Method and system for acquiring Gaussian model surface based on point cloud data, and storage medium

Abstract

The invention discloses a method and system for obtaining a Gaussian model surface based on point cloud data and a storage medium, and belongs to the technical field of virtuality and reality, and the method comprises the steps: obtaining a Gaussian model, and importing the Gaussian model into a three-dimensional coordinate system; generating bounding boxes of which the lengths, widths and heights are L1, L2 and L3 in sequence; obtaining a central point of the bounding box, and making cutting sections for the bounding box along three directions perpendicular to the X axis, the Y axis and the Z axis; making a chessboard segmentation line on each tangent plane, and making a normal perpendicular to the tangent plane by taking each line intersection point of the chessboard segmentation lines as a starting point; recording all point cloud data on each normal, finding out the point cloud data with the maximum distance from the line intersection point on the positive normal and the negative normal, and storing the point cloud data in a surface point set; repeated point cloud data in the surface point set are removed, and the reserved point cloud data are sequentially connected to form a continuous edge contour model. According to the method, the outer surface profile of the Gaussian model can be rapidly obtained, the calculation mode is simple and efficient, and the use cost is low.

Description

Technical Field

[0001] The present invention relates to the field of virtual and reality technology, and in particular to a method, system and storage medium for acquiring a Gaussian model surface based on point cloud data. Background Art

[0002] Virtual reality technology (English name: Virtual Reality, abbreviated as VR) encompasses computers, electronic information, and simulation technologies. Its basic implementation method is based on computer technology, utilizing and integrating the latest developments in three-dimensional graphics technology, multimedia technology, simulation technology, display technology, servo technology, and other high-tech technologies. With the help of computers and other equipment, it creates a realistic three-dimensional virtual world with multiple sensory experiences such as vision, touch, and smell, thereby giving people in the virtual world an immersive feeling.

[0003] Gaussian models have important applications in the field of panoramic imaging in virtual and real technology. Gaussian models can be used to build realistic virtual scenes. Through Gaussian sputtering technology, real-time high-quality rendering effects can be achieved, allowing users to immerse themselves in virtual scenes. In particular, 3D Gaussian models are often used to represent geometric and color information in virtual scenes. Through the distribution of Gaussian balls (point cloud data), geometric elements such as points, lines, and surfaces in the scene can be explicitly represented, while carrying color information, thereby achieving realistic rendering effects. Since the Gaussian model is composed of a large amount of point cloud data, the point cloud data exists both inside and outside the Gaussian model, and its point cloud data is not continuous, compared with traditional virtual models, it is impossible to directly obtain surface information. When the Gaussian model is applied to a virtual scene, it is often necessary to perform secondary processing on the Gaussian model. Therefore, it is particularly important to obtain the surface information of the model. However, the traditional acquisition method can only rely on manually depicting the outer surface of the model. The outer surface contour obtained in this way is rough, and the depiction efficiency is low and prone to errors. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the outer surface of the traditional Gaussian model is difficult to obtain, and to propose a method, system and storage medium for obtaining the Gaussian model surface based on point cloud data. The method has the advantages of being able to quickly generate the outer surface contour, high operating efficiency and more refined display effect.

[0005] In a first aspect, the present invention achieves the above-mentioned purpose through the following technical solutions: a method for obtaining a Gaussian model surface based on point cloud data, the method comprising the following steps: Obtain the Gaussian model and import it into the three-dimensional coordinate system; Calculate the maximum length L1 of the Gaussian model along the X-axis, the maximum length L2 along the Y-axis, and the maximum length L3 along the Z-axis, and generate a bounding box whose length, width, and height are L1, L2, and L3 respectively; Obtain the center point of the bounding box, and divide the bounding box into sections along three directions perpendicular to the X-axis, the Y-axis, and the Z-axis, respectively, so that the intersection of the three sections coincides with the center point; Draw a chessboard dividing line on each of the cut surfaces, and draw a normal line perpendicular to the cut surface with each intersection of the chessboard dividing line as a starting point, wherein each normal line includes two directions, a positive direction and a negative direction; Record all point cloud data on each normal line, and calculate the distance between each point cloud data and the intersection of the line, and find the point cloud data with the largest distance on both the positive normal line and the negative normal line and save them in the surface point set; The repeated point cloud data in the surface point set are removed, and the retained point cloud data are sequentially connected to form a continuous edge contour model.

[0006] Preferably, the method for obtaining the center point of the bounding box is: Calculate the coordinates of the vertex (x0, y0, z0) of the bounding box closest to the origin of the 3D coordinate system; Then the x-axis coordinate of the center point is: x=x0+L1 / 2; The y-axis coordinate of the center point is: y=y0+L2 / 2; The z-axis coordinate of the center point is: z=z0+L3 / 2; Get the center point coordinates (x, y, z) of the bounding box.

[0007] Preferably, the method further comprises setting the center point of the bounding box as the coordinate origin of the three-dimensional coordinate system.

[0008] Preferably, the method of drawing a normal line perpendicular to the section surface with the intersection points of the chessboard dividing lines as starting points is: Get the section of the bounding box perpendicular to the X axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the X axis, which is the positive normal line, and emit a ray in the negative direction of the X axis, which is the negative normal line; Get the section of the bounding box perpendicular to the Y axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the Y axis, which is the positive normal line, and emit a ray in the negative direction of the Y axis, which is the negative normal line; Get the slice of the bounding box perpendicular to the Z axis, select the intersection of the chessboard dividing lines on the slice as the starting point to emit a ray in the positive direction of the Z axis, which is the positive normal, and emit a ray in the negative direction of the Z axis, which is the negative normal.

[0009] Preferably, the method further comprises setting the length and width of the cells divided by the chessboard dividing lines.

[0010] Preferably, the method of connecting the retained point cloud data in sequence to form a continuous edge contour model is: selecting one of the point cloud data as the starting point, connecting the starting point to all adjacent point cloud data until all the point cloud data are connected. The closed contour formed at this time is the outer surface of the Gaussian model.

[0011] In a second aspect, the present invention achieves the above-mentioned purpose through the following technical solutions: a system for obtaining a Gaussian model surface based on point cloud data, the system comprising: Model import unit, used to obtain Gaussian model and import it into three-dimensional coordinate system; A bounding box generating unit, used to calculate the maximum length L1 of the Gaussian model along the X-axis direction, the maximum length L2 along the Y-axis direction, and the maximum length L3 along the Z-axis direction, and generate a bounding box whose length, width, and height are L1, L2, and L3 respectively; A segmentation section unit is used to obtain the center point of the bounding box, and to segment the bounding box along three directions perpendicular to the X-axis, the Y-axis and the Z-axis, so that the intersection of the three segments is consistent with the center point; A normal line generating unit, used for making a chessboard dividing line on each of the cut surfaces, and making a normal line perpendicular to the cut surface with each intersection of the chessboard dividing line as a starting point, wherein each normal line includes two directions, a positive direction and a negative direction; A point cloud screening unit is used to record all point cloud data on each normal line, calculate the distance between each point cloud data and the intersection of the line, and find the point cloud data with the largest distance on both the positive normal line and the negative normal line and save it in the surface point set; The outer surface shaping unit is used to remove repeated point cloud data in the surface point set and connect the retained point cloud data in sequence to form a continuous edge contour model.

[0012] Preferably, the system further comprises setting the center point of the bounding box as the coordinate origin of the three-dimensional coordinate system.

[0013] Preferably, the normal generation unit further includes setting the length and width of the cells divided by the chessboard dividing lines.

[0014] In a third aspect, the present invention achieves the above-mentioned purpose through the following technical scheme: a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for obtaining a Gaussian model surface based on point cloud data as described in the first aspect.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention utilizes the method of importing a Gaussian model into a three-dimensional coordinate system, then cutting the Gaussian model with the X-axis, Y-axis and Z-axis of the vertical coordinate system, generating a chessboard-like dividing line on each cut surface, taking the intersection of the dividing lines as the normal perpendicular to the surface, selecting the point cloud data on the normal farthest from the intersection of the lines, thereby finding the point cloud data of the outer surface of the Gaussian model, de-duplicating and connecting the point cloud data, and finally obtaining the entire outer surface contour. In this way, the outer surface of the Gaussian model can be quickly obtained, the calculation is simple and efficient, and the efficiency and accuracy of obtaining the Gaussian model surface are greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of the method for obtaining a Gaussian model surface based on point cloud data of the present invention.

[0017] Figure 2 It is a schematic diagram of a Gaussian model enclosed by a bounding box in the coordinate system of the present invention.

[0018] Figure 3 It is a schematic diagram of segmenting a Gaussian model according to the present invention.

[0019] Figure 4 It is a schematic diagram of the chessboard dividing lines on the cross section of the present invention.

[0020] Figure 5 It is a schematic diagram of the system composition for acquiring Gaussian model surface based on point cloud data of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Example 1 like Figure 1 As shown, a method for obtaining a Gaussian model surface based on point cloud data comprises the following steps: Step S1, obtain the Gaussian model and import it into a three-dimensional coordinate system. The three-dimensional coordinate system includes three axes: X, Y, and Z. The Gaussian model is composed of a number of point cloud data. Due to the application of Gaussian sputtering technology, the point cloud data is not only distributed on the outer surface of the Gaussian model but also irregularly distributed inside. By importing the Gaussian model into the three-dimensional coordinate system, each point cloud data on the Gaussian model can be represented by coordinates, which is convenient for subsequent calculations.

[0023] Step S2, calculate the maximum length L1 of the Gaussian model along the X-axis, the maximum length L2 along the Y-axis, and the maximum length L3 along the Z-axis, and generate a bounding box whose length, width, and height are L1, L2, and L3, respectively. In step S1, all coordinates of the point cloud data constituting the Gaussian model can be known, so by calculating the maximum value, the maximum value along the X-axis direction can be obtained, which is recorded as L1, the maximum value along the Y-axis direction is recorded as L2, and the maximum value along the Z-axis direction is recorded as L3. The bounding box can be generated by the maximum value, such as Figure 2 The figure shows a schematic diagram of the Gaussian model in the bounding box. Since L1, L2 and L3 are the maximum distances of the Gaussian model in three directions, the bounding box can fully include the Gaussian model. For irregularly shaped Gaussian models, the bounding box, which is a cuboid or cube, can replace the irregular Gaussian model for subsequent calculations, reducing the difficulty of calculations.

[0024] Step S3, obtaining the center point of the bounding box, dividing the bounding box into sections along three directions perpendicular to the X-axis, Y-axis and Z-axis, so that the intersection of the three sections coincides with the center point, and the center point of the bounding box is located at half the length, half the width and half the height of the bounding box. The method also includes setting the center point of the bounding box as the origin of the three-dimensional coordinate system. Figure 3 As shown, the three sections generated by the segmentation are x1, y1 and z1, x1 is perpendicular to the X-axis, y1 is perpendicular to the Y-axis, and z1 is perpendicular to the Z-axis. In order to ensure that each section can divide the Gaussian model into two parts, the intersection of the three sections is coincident with the center point of the bounding box to ensure that each segmentation can divide the model into two halves. The center point is the origin of the three-dimensional coordinate system. One is to better calculate the coordinates of the point cloud data, and the other is to ensure that half of the Gaussian model after segmentation is located on the positive direction side of one of the X-axis, Y-axis and Z-axis, and the other half is located on the opposite direction side of the axis.

[0025] Step S4: draw a chessboard dividing line on each of the sections, and draw a normal line perpendicular to the section with each intersection of the chessboard dividing line as the starting point, each of the normal lines includes two directions, positive and negative, such as Figure 4 As shown, taking section y1 as an example, several cells can be formed on the surface of the section through the chessboard dividing lines. The vertex of each cell is the line intersection point, and the number of normals is the number of line intersection points. Since section y1 is perpendicular to the Y-axis and the origin of the Y-axis is on section y1, a normal is drawn in the positive direction of the section from the line intersection point as the starting point, and the normal is consistent with the positive direction of the Y-axis. A normal is drawn in the reverse direction of the section, and the normal is consistent with the negative direction of the Y-axis. The reason for choosing normals in two directions here is that each section divides the Gaussian model into two parts, so there will be point cloud data on both normals.

[0026] Step S5, record all the point cloud data on each normal line, and calculate the distance between each point cloud data and the intersection of the line, find the point cloud data with the largest distance on both the positive normal line and the negative normal line and save them in the surface point set. According to step S4, there will be point cloud data of the Gaussian model in the positive and negative directions of each normal line. In order to find out what kind of point cloud data is the outer surface point cloud data of the Gaussian model, the distance between all the point cloud data on the same normal line and the intersection of the line is calculated. The point cloud data with the largest distance indicates that it is the farthest distance from the section, so it is also the outer surface point cloud data of the Gaussian model. Save these point cloud data in the surface point set for calculation in step S6.

[0027] Step S6: remove the duplicate point cloud data in the surface point set, and connect the retained point cloud data in sequence to form a continuous edge contour model. Since we make a chessboard dividing line for each section, the point cloud data of the maximum distance calculated after making the normal line will overlap, such as Figure 2 As shown, when each section has a positive or negative normal, a part of the outer surface of the Gaussian model will exist on the positive or negative normal corresponding to other sections. Therefore, the calculated surface point set will have some repeated point cloud data. In this step, the repeated point cloud data is removed, and the uniqueness of each point cloud data is retained to avoid errors in the subsequent connection process. After removing the repeated point cloud data, the surface point set only includes the point cloud data of the outer surface of the Gaussian model. At this time, by connecting each point cloud data, a continuous edge contour model can be formed, and the generated contour model fits the outer surface of the original Gaussian model accurately, and errors are not easy to occur.

[0028] In step S3, the method for determining the center point is roughly described. Now, the coordinates of the center point are accurately calculated. The method for obtaining the center point of the bounding box is: Calculate the coordinates of the vertex (x0, y0, z0) of the bounding box closest to the origin of the three-dimensional coordinate system. The positions of the vertex of the bounding box in the three-dimensional coordinate system are determined, so it is easy to calculate the coordinates of the vertex closest to the origin. Using the vertex coordinates as the basic quantity, step S3 mentions that the center point of the bounding box is located at half the length, half the width and half the height of the bounding box. When the vertex coordinates are determined, the coordinates of the center point can be calculated; Then the x-axis coordinate of the center point is: x=x0+L1 / 2; The y-axis coordinate of the center point is: y=y0+L2 / 2; The z-axis coordinate of the center point is: z=z0+L3 / 2; Get the center point coordinates (x, y, z) of the bounding box.

[0029] The method of drawing a normal line perpendicular to the section surface with the intersection points of the chessboard dividing lines as starting points is: Get the section of the bounding box perpendicular to the X axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the X axis, which is the positive normal line, and emit a ray in the negative direction of the X axis, which is the negative normal line; Get the section of the bounding box perpendicular to the Y axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the Y axis, which is the positive normal line, and emit a ray in the negative direction of the Y axis, which is the negative normal line; Get the section of the bounding box perpendicular to the Z axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the Z axis, which is the positive normal, and emit a ray in the negative direction of the Z axis, which is the negative normal. Each axis of the three-dimensional coordinate system has a positive direction and a negative direction. Keeping the direction of the normal consistent with the positive and negative directions of the axis can quickly find out which point cloud data are in the positive or negative direction of the axis during calculation, which is also conducive to subsequent calculations.

[0030] According to step S6, the edge contour model finally obtained is obtained by connecting the point cloud data in the surface point set after deduplication. We cannot keep the cell size of the chessboard dividing line consistent with the spacing between the outer surface point cloud data of the original Gaussian model. The method also includes setting the length and width of the cell divided by the chessboard dividing line. By adjusting the length and width of the cell, the spacing between the point cloud data finally obtained can be changed. If the length and width of the cell are too large, the spacing between the normals is also large, so the number of point cloud data obtained is small, and the edge contour model finally formed may not be able to achieve the phenomenon of perfectly fitting the outer surface of the original Gaussian model. On the contrary, the cell size is small, the normal density is large, and the number of point cloud data is large. The edge contour model obtained can perfectly fit the outer surface of the original Gaussian model. Due to different application requirements, for example, for the need to completely restore the information of the outer surface of the Gaussian model, it is necessary to pursue details, but the amount of calculation brought by the dense normals is also large. If only the information of the outer surface of the Gaussian model needs to be roughly restored, then the amount of calculation becomes smaller. Therefore, by adjusting the size of the chessboard dividing line cell, the method can be flexibly applied to different usage scenarios, and the applicability is wider.

[0031] The method of connecting the retained point cloud data in sequence to form a continuous edge contour model is: select one of the point cloud data as the starting point, and connect the starting point to all adjacent point cloud data until all the point cloud data are connected. The closed contour formed at this time is the outer surface of the Gaussian model. The method of connecting point cloud data to form edge contours is a relatively mature technical means, such as the Poisson reconstruction algorithm or other point cloud reconstruction algorithms disclosed in the industry, so its principle will not be elaborated in detail. It is worth noting that since this method connects point cloud data, the edge contour models formed are all closed model contours. If the original Gaussian model has an unclosed aperture, the outer surface cannot be completely restored.

[0032] Example 2 like Figure 5 As shown, a system for obtaining a Gaussian model surface based on point cloud data, the system comprising: Model import unit, used to obtain Gaussian model and import it into three-dimensional coordinate system; A bounding box generating unit, used to calculate the maximum length L1 of the Gaussian model along the X-axis direction, the maximum length L2 along the Y-axis direction, and the maximum length L3 along the Z-axis direction, and generate a bounding box whose length, width, and height are L1, L2, and L3 respectively; A segmentation section unit is used to obtain the center point of the bounding box, and to segment the bounding box along three directions perpendicular to the X-axis, the Y-axis and the Z-axis, so that the intersection of the three segments is consistent with the center point; A normal line generating unit, used for making a chessboard dividing line on each of the cut surfaces, and making a normal line perpendicular to the cut surface with each intersection of the chessboard dividing line as a starting point, wherein each normal line includes two directions, a positive direction and a negative direction; A point cloud screening unit is used to record all point cloud data on each normal line, calculate the distance between each point cloud data and the intersection of the line, and find the point cloud data with the largest distance on both the positive normal line and the negative normal line and save it in the surface point set; The outer surface shaping unit is used to remove repeated point cloud data in the surface point set and connect the retained point cloud data in sequence to form a continuous edge contour model.

[0033] The system also includes setting the center point of the bounding box as the coordinate origin of the three-dimensional coordinate system.

[0034] The normal generation unit also includes setting the length and width of the cells divided by the chessboard dividing line.

[0035] The method of drawing a normal line perpendicular to the section surface with the intersection points of the chessboard dividing lines as starting points is: Get the section of the bounding box perpendicular to the X axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the X axis, which is the positive normal line, and emit a ray in the negative direction of the X axis, which is the negative normal line; Get the section of the bounding box perpendicular to the Y axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the Y axis, which is the positive normal line, and emit a ray in the negative direction of the Y axis, which is the negative normal line; Get the slice of the bounding box perpendicular to the Z axis, select the intersection of the chessboard dividing lines on the slice as the starting point to emit a ray in the positive direction of the Z axis, which is the positive normal, and emit a ray in the negative direction of the Z axis, which is the negative normal.

[0036] The method of connecting the retained point cloud data in sequence to form a continuous edge contour model is: select one of the point cloud data as the starting point, and connect the starting point to all adjacent point cloud data until all the point cloud data are connected. The closed contour formed at this time is the outer surface of the Gaussian model.

[0037] Embodiment 2 is essentially the same as Embodiment 1, so the specific working principles of each unit will not be described in detail.

[0038] Example 3 This embodiment proposes a storage medium, which includes a program storage area and a data storage area, wherein the program storage area can store an operating system and programs required for running instant messaging functions, etc.; the data storage area can store various instant messaging information and operation instruction sets, etc. The computer program is stored in the program storage area, and when the computer program is executed by the processor, the method for obtaining a Gaussian model surface based on point cloud data as described in Example 1 is implemented. The processor may include one or more central processing units, or a digital processing unit, etc.

[0039] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

[0040] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A method for obtaining a Gaussian model surface based on point cloud data, characterized in that: The method comprises the following steps: Obtain the Gaussian model and import it into the three-dimensional coordinate system; Calculate the maximum length L1 of the Gaussian model along the X-axis, the maximum length L2 along the Y-axis, and the maximum length L3 along the Z-axis, and generate a bounding box whose length, width, and height are L1, L2, and L3 respectively; Obtain the center point of the bounding box, and divide the bounding box into sections along three directions perpendicular to the X-axis, the Y-axis, and the Z-axis, respectively, so that the intersection of the three sections coincides with the center point; Draw a chessboard dividing line on each of the cut surfaces, and draw a normal line perpendicular to the cut surface with each intersection of the chessboard dividing line as a starting point, wherein each normal line includes two directions, a positive direction and a negative direction; Record all point cloud data on each normal line, and calculate the distance between each point cloud data and the intersection of the line, and find the point cloud data with the largest distance on both the positive normal line and the negative normal line and save them in the surface point set; The repeated point cloud data in the surface point set are removed, and the retained point cloud data are sequentially connected to form a continuous edge contour model.

2. The method for obtaining a Gaussian model surface based on point cloud data according to claim 1, characterized in that: The method for obtaining the center point of the bounding box is: Calculate the coordinates of the vertex (x0, y0, z0) of the bounding box closest to the origin of the 3D coordinate system; Then the x-axis coordinate of the center point is: x=x0+L1 / 2; The y-axis coordinate of the center point is: y=y0+L2 / 2; The z-axis coordinate of the center point is: z=z0+L3 / 2; Get the center point coordinates (x, y, z) of the bounding box.

3. The method for obtaining a Gaussian model surface based on point cloud data according to claim 1 or 2, characterized in that: The method also includes setting the center point of the bounding box as the coordinate origin of the three-dimensional coordinate system.

4. The method for obtaining a Gaussian model surface based on point cloud data according to claim 1, characterized in that: The method of drawing a normal line perpendicular to the section surface with the intersection points of the chessboard dividing lines as starting points is: Get the section of the bounding box perpendicular to the X axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the X axis, which is the positive normal line, and emit a ray in the negative direction of the X axis, which is the negative normal line; Get the section of the bounding box perpendicular to the Y axis, select the intersection of the chessboard dividing lines on the section as the starting point to emit a ray in the positive direction of the Y axis, which is the positive normal line, and emit a ray in the negative direction of the Y axis, which is the negative normal line; Get the slice of the bounding box perpendicular to the Z axis, select the intersection of the chessboard dividing lines on the slice as the starting point to emit a ray in the positive direction of the Z axis, which is the positive normal, and emit a ray in the negative direction of the Z axis, which is the negative normal.

5. The method for obtaining a Gaussian model surface based on point cloud data according to claim 1 or 4, characterized in that: The method also includes setting the length and width of the cells divided by the chessboard dividing line.

6. The method for obtaining a Gaussian model surface based on point cloud data according to claim 1, characterized in that: The method of connecting the retained point cloud data in sequence to form a continuous edge contour model is: select one of the point cloud data as the starting point, and connect the starting point to all adjacent point cloud data until all the point cloud data are connected. The closed contour formed at this time is the outer surface of the Gaussian model.

7. A system for obtaining a Gaussian model surface based on point cloud data, characterized in that: The system includes: Model import unit, used to obtain Gaussian model and import it into three-dimensional coordinate system; A bounding box generating unit, used to calculate the maximum length L1 of the Gaussian model along the X-axis direction, the maximum length L2 along the Y-axis direction, and the maximum length L3 along the Z-axis direction, and generate a bounding box whose length, width, and height are L1, L2, and L3 respectively; A segmentation section unit is used to obtain the center point of the bounding box, and to segment the bounding box along three directions perpendicular to the X-axis, the Y-axis and the Z-axis, so that the intersection of the three segments is consistent with the center point; A normal line generating unit, used for making a chessboard dividing line on each of the cut surfaces, and making a normal line perpendicular to the cut surface with each intersection of the chessboard dividing line as a starting point, wherein each normal line includes two directions, a positive direction and a negative direction; A point cloud screening unit is used to record all point cloud data on each normal line, calculate the distance between each point cloud data and the intersection point of the line, and find the point cloud data with the largest distance on both the positive normal line and the negative normal line and save it in the surface point set; The outer surface shaping unit is used to remove repeated point cloud data in the surface point set and sequentially connect the retained point cloud data to form a continuous edge contour model.

8. The system for acquiring Gaussian model surface based on point cloud data according to claim 7, characterized in that: The system also includes setting the center point of the bounding box as the coordinate origin of the three-dimensional coordinate system.

9. The system for acquiring Gaussian model surface based on point cloud data according to claim 7, characterized in that: The normal generation unit also includes setting the length and width of the cells divided by the chessboard dividing line.

10. A storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for obtaining a Gaussian model surface based on point cloud data as described in any one of claims 1 to 6 is implemented.

Images