A point cloud-image based three-dimensional imaging method and system

Abstract

The application relates to a point cloud-image-based three-dimensional imaging method and system, and belongs to the technical field of optical imaging. The method comprises the following steps: inputting an image, performing two-dimensional image instance segmentation on the image to obtain a two-dimensional mask; extracting point cloud corresponding to the two-dimensional mask from three-dimensional point cloud data of the image, processing the point cloud, and performing three-dimensional modeling to obtain a three-dimensional research target; performing triangulation on the three-dimensional research target to convert discrete point cloud into a grid structure; projecting the grid structure into a two-dimensional coordinate system, and finding the intersection of the edges of the triangles in the grid structure and image edge information; wherein the image edge information is obtained by performing edge detection on the two-dimensional research target; the two-dimensional research target is obtained by extracting the two-dimensional mask; and the intersection is added to the point cloud structure and triangulated to complete three-dimensional imaging. The application can simultaneously obtain three-dimensional space information and texture characteristics, reduces the difference between the scales of the point cloud and the image data, and further reduces the loss of texture information.

Description

Technical Field

[0001] This invention relates to the field of optical imaging technology, and in particular to a three-dimensional imaging method and system based on point cloud-image. Background Technology

[0002] 3D imaging technology, combining LiDAR and optical cameras, can provide a comprehensive description of real-world objects. This technology combines depth information from point clouds with texture information from images, providing both spatial information and surface texture of the object. However, due to the difference in scale between point cloud and image data, traditional processing methods face challenges in achieving accurate 3D reconstruction and texture mapping. To address this issue, researchers have developed a point cloud-image pre-fusion technique, which fuses data at the initial stages of point cloud generation and image capture, thereby better preserving the depth information and image texture of each pixel. This technique not only improves the overall performance and robustness of the system but also has broad application prospects in fields such as autonomous driving, robot navigation, and intelligent monitoring.

[0003] Despite this, existing optical 3D imaging methods often fail to simultaneously capture 3D spatial information and texture features in the imaging results, and the issue of scale differences between point cloud and image data remains unresolved. Furthermore, in 3D imaging methods based on point cloud and image registration, the discreteness of point cloud coordinate distribution and the continuity of image coordinate distribution lead to the loss of significant image texture information during imaging, especially edge features. Simultaneously, because depth information cannot be acquired pixel-by-pixel, the imaging results of point clouds and images tend to be similar, making it impossible for existing methods to fully preserve the original texture features when completing 3D imaging tasks. Summary of the Invention

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the inability to simultaneously obtain three-dimensional spatial information and texture features, the difference in scale between point cloud and image data, and the loss of texture information in the prior art, and to propose a three-dimensional imaging method and system based on point cloud-image.

[0005] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a three-dimensional imaging method based on point cloud-image, comprising:

[0006] Input an image, perform two-dimensional image instance segmentation on the image, and obtain a two-dimensional mask;

[0007] The point cloud corresponding to the two-dimensional mask is extracted from the three-dimensional point cloud data of the image, the point cloud is processed and three-dimensional modeling is performed to obtain the three-dimensional research target;

[0008] The three-dimensional research target is triangulated to convert the discrete point cloud into a mesh structure;

[0009] The mesh structure is projected into a two-dimensional coordinate system to find the intersection points of the sides of the triangles in the mesh structure and the image edge information; wherein, the image edge information is obtained by edge detection of the two-dimensional research target; the two-dimensional research target is obtained by extraction from the two-dimensional mask;

[0010] The intersection points are added to the point cloud structure and triangulation is performed to complete the three-dimensional imaging.

[0011] In one embodiment of the present invention, the method for extracting the point cloud corresponding to the two-dimensional mask from the three-dimensional point cloud data of the image, processing the point cloud and performing three-dimensional modeling to obtain the three-dimensional research target is as follows:

[0012] The point cloud is projected from the three-dimensional coordinate system to the two-dimensional coordinate system using the following formula:

[0013]

[0014] Among them, (x i ,y i ,z i (u) represents the 3D coordinates of one of the points in the point cloud. i ,v i ) represents the coordinates of the entire point cloud projected onto the two-dimensional coordinate system, K is the intrinsic parameter matrix, R is the rotation matrix, and t is the translation vector;

[0015] Compare the two-dimensional coordinate region covered by the two-dimensional mask and (u i ,v i ), determine (u i ,v i If it is within the area of ​​the 2D mask, then search for (u) within the 2D mask area. i ,v i (x) i ,y i ,z i This allows us to obtain all the 3D point clouds projected onto the 2D mask.

[0016] Based on the clustering state of points in the point cloud, the cluster with the smallest x-value in the x-axis direction is identified as the 3D research target.

[0017] In one embodiment of the present invention, the triangulation method is as follows:

[0018] For each triangle in the mesh structure, calculate the area inside the two-dimensional mask and the area outside the two-dimensional mask using the following formula:

[0019]

[0020] Among them, S n,i S represents the area within the two-dimensional mask. n,o T is the area outside the two-dimensional mask. n Let A be one of the triangles in the mesh structure, let mask represent the two-dimensional mask, and let A be the integration object of the double integral.

[0021] In one embodiment of the present invention, for each triangle in the mesh structure, the area inside the two-dimensional mask and the area outside the two-dimensional mask are calculated. When the area inside the two-dimensional mask is not greater than the area outside the two-dimensional mask, the triangle is deleted.

[0022] In one embodiment of the present invention, the method for performing two-dimensional image instance segmentation on the image is the GroundedSAM method.

[0023] In one embodiment of the present invention, the edge detection is Canney edge detection.

[0024] Secondly, to solve the above-mentioned technical problems, the present invention provides a computer program product, which includes a computer program that, when the computer program is run, causes the above-mentioned three-dimensional imaging method based on point cloud-image to be executed.

[0025] Thirdly, to solve the above-mentioned technical problems, the present invention provides a three-dimensional imaging system based on point cloud-image, comprising:

[0026] A two-dimensional mask acquisition module is used to input an image, perform two-dimensional image instance segmentation on the image, and obtain a two-dimensional mask;

[0027] The three-dimensional research target acquisition module is used to extract the point cloud corresponding to the two-dimensional mask from the three-dimensional point cloud data of the image, process the point cloud and perform three-dimensional modeling to obtain the three-dimensional research target;

[0028] The triangulation module is used to triangulate the three-dimensional research target, converting the discrete point cloud into a mesh structure;

[0029] The intersection point acquisition module is used to project the mesh structure onto a two-dimensional coordinate system and find the intersection points of the sides of the triangles in the mesh structure and the image edge information; wherein, the image edge information is obtained by edge detection of the two-dimensional research target; the two-dimensional research target is obtained by extraction from the two-dimensional mask;

[0030] The three-dimensional imaging module is used to add the intersection points to the point cloud structure and perform triangulation to complete the three-dimensional imaging.

[0031] Fourthly, to solve the above-mentioned technical problems, the present invention provides a 3D scanner, including the above-mentioned point cloud-image-based three-dimensional imaging system.

[0032] Fifthly, to solve the above-mentioned technical problems, the present invention provides a medical device, including the aforementioned 3D scanner.

[0033] Compared with the prior art, the above-described technical solution of the present invention has the following advantages:

[0034] This invention discloses a point cloud-image-based 3D imaging method and system. It employs edge detection and triangulation techniques, and uses the point cloud to find a unit scale to ensure that the image texture within each unit scale is highly approximate. This innovative strategy effectively alleviates the order-of-magnitude mismatch between point cloud and image data, reduces texture information loss, and achieves clear and sharp 3D imaging. Simultaneously, this invention fully utilizes point cloud data and image edge information to simultaneously acquire 3D spatial information and texture features, improving imaging accuracy and realism, and providing a solid foundation for high-quality 3D reconstruction. Furthermore, extracting the point cloud corresponding to the 2D mask from the 3D point cloud data of the image allows for the selection of the corresponding 3D point cloud from the original 3D point cloud, ensuring the integrity and accuracy of the research target in 3D space. Attached Figure Description

[0035] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein...

[0036] Figure 1 This is a flowchart of a point cloud-image-based three-dimensional imaging method according to a preferred embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of a point cloud acquisition method based on a two-dimensional mask within a preferred embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of Delauney triangulation correction based on a two-dimensional mask in a preferred embodiment of the present invention;

[0039] Figure 4 This is a result diagram of a practical application of a point cloud-image-based three-dimensional imaging system according to a preferred embodiment of the present invention;

[0040] Figure 5 This is another result diagram of the practical application of a point cloud-image-based three-dimensional imaging system according to a preferred embodiment of the present invention. Detailed Implementation

[0041] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0042] Reference Figure 1 As shown, this embodiment of the invention provides a three-dimensional imaging method based on point cloud-image, including:

[0043] Input an image, perform two-dimensional image instance segmentation on the image, and obtain a two-dimensional mask;

[0044] Extract the point cloud corresponding to the two-dimensional mask from the three-dimensional point cloud data of the image, process the point cloud and perform three-dimensional modeling to obtain the three-dimensional research target;

[0045] Triangulation is performed on the three-dimensional research target to convert the discrete point cloud into a mesh structure;

[0046] The mesh structure is projected onto a two-dimensional coordinate system, and the intersection points of the edges of the triangles in the mesh structure and the image edge information are found. The image edge information is obtained by edge detection of the two-dimensional research target. The two-dimensional research target is obtained by extraction of a two-dimensional mask. At this time, the image texture corresponding to each triangle is approximate, and each triangle is the unit scale.

[0047] The intersection points are added to the point cloud structure and triangulation is performed to complete the 3D imaging.

[0048] This invention provides a point cloud-image-based 3D imaging method that employs edge detection and triangulation techniques. It finds a unit scale using the point cloud to approximate the image texture within each unit scale, thereby mitigating the information loss caused by the order-of-magnitude mismatch between point cloud and image data. This results in clear and sharp 3D imaging. Simultaneously, this method fully utilizes point cloud data and image edge information to simultaneously acquire 3D spatial information and texture features, improving imaging accuracy and realism, and providing a solid foundation for high-quality 3D reconstruction.

[0049] In the acquisition of 3D and 2D research targets, traditional methods rely on performing instance segmentation algorithms on images and point clouds separately to efficiently extract specific 2D and 3D research targets from large-scale scenes. However, this method has limitations, mainly because small error accumulation during the processing of multiple neural networks can lead to inconsistent results, and it also requires high computational resources. The Grounded SAM (Grounded Spatial Attention Module) method effectively reduces the risk of error accumulation, simplifies the processing flow, and reduces reliance on high-performance computing resources by reducing the steps that rely on multiple neural networks. The Grounded SAM method combines two methods: Grounding DINO and Segment Anything Model (SAM). Grounding DINO is responsible for understanding text prompts and locating corresponding targets in the image, while SAM is responsible for instance segmentation and generating masks. This method directly maps 2D masks to 3D space, which not only speeds up data processing but also ensures the accuracy of the 2D-to-3D conversion, improving the accuracy of research target extraction. Therefore, in this embodiment of the invention, the preferred method for two-dimensional image instance segmentation is the Grounded SAM method.

[0050] Specifically, during instance segmentation, the detection results of Grounding DINO are used as input data. SAM performs segmentation operations on the image based on these detection results, generating a mask for each instance. These masks closely fit the target instances in the image, ensuring segmentation accuracy. Through this coherent segmentation process, one or more two-dimensional masks are obtained, representing the segmented target instances in the image. Finally, these masks are used to extract the desired image regions, which constitute the two-dimensional research targets. This method not only improves segmentation efficiency but also enhances the accuracy and reliability of the results.

[0051] After obtaining the 2D mask, the next step is to map the 2D mask into 3D space using the precise projection relationship between the image and the point cloud. This process involves associating 2D image data with corresponding 3D point cloud data, thereby accurately locating and extracting point cloud data within the study area in 3D space. Through this mapping, the 3D point cloud corresponding to the 2D mask can be filtered from the original 3D point cloud, ensuring the integrity and accuracy of the research target in 3D space. This method effectively converts 2D information into 3D information, providing rich 3D data support for subsequent analysis and applications.

[0052] Specifically, refer to Figure 2 , Figure 2 This diagram illustrates a method for acquiring point clouds within a research area based on a 2D mask. Point clouds corresponding to the 2D mask are extracted from the 3D point cloud data of an image. These point clouds are then processed and 3D modeled to obtain the 3D research target. Point cloud processing includes smoothing and denoising to optimize data quality. The denoising step aims to eliminate outliers and noise in the point cloud, while the smoothing step reduces small-scale fluctuations in the data, providing a continuous and accurate surface representation for subsequent 3D modeling, analysis, and visualization. The method for obtaining the 3D research target is as follows:

[0053] S201. Based on the projection relationship between the point cloud and the image, project the point cloud from the three-dimensional coordinate system to the two-dimensional coordinate system. The projection formula is:

[0054]

[0055] Among them, (x i ,y i ,z i (u) represents the 3D coordinates of one of the points in the point cloud; i ,v i ) represents the coordinates of the entire point cloud projected onto a two-dimensional coordinate system; K is the intrinsic parameter matrix, determined by the camera that acquired the image; R is the rotation matrix; and t is the translation vector, determined by the pose of the camera and the radar or other equipment that acquired the point cloud.

[0056] S202, Compare the two-dimensional coordinate region covered by the two-dimensional mask with (u i ,v i ), determine (u i ,v i Whether it is within the area of ​​the two-dimensional mask.

[0057] S203. If it is within the area of ​​the two-dimensional mask, then find (u) within the area of ​​the two-dimensional mask. i ,v i (x) i ,y i ,z i This method obtains all 3D point clouds projected onto a 2D mask, where the same point can be searched consistently according to its index order throughout the array.

[0058] S204. Based on the clustering state of points in the point cloud, the cluster with the smallest x-value in the x-axis direction is the three-dimensional research target.

[0059] In three-dimensional space, points on the depth direction (usually the x-axis) of a point cloud may fall within a two-dimensional mask area when projected onto a two-dimensional plane. However, not all of these projected points belong to the object under study. Due to the characteristics of point cloud acquisition, only the points of objects closest to the acquisition device will be visible in the two-dimensional mask, while objects further away may be occluded. Therefore, in step S204, the depth information and occlusion relationships of the point cloud are used to accurately filter out the valid points belonging to the object under study to ensure the accuracy and relevance of the data.

[0060] After obtaining the 3D research target, the next step is to triangulate it to generate a mesh for 3D modeling. Delaunay triangulation is a widely used method that can quickly generate meshes and ensures that the generated mesh is hole-free, meaning that the generated mesh is a complete surface without internal voids. However, it is important to note that the mesh generated by Delaunay triangulation may form a convex hull, which may lead to the generated mesh not perfectly matching the shape of the actual object. A convex hull is the smallest convex polygon or convex polyhedron that can enclose all points, while the surface of the actual object may be non-convex. Therefore, directly using Delaunay triangulation may cause some geometric distortions. To solve this problem, the results of Delaunay triangulation usually need to be optimized and adjusted to better match the shape of the actual object. This may include removing some triangles that do not conform to the object's surface, or rearranging and connecting the triangles to generate a more accurate and smoother mesh model. In addition, other triangulation algorithms can be used, such as Constrained Delaunay Triangulation, which allows triangulation under certain constraints, thereby better adapting to the actual boundaries of the object. In a preferred embodiment of the present invention, the Delauney triangulation is corrected using the obtained two-dimensional mask and formula (1).

[0061] Specifically, refer to Figure 3 , Figure 3 This is a schematic diagram of Delauney triangulation correction based on a 2D mask. The specific methods for triangulation and its correction are as follows:

[0062] For each triangle in the mesh structure, calculate the area inside the 2D mask and the area outside the 2D mask using the following formula:

[0063]

[0064] Among them, S n,i S is the area within the two-dimensional mask. n,o Let T be the area outside the two-dimensional mask. nLet S be one of the triangles in the mesh structure, mask represent a two-dimensional mask, and A be the object of the double integral. When the area inside the two-dimensional mask is no greater than the area outside the two-dimensional mask, i.e., S... n,i ≤S n,o When the time comes, delete the triangle.

[0065] Furthermore, the mesh structure is projected onto a two-dimensional coordinate system. To accurately align the 3D mesh with the 2D image data, edge detection is first performed on the 2D research target to extract edge information from the image. Secondly, the edges of all triangles are extracted from the mesh structure of the 3D model. By calculating the intersections of these edges with the image edges, the correspondence between the 3D structure and the 2D image is determined. This method improves the accuracy and reliability of 3D reconstruction and image analysis.

[0066] Among the edge detection algorithms, the Canny algorithm stands out due to its excellent performance. It can not only accurately identify real edges in an image but also precisely locate edge pixels. Through Gaussian filter preprocessing, the Canny algorithm significantly reduces the impact of image noise, thereby improving the accuracy of edge detection. Furthermore, the Canny algorithm provides a flexible parameter adjustment mechanism, including the standard deviation of the Gaussian filter and dual threshold settings, allowing users to optimize the edge detection results according to specific application scenarios. Its design aims to ensure high accuracy of detection results by minimizing false detections and false localizations, making the detected edges as close as possible to the actual edges. Therefore, this embodiment of the invention preferably uses the Canny edge detection method to obtain image edge information.

[0067] Example 2

[0068] This embodiment provides a computer program product, which includes a computer program that, when executed, causes the point cloud-image-based three-dimensional imaging method described in Embodiment 1 to be executed.

[0069] Example 3

[0070] Based on the same inventive concept, this embodiment provides a three-dimensional imaging system based on point cloud-image. The principle of solving the problem is similar to that of the three-dimensional imaging method based on point cloud-image provided in Embodiment 1, and the repeated parts will not be described again.

[0071] This invention provides a point cloud-image based three-dimensional imaging system, comprising:

[0072] The 2D mask acquisition module is used to input an image, perform 2D image instance segmentation on the image, and obtain a 2D mask.

[0073] The 3D research target acquisition module is used to extract the point cloud corresponding to the 2D mask from the 3D point cloud data of the image, process the point cloud and perform 3D modeling to obtain the 3D research target.

[0074] The triangulation module is used to triangulate the three-dimensional research target, converting the discrete point cloud into a mesh structure.

[0075] The intersection point acquisition module is used to project the mesh structure onto a two-dimensional coordinate system and find the intersection points between the sides of the triangles in the mesh structure and the image edge information; wherein, the image edge information is obtained by edge detection of the two-dimensional research target; the two-dimensional research target is obtained by two-dimensional mask extraction;

[0076] The 3D imaging module is used to add intersection points to the point cloud structure and perform triangulation to complete 3D imaging.

[0077] Reference Figure 4 and Figure 5 , Figure 4 and Figure 5 Two result images generated by the point cloud-image 3D imaging system in a practical application are shown. It is clear from the images that the system provided by this embodiment of the invention can not only simultaneously capture 3D spatial information and texture features, but also effectively reduce the scale difference between point cloud and image data, thereby preventing the loss of texture information.

[0078] Example 4

[0079] This embodiment provides a 3D scanner, including the point cloud-image-based three-dimensional imaging system provided in Embodiment 3.

[0080] Example 5

[0081] This embodiment provides a medical device, including a 3D scanner as provided in Embodiment 4.

[0082] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0083] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A three-dimensional imaging method based on point cloud-image, characterized in that, include: Input an image, perform two-dimensional image instance segmentation on the image, and obtain a two-dimensional mask; The point cloud corresponding to the two-dimensional mask is extracted from the three-dimensional point cloud data of the image; the point cloud is processed and three-dimensional modeling is performed to obtain the three-dimensional research target; wherein, the steps for obtaining the three-dimensional research target are as follows: The point cloud is projected from the three-dimensional coordinate system to the two-dimensional coordinate system using the following formula: ； in, Let be the 3D coordinates of one of the points in the point cloud. The coordinates are the projections of the entire point cloud onto a two-dimensional coordinate system. This is the intrinsic parameter matrix. For rotation matrix, It is a translation vector; Compare the two-dimensional coordinate region covered by the two-dimensional mask and ,judge Is it within the area of ​​the 2D mask? If it is within the area of ​​the 2D mask, then search for [something] within the 2D mask area. corresponding This yields all the three-dimensional point clouds projected onto the two-dimensional mask. Based on the clustering state of points in the point cloud, the cluster with the smallest x-value in the x-axis direction is the 3D research target; The three-dimensional research target is triangulated to convert the discrete point cloud into a mesh structure; wherein the triangulation method is as follows: For each triangle in the mesh structure, calculate the area inside the 2D mask and the area outside the 2D mask; if the area inside the 2D mask is not greater than the area outside the 2D mask, delete the triangle; the calculation formula is: ； ； in, The area within the two-dimensional mask. The area outside the two-dimensional mask. Let mask be one of the triangles in the mesh structure, and let mask represent the two-dimensional mask. It is the object of integration for a double integral; The mesh structure is projected into a two-dimensional coordinate system to find the intersection points of the sides of the triangles in the mesh structure and the image edge information; wherein, the image edge information is obtained by edge detection of the two-dimensional research target; the two-dimensional research target is obtained by extraction from the two-dimensional mask; The intersection points are added to the point cloud structure and triangulation is performed to complete the three-dimensional imaging.

2. The three-dimensional imaging method based on point cloud-image according to claim 1, characterized in that, The method for performing two-dimensional image instance segmentation on the image is the Grounded SAM method.

3. The three-dimensional imaging method based on point cloud-image according to claim 1, characterized in that, The edge detection is Canney edge detection.

4. A computer program product, characterized in that, The computer program product includes a computer program that, when executed, causes the method as described in any one of claims 1 to 3 to be performed.

5. A three-dimensional imaging system based on point cloud-image, characterized in that, A method for implementing a point cloud-image based three-dimensional imaging method according to any one of claims 1 to 3 includes: A two-dimensional mask acquisition module is used to input an image, perform two-dimensional image instance segmentation on the image, and obtain a two-dimensional mask; The three-dimensional research target acquisition module is used to extract the point cloud corresponding to the two-dimensional mask from the three-dimensional point cloud data of the image, process the point cloud and perform three-dimensional modeling to obtain the three-dimensional research target; The triangulation module is used to triangulate the three-dimensional research target, converting the discrete point cloud into a mesh structure; The intersection point acquisition module is used to project the mesh structure onto a two-dimensional coordinate system and find the intersection points of the sides of the triangles in the mesh structure and the image edge information; wherein, the image edge information is obtained by edge detection of the two-dimensional research target; the two-dimensional research target is obtained by extraction from the two-dimensional mask; The three-dimensional imaging module is used to add the intersection points to the point cloud structure and perform triangulation to complete the three-dimensional imaging.

6. A 3D scanner, characterized in that, This includes the point cloud-image-based three-dimensional imaging system described in claim 5.

7. A medical device, characterized in that, Including the 3D scanner described in claim 6.

Images