertex belonging to the source vertex set in the filtered source point cloud is recorded to obtain a set of overlapping point pairs.
[0016] Furthermore, the overlapping area of the filtered source point cloud and the target point cloud is detected, including detecting the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud; the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud are detected for intersection, and if the two bounding boxes intersect, it is determined that the filtered source point cloud and the target point cloud have an overlapping area.
[0017] Furthermore, according to the overlapping point pair index set, the overlapping vertices in the filtered source point cloud and the target point cloud are fused, including taking the vertices of the filtered source point cloud that fall within the bounding box of the overlapping area as the first vertices, and taking the vertices of the filtered target point cloud that fall within the bounding box of the overlapping area as the second vertices, and the serial numbers of the first vertex and the second vertex belong to a pair of overlapping point pairs in the overlapping point pair index set; calculating the coordinate value of the fused vertex according to the coordinate value of the first vertex and the coordinate value of the second vertex; replacing the coordinate value of the first vertex and the coordinate value of the second vertex with the coordinate value of the fused vertex, and obtaining the source point cloud and target point cloud after the vertex fusion.
[0018] Furthermore, the vertex-fused source point cloud is spliced onto the vertex-fused target point cloud to obtain a spliced point cloud, including traversing each vertex in the vertex-fused source point cloud. If the index of the vertex belongs to the overlapping point pair index set, no processing is performed; otherwise, the vertex is added to the set of serial numbers of the vertices in the bounding box of the filtered target point cloud that fall into the overlapping area to obtain a spliced point cloud.
[0019] According to a second aspect of the present invention, there is also provided a computer device, which includes a memory, a processor and a computer program stored in the memory, and the processor executes the computer program to implement the steps of any one of the above methods.
[0020] According to a third aspect of the present invention, there is also provided a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.
[0021] According to a fourth aspect of the present invention, there is also provided a computer program product, comprising a computer program, which implements the steps of any of the above methods when executed by a processor.
[0022] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:
[0023] The present invention provides a projection-based 3D point cloud stitching method, which can quickly detect the overlapping area of the source point cloud and the target point cloud through the AABB bounding box; the vertices of the filtered source point cloud and the target point cloud falling into the bounding box of the overlapping area are projected to the xoy plane, and an overlapping point pair index set is constructed to overcome the influence of the point cloud's own z-direction accuracy and improve the speed of point cloud stitching; the overlapping vertices in the filtered source point cloud and the target point cloud are fused according to the overlapping point pair index set, and the source point cloud after the vertex fusion is spliced to the target point cloud after the vertex fusion to obtain the spliced point cloud, thereby improving the speed of point cloud splicing, and realizing the deduplication of overlapping vertices in the overlapping area of the point cloud, and greatly reducing the residual overlapping vertices in the spliced point cloud. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0025] Figure 1 A schematic diagram of a process of a projection-based 3D point cloud stitching method provided in an embodiment of the present application;
[0026] Figure 2 A schematic diagram of a flow chart of the steps of constructing an overlapping point pair index set provided in an embodiment of the present application;
[0027] Figure 3 A schematic diagram of a bounding box of an overlapping area provided in an embodiment of the present application;
[0028] Figure 4 A schematic diagram of the effect of the projection-based 3D point cloud stitching method provided in an embodiment of the present application;
[0029] Figure 5 A comparison diagram of the effects of a traditional point cloud stitching method and a projection-based 3D point cloud stitching method provided in an embodiment of the present application;
[0030] Figure 6 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0031] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0032] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0033] like Figure 1 As shown, a projection-based 3D point cloud stitching method is provided, which can be executed by a terminal or by a server communicating with the terminal through a network. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The server can be an independent server, or a server cluster composed of multiple servers can be used. Taking the method applied to a terminal as an example, the method includes the following steps:
[0034] Step S1, obtaining source point cloud and target point cloud.
[0035] Among them, both the source point cloud and the target point cloud are 3D point clouds, and both include multiple vertices.
[0036] For example, the terminal obtains the source point cloud C in the same coordinate system through a 3D scanning device. src and target point cloud C dest .
[0037] Step S2, performing through filtering on the source point cloud and the target point cloud to obtain filtered source point cloud and target point cloud.
[0038] For example, the terminal receives the source point cloud C src and target point cloud C dest Perform straight-through filtering to remove background point clouds. Set the filter axis to the z direction and the threshold range to (z min ,z max ), cut out the points whose z coordinates do not fall within the threshold range, retain the points whose z coordinates fall within the threshold range, and obtain the filtered point cloud. The filtered source point cloud is recorded as B src , the filtered target point cloud is recorded as B dest .
[0039] Step S3, detecting the overlapping area of the filtered source point cloud and the target point cloud, projecting the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the Xoy plane, and constructing an overlapping point pair index set.
[0040] The bounding box of the overlapping area is the AABB bounding box of the overlapping area.
[0041] Exemplarily, the terminal detects the filtered source point cloud B src and target point cloud B dest Whether there is an overlapping area, if an overlapping area is detected, determine the bounding box of the overlapping area; project the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the xoy plane, and construct the overlapping point pair index set O I .
[0042] Step S4, merging the overlapping vertices in the filtered source point cloud and the target point cloud according to the overlapping point pair index set.
[0043] Step S5, splicing the vertex-fused source point cloud to the vertex-fused target point cloud.
[0044] In the above-mentioned projection-based 3D point cloud stitching method, the overlapping area of the source point cloud and the target point cloud can be quickly detected through the AABB bounding box; the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area are projected to the xoy plane, and an overlapping point pair index set is constructed to overcome the influence of the point cloud's own z-direction accuracy and improve the speed of point cloud stitching; the overlapping vertices in the filtered source point cloud and the target point cloud are fused according to the overlapping point pair index set, and the source point cloud after the vertex fusion is spliced to the target point cloud after the vertex fusion to obtain the spliced point cloud, which improves the speed of point cloud stitching, and realizes the deduplication of overlapping vertices in the overlapping area of the point cloud, reducing the residual overlapping vertices in the spliced point cloud.
[0045] In one embodiment, Figure 2 As shown, the above step S3 detects the overlapping area of the filtered source point cloud and the target point cloud, projects the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the xoy plane, and constructs an overlapping point pair index set, including:
[0046] Step S301 , detecting the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud.
[0047] The bounding box of the filtered source point cloud is the AABB bounding box of the filtered source point cloud, and the bounding box of the filtered target point cloud is the AABB bounding box of the filtered target point cloud.
[0048] Exemplarily, the terminal detects the filtered source point cloud B src The bounding box S src And the filtered target point cloud B dest The bounding box S dest , record the bounding box S src The minimum point coordinates Maximum point coordinates and bounding box S dest The minimum point coordinates Maximum point coordinates
[0049] Step S302 , performing intersection detection on the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud. If the two bounding boxes intersect, it is determined that there is an overlapping area between the filtered source point cloud and the target point cloud.
[0050] Exemplarily, the terminal bounding box S src and bounding box S dest Perform intersection detection. If the center distance between the two bounding boxes is d c In a certain coordinate axis direction, it is smaller than the sum of the radii of the two bounding boxes d R , then the two bounding boxes intersect, that is, the filtered source point cloud B is determined src and target point cloud B dest There are overlapping areas.
[0051] The following describes the bounding box S src and bounding box S dest Computational steps for intersection detection.
[0052] Step S30201, calculate the bounding box S respectively src and bounding box S dest The center coordinates of:
[0053]
[0054] Step S30202, calculate the bounding box S respectively src and bounding box S dest Radius:
[0055]
[0056] Step S30203, calculate the bounding box S src and bounding box S dest The center distance d c :
[0057]
[0058] Step S30204, calculate the bounding box S src and bounding box S dest The sum of the radii d R :
[0059]
[0060] Step S30205, if dc In the direction of a certain coordinate axis, it is smaller than d R , then it is determined that the two bounding boxes intersect, that is, B src and B dest If there is an overlapping area, continue to execute step S303; otherwise, it is determined that the two bounding boxes do not intersect, and execute step S5.
[0061] In this embodiment, by detecting the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud and performing intersection detection on the two bounding boxes, fast detection of the overlapping area of the filtered source point cloud and the target point cloud is achieved.
[0062] Step S303 , calculating the bounding box of the overlapping area according to the minimum point coordinates and the maximum point coordinates of the bounding box of the filtered source point cloud and the minimum point coordinates and the maximum point coordinates of the bounding box of the filtered target point cloud.
[0063] Exemplarily, step S303 includes step S30301, determining the minimum value of the bounding box of the overlapping area, that is, determining the bounding box S src The minimum point coordinates and bounding box S dest The minimum point coordinates The larger value of x, y, and z coordinates is denoted by Step S30302, determine the maximum value of the bounding box of the overlapping area, that is, determine the bounding box S src The maximum point coordinates and bounding box S dest The maximum point coordinates The smaller value of x, y, and z coordinates is denoted by Step S30303, set the threshold γ, determine and is the bounding box of the overlapping area of vertices, which can be divided into B src and B dest The overlapping area of Figure 3 As shown, the AABB bounding box of the overlapping area is the box marked by the purple frame line in the figure, and Figure 3 The left side shows the original image of the AABB bounding box of the overlapping area, and the right side shows the enlarged AABB bounding box of the overlapping area.
[0064] Step S304, projecting the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the xoy plane, and constructing an overlapping point pair index set.
[0065] In one embodiment, the above step S304, projecting the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the xoy plane, and constructing an overlapping point pair index set, includes the following steps:
[0066] Step S30401, respectively detect the filtered source point cloud B src and target point cloud B dest The vertices that fall into the bounding box of the overlapping area are recorded, and the serial number (ID) of the vertex is recorded as the source vertex set. and the target vertex set
[0067] Step S30402: filter the target point cloud B dest The vertices falling into the bounding box of the overlapping area are projected onto the Xoy plane to construct the target overlapping area point cloud, and a kd-tree search tree is constructed based on the target overlapping area point cloud.
[0068] Exemplarily, step S30402 includes step S3040201, setting the target vertex set The vertex corresponding to the vertex id in is taken out and projected onto the xoy plane (that is, the filtered target point cloud B dest The vertices falling into the bounding box of the overlapping area are projected onto the xoy plane, that is, the z coordinate of the vertex is set to 0, and the point cloud of the target overlapping area is constructed. Record The vertex id and target vertex set in The correspondence between the vertex ids in Step S3040202: based on the target overlapping area point cloud Construct a kd-tree search tree.
[0069] Step S30403, set the distance threshold λ, and traverse B src belongs to the source vertex set The vertex (or B src Vertices that fall within the bounding box of the overlapping area) The vertex Projection to the xoy plane (that is, the filtered source point cloud B src The vertices that fall into the bounding box of the overlapping area are projected onto the xoy plane. The z coordinate value is 0; the target overlapping area point cloud is found through the kd-tree search tree The vertex The point with the shortest distance (denoted as the shortest distance d) between the projection points on the xoy plane is denoted as the closest point If the shortest distance d<λ, then record the closest point And the corresponding filtered source point cloud B src The serial numbers of the vertices that fall into the bounding box of the overlapping area. These two serial numbers constitute a set of overlapping point pairs.
[0070] according to and Get the filtered source point cloud B src and target point cloud B dest All the overlapping points of All overlapping point pairs constitute the overlapping point pair index set O I .
[0071] In one embodiment, the above step S4, fusing the overlapping vertices in the filtered source point cloud and the target point cloud according to the overlapping point pair index set, includes:
[0072] Step S401, take the vertex of the filtered source point cloud that falls within the bounding box of the overlapping area as the first vertex, and take the vertex of the filtered target point cloud that falls within the bounding box of the overlapping area as the second vertex, and the serial numbers of the first vertex and the second vertex belong to a pair of overlapping point pairs in the overlapping point pair index set; calculate the coordinate value of the fused vertex based on the coordinate value of the first vertex and the coordinate value of the second vertex.
[0073] Exemplarily, the coordinate value of the fused vertex is calculated based on the coordinate value of the first vertex and the coordinate value of the second vertex, which can be a weighted average or arithmetic mean of the coordinate value of the first vertex and the coordinate value of the second vertex, and the weighted average or arithmetic mean is used as the coordinate value of the fused vertex.
[0074] In other embodiments, step S401 may also be to obtain several vertices in the neighborhood of the overlapping point pair, calculate a weighted average or arithmetic mean based on the several vertices, and use the weighted average or arithmetic mean as the coordinate value of the fused vertex.
[0075] The following takes the arithmetic mean of the coordinate value of the first vertex and the coordinate value of the second vertex as an example to illustrate the specific steps of calculating the coordinate value of the fused vertex.
[0076] According to the following formula, calculate the coordinate value of the fused vertex
[0077]
[0078] in, and is the index of a pair of overlapping vertices, is the filtered source point cloud B src belongs to the source vertex set The vertices (filtered source point cloud B src The coordinate value of the vertex that falls into the bounding box of the overlapping area, that is, the first vertex, is the filtered target point cloud Bdest The target vertex set Vertex (filtered target point cloud B dest The coordinate value of the vertex that falls within the bounding box of the overlapping area, that is, the second vertex).
[0079] Step S402: Replace the coordinate values of the first vertex and the second vertex with the coordinate values of the fused vertex to obtain the source point cloud and the target point cloud after vertex fusion, that is:
[0080] In one embodiment, the above step S5, splicing the vertex-fused source point cloud to the vertex-fused target point cloud to obtain a spliced point cloud, includes the following steps:
[0081] Step S501, traverse the source point cloud B after vertex fusion src For each vertex v i , if the vertex v i The index i belongs to the overlapping point pair index set O I (i.e., satisfying i∈O I ), no processing is performed; otherwise, step S502 is executed;
[0082] Step S502: If the vertex v i The index i does not belong to the overlapping point pair index set O I (i.e. satisfy ), then the vertex v i Add to the target vertex set Get as Figure 4 The stitched point cloud shown.
[0083] Figure 5 3D point cloud stitching method based on projection is provided. src The number of vertices is 1518487, and the target point cloud C dest The number of vertices of the source point cloud C is 2279614. src and target point cloud C dest Based on the overall three-dimensional coordinates, we search for overlapping point pairs and stitch them together to obtain Figure 5 The stitching effect shown on the left takes 2.2 seconds in total. src and target point cloud C dest Based on the projection-based 3D point cloud stitching method provided in this embodiment, stitching is performed to obtain Figure 5 The stitching effect shown on the right takes a total of 0.38 seconds. It can be seen that compared with the traditional point cloud stitching method, the stitching speed of this method is nearly 6 times faster, and the efficiency is significantly improved.
[0084] The present application also provides a computer device, the internal structure diagram of which can be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a projection-based 3D point cloud stitching method is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.
[0085] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0086] like Figure 6 As shown, the present application also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps in the above-mentioned method embodiments.
[0087] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.
[0088] The present application also provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.
[0089] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0090] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0091] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.
[0092] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0093] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A projection-based 3D point cloud stitching method, characterized in that: include: Get source point cloud and target point cloud; Performing through filtering on the source point cloud and the target point cloud to obtain filtered source point cloud and target point cloud; Detecting the overlapping area of the filtered source point cloud and the target point cloud, projecting the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area to the xoy plane, and constructing an overlapping point pair index set; According to the overlapping point pair index set, fusing the overlapping vertices in the filtered source point cloud and the target point cloud; The vertex-fused source point cloud is spliced onto the vertex-fused target point cloud to obtain a spliced point cloud.
2. The method according to claim 1, characterized in that The vertices of the filtered source point cloud and the target point cloud falling into the bounding box of the overlapping area are projected onto the Xoy plane, and an overlapping point pair index set is constructed, including: Calculate the bounding box of the overlapping area according to the minimum point coordinates and the maximum point coordinates of the bounding box of the filtered source point cloud and the minimum point coordinates and the maximum point coordinates of the bounding box of the filtered target point cloud; The vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area are projected onto the Xoy plane, and an overlapping point pair index set is constructed.
3. The method according to claim 1, characterized in that The vertices of the filtered source point cloud and the target point cloud falling into the bounding box of the overlapping area are projected onto the Xoy plane, and an overlapping point pair index set is constructed, including: Detect the vertices of the filtered source point cloud and the target point cloud that fall into the bounding box of the overlapping area respectively, and record the sets of the serial numbers of the vertices as the source vertex set and the target vertex set respectively; The vertices of the filtered target point cloud that fall into the bounding box of the overlapping area are projected to the Xoy plane to construct the target overlapping area point cloud, and a kd-tree search tree is constructed based on the target overlapping area point cloud; A set of overlapping point pairs is obtained through the source vertex set and the kd-tree search tree; an overlapping point pair index set is constructed according to the correspondence between the vertex serial numbers in the target overlapping area point cloud and the vertex serial numbers in the target vertex set, as well as the overlapping point pairs.
4. The method according to claim 3, characterized in that The method projects the vertices of the filtered target point cloud that fall into the bounding box of the overlapping area to the Xoy plane, constructs the target overlapping area point cloud, and constructs a kd-tree search tree based on the target overlapping area point cloud, including: The vertices of the filtered target point cloud that fall into the bounding box of the overlapping area are projected to the xoy plane, the target overlapping area point cloud is constructed, and the corresponding relationship between the vertex serial numbers in the target overlapping area point cloud and the vertex serial numbers in the target vertex set is recorded; According to the target overlapping area point cloud, a kd-tree search tree is constructed.
5. The method according to claim 3, characterized in that The step of obtaining a set of overlapping point pairs through the source vertex set and the kd-tree search tree includes: Traversing the vertices belonging to the source vertex set in the filtered source point cloud, and projecting the vertices belonging to the source vertex set in the filtered source point cloud to the xoy plane; Searching the kd-tree search tree for the point in the target overlapping area point cloud that is closest to the projection point of the vertex belonging to the source vertex set in the filtered source point cloud on the xoy plane, and recording it as the closest point; If the distance is less than a set distance threshold, the closest point and the corresponding serial number of the vertex belonging to the source vertex set in the filtered source point cloud are recorded to obtain a set of overlapping point pairs.
6. The method according to claim 1, characterized in that The detecting the overlapping area of the filtered source point cloud and the target point cloud comprises: Detecting the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud; An intersection check is performed on the bounding box of the filtered source point cloud and the bounding box of the filtered target point cloud. If the two bounding boxes intersect, it is determined that there is an overlapping area between the filtered source point cloud and the target point cloud.
7. The method according to claim 1, characterized in that The step of fusing overlapping vertices in the filtered source point cloud and the target point cloud according to the overlapping point pair index set comprises: The vertex of the filtered source point cloud that falls within the bounding box of the overlapping area is taken as the first vertex, and the vertex of the filtered target point cloud that falls within the bounding box of the overlapping area is taken as the second vertex, and the sequence numbers of the first vertex and the second vertex belong to a pair of overlapping point pairs in the overlapping point pair index set; Calculate the coordinate value of the fused vertex according to the coordinate value of the first vertex and the coordinate value of the second vertex; The coordinate values of the first vertex and the second vertex are replaced with the coordinate values of the fused vertex to obtain the source point cloud and the target point cloud after the vertex fusion.
8. The method according to claim 7, characterized in that The step of splicing the vertex-fused source point cloud to the vertex-fused target point cloud to obtain a spliced point cloud includes: Traverse each vertex in the source point cloud after vertex fusion. If the index of the vertex belongs to the overlapping point pair index set, it will not be processed; otherwise, add the vertex to the set of serial numbers of the vertices in the bounding box of the overlapping area of the filtered target point cloud to obtain the spliced point cloud.
9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.
Citation Information
Patent Citations
Overlapped point cloud fusion method
CN110415362A
Large-range splicing method and device for vehicle-mounted laser scanning point clouds
CN110544201A
Splicing method and device for three-dimensional point cloud data
CN113706381A
Multi-camera laser point cloud splicing automatic de-weighting method
CN116091340A
3D target detection method and system based on environmental perception
CN116203576A
