A method for extracting visible area of 3D model group based on layered superposition and region growing
By performing perspective projection transformation and layered superposition algorithm on the triangular element of the three-dimensional model, the problem of low visual area extraction efficiency of fine three-dimensional model groups is solved, and efficient and accurate visual area extraction is achieved, suitable for complex three-dimensional model groups Group.
Patent Information
- Application Number
- CN202210008670.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-05
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2042-01-05
AI Technical Summary
The prior art is inefficient when detecting visual area and extracting visual area on fine three-dimensional model groups, and the traditional projection area extraction algorithm is too low in efficiency in complex three-dimensional model groups, resulting in high labor costs and low accuracy.
Using the three-dimensional model group visual area extraction method based on layered superposition and regional growth, the three-dimensional space and two-dimensional are quickly judged by perspective projection transformation of the three-dimensional model triangle element. The mapping model of space optimizes extraction efficiency and accuracy.
It improves the efficiency and accuracy of visual area extraction, is suitable for complex three-dimensional model groups, reduces the complexity of the algorithm, and is suitable for many fields such as architecture, surveying and mapping, and culture. It has high extraction accuracy and fast efficiency.
Smart Images

Figure CN114359226B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of computer technology and relates to a visible area extraction method, in particular to a visible area extraction method based on segmentation of triangular surface primitives of a three-dimensional model. Background Art
[0002] With the rapid development of digital technology and its industry in my country, large-scale rough three-dimensional models are being gradually replaced by fine three-dimensional model groups composed of multiple small-scale fine three-dimensional models due to their lack of refinement and individualization capabilities. A series of new digital products such as refined three-dimensional model groups are playing an increasingly important role in digital city construction, industrial simulation systems, and even cultural and entertainment industries. The development and application of refined three-dimensional model groups are also receiving more and more widespread attention and attention. However, due to the relatively complex geometric structure of the fine three-dimensional and the relatively large amount of data, the existing technical solutions are quite complicated in the development and application operations of the fine three-dimensional model group, such as visible area detection and visible area extraction. Therefore, the present invention conducts in-depth research on how to quickly and conveniently extract the visible area of the fine three-dimensional model group.
[0003] At present, operators can observe independent fine 3D models, such as individual buildings or fine 3D model groups, such as building complexes, at various angles and positions in computer equipment with a simulation environment. For example, for a certain simulated fine 3D model of a certain object, the operator may need to extract the surface area and model surface number of all observable objects in the simulated 3D model of the object under the current viewing angle. For example, in the data analysis process of reconstructing the 3D model of a drone, the operator often needs to manually select the visible object surface, and some partially covered object surfaces cannot be accurately selected, which takes a long time, has a huge selection workload, and has low accuracy, which increases labor costs and reduces extraction efficiency. For multi-component 3D models, the traditional projection area extraction algorithm has the disadvantage of being too inefficient when judging the visible area of the model and extracting the area of the visible area. Summary of the invention
[0004] In order to solve the above problems, the present invention takes a fine three-dimensional model group as the research object and proposes a visible area extraction method for a fine three-dimensional model group based on layered superposition. The method performs perspective projection transformation on the vertices of the triangular surface primitives of the three-dimensional model, quickly judges the vertices of the visible part of the visible surface, improves the efficiency of visible area extraction, and enhances the degree of automation of visible area extraction. The method meets the needs of fast, accurate and stable visible area extraction.
[0005] In order to achieve the above object, the present invention is achieved through the following technical solutions:
[0006] The present invention is a method for extracting visible area of a three-dimensional model group based on layered superposition and region growth, comprising the following steps:
[0007] Step 1: Import the 3D model of the triangular surface primitive, read the vertices of each triangular surface, and construct the vertex coordinate sequence V{(x0, y0, z0), (x1, y1, z1)...(x n ,y n ,z n )};
[0008] Step 2: Perform coordinate transformation and view volume clipping, find all triangles in the view volume, and clip out the part in the view volume; the specific steps are as follows:
[0009] Initialize the camera parameters, including the camera position P(x p ,y p ,z p ), z coordinate f of the far section of the view volume, z coordinate n of the near section of the view volume, y coordinate t of the top plane of the near section of the view volume, y coordinate b of the bottom plane of the near section of the view volume, x coordinate l of the left plane of the near section of the view volume, x coordinate r of the right plane of the near section of the view volume, right vector R of the current camera observation direction, upper vector U of the camera observation direction, direction vector D of the camera observation direction, construct the projection matrix PRO, the observation matrix LOOKAT and the surface equation of the view volume.
[0010] PRO is:
[0011]
[0012] LOOKAT is:
[0013]
[0014] The normalized horizon function is:
[0015]
[0016] The coordinates in V are transformed into perspective projection by the projection matrix PRO according to the following formula, and normalized, and the coordinates after projection transformation are converted into ndc coordinates;
[0017] V=V / Vw
[0018] V′=PRO T *LOOKAT T *V.
[0019] Traverse the vertex coordinate sequence V_ndc of each triangular face of the triangular face model after the transformation, calculate the situation of each triangular face in V_ndc inside and outside the hexahedron surrounded by each face of the view volume, extract the vertex sequence of the triangular face piece that is partially inside the view volume and the vertex sequence of the triangular face piece that is completely inside the view volume, which are V1 and V2 respectively. For the vertex coordinate vector V1 of the triangular face primitive that is partially inside the view volume, traverse and determine the intersection of the view volume plane and each triangular face in the container V1. If they intersect, find the intersection point and reassemble them into a new triangular face, perform triangular face segmentation, and merge the segmented triangular face set with V2 to obtain the triangular face vertex sequence set LT inside the view volume i .
[0020] Step 3: Perform projection transformation of the 3D model. i , LT i The vertices of the middle triangle are orthogonally projected onto the near section of the view volume, and the x and y coordinates of each triangle vertex are taken to convert the three-dimensional coordinates into two-dimensional coordinates.
[0021] Step 4: Calculate the projection area conversion coefficient and construct a mapping relationship between three-dimensional space and two-dimensional space. The mapping relationship refers to the ratio of the area of the triangle in three-dimensional space to the projection area of the triangle after projection deformation on the far plane of the view volume. Calculate the area S of each triangle after orthogonal projection i _p and the unprojected triangular surface area S i_ a, so that
[0022] λ i =S i_ p / S i_ a
[0023] Among them, λ i is the current projection area conversion coefficient.
[0024] Step 5: Select and create the initial polygon poly0, and create the initial polygon chain structure POLY;
[0025] {num_of_link; / / Number of links
[0026] Linkset
[0027] {pointset / / chain vertex;
[0028] direct_of_ply; / / Edges and directions of the ring chain;
[0029] attribute / / each link attribute}
[0030] / / chain collection}
[0031] Here, the attribute is used to distinguish whether the ring chain is an internal island ring chain or an external contour ring chain of a complex polygon. i All triangles in the triangle sequence LT i All triangular faces in the are sorted from small to large according to the minimum z coordinate value of the three vertices of the triangular face, that is, the triangular faces are arranged from front to back, and the triangle determined in the x and y coordinate space of the triangular face LT0 with the smallest z coordinate value is used as the initial polygon face poly0 and is stored in the structure of a polygon ring chain.
[0032] Step 6: Calculate the visible area of the projected 3D model, i.e. the area of polygon poly0;
[0033] Specifically, the step 6 is as follows: the area of the visible region of the projected three-dimensional model, i.e., the area of the polygon poly0, is calculated as:
[0034]
[0035] Among them, Sm is the visible area, Sm i is the visible area of each triangle during the iteration, λ i is the current projection area conversion coefficient,
[0036] Sm i The calculation process is: extract LT i The x and y coordinates of the vertices of the middle triangle surface are used to obtain the current triangle sequence set LTO to be judged i , judge from back to front, and the judgment results are:
[0037] 5-1. If the three vertices of triangle LTO1 are completely inside poly0, then i and LT i Eliminate LTO1 and the corresponding LT1, and update poly0 to poly1;
[0038] 5-2. If the triangle LTO1 intersects with poly0, find the intersection of LTO1 area S1 and poly0 OVER and its area Sover.
[0039] Sm1=λ0Sp+λ1(S1-Sover)
[0040] And from LTO i and LT i Eliminate LTO1 and the corresponding LT1, and find the union of LTO1 and poly0 as poly1;
[0041] 5-3. The triangle LTO1 is separated from poly0. LTO1 is included in poly0 as the outer ring of the complex polygon contour, and poly0 is updated to poly1. Then
[0042] Sm1=λ0Sp+λ1S1
[0043] And from LTO i and LT i LTO1 and the corresponding LT1 are excluded.
[0044] Step 7: Repeat step 6 until LTO is traversed i All triangles in the 3D model group are obtained to obtain the visible area Sm n .
[0045] The beneficial effects of the present invention are:
[0046] The present invention adopts the method of projecting the three-dimensional model onto the near cross section of the horizon volume, converting the three-dimensional space into a two-dimensional space, which greatly reduces the complexity of visual area extraction; at the same time, the initial area is selected according to the bounding box range of the projected triangle surface set, and the regional growth and layered superposition algorithms are combined to further optimize the extraction efficiency and algorithm complexity;
[0047] The present invention establishes a mapping model between three-dimensional space and two-dimensional projection space by introducing projection area transformation coefficients, and expands the function of the algorithm so that it can not only extract the visible area of the model group after projection, but also extract important information such as the visible range of the model group and the area of the visible range;
[0048] The present invention has a wide range of applications. Traditional projection area extraction algorithms are mostly applicable to single simple three-dimensional models, while this method is suitable for extracting the visible area of complex three-dimensional model groups. It can be widely used in many fields such as architecture, surveying and mapping, culture, agriculture and forestry, and has relatively high extraction accuracy and relatively high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 Enter the original triangular model diagram for the instance.
[0050] Figure 2 It is a schematic diagram of the triangular face model after the view volume is clipped in the embodiment.
[0051] Figure 3 Schematic diagram of the positional relationship between LTO2 and poly1 in step 1 of the embodiment.
[0052] Figure 4 Schematic diagram of the positional relationship between LTO3 and poly2 in step 2 of the embodiment.
[0053] Figure 5 Schematic diagram of the positional relationship between LTO4 and poly3 in step 3 of the embodiment.
[0054] Figure 6Schematic diagram of the positional relationship between LTO5 and poly4 in step 4 of the embodiment.
[0055] Figure 7 Schematic diagram of the positional relationship between LTO6 and poly5 in step 5 of the embodiment.
[0056] Figure 8 Schematic diagram of the positional relationship between LTO7 and poly6 in step 6 of the embodiment.
[0057] Fig. 9 Schematic diagram of the positional relationship between LTO8 and poly7 in step 7 of the embodiment.
[0058] Fig.10 Schematic diagram of the positional relationship between LTO8 and poly8 in step 8 of the embodiment. DETAILED DESCRIPTION
[0059] The following will disclose the embodiments of the present invention with drawings. For the purpose of clear description, many practical details will be described together in the following description. However, it should be understood that these practical details should not be used to limit the present invention. That is to say, in some embodiments of the present invention, these practical details are not necessary.
[0060] The present invention is a method for extracting the visible area of a three-dimensional model group based on layered superposition and region growth. The method for extracting the visible area of a three-dimensional model group comprises the following steps:
[0061] Step 1: Use the assimp library as a reading and writing tool, read in the triangular face format, and obtain the vertex coordinates and face sequence of each triangular face in the model. In this example, a vector container LT_a and a triangular face model vertex sequence will be constructed:
[0062] {(40,60,0),(60,60,0),(60,40,0),
[0063] (50,50,10),(70,50,10),(70,60,10),
[0064] (70,60,20),(70,30,20),(60,30,20),
[0065] (40,40,30),(40,70,30),(20,40,30),
[0066] (40,10,50),(10,50,50),(40,50,50),
[0067] (80,10,60),(80,90,60),(20,90,60),
[0068] (80,10,70),(80,40,70),(50,40,70),
[0069] (40,80,80),(70,80,80),(70,70,80),
[0070] (10,10,90),(20,10,90),(10,20,90),(150,50,50),(170,50,50),(150,30,50)}
[0071] The vertex coordinate data of each face is stored in the vector container LT_a in sequence. Figure 1 shown.
[0072] Step 2: Initialize the camera parameters, including the camera position P(50,50,-20), the field of view angle β(45°), the aspect ratio of the view interface a(1), the far section distance of the view volume F(100), the near section distance of the view volume N(0), the camera pitch angle γ(0°), and the roll angle Yaw angle λ(0°), construct projection matrix PRO and observation matrix LOOKAT. Among them:
[0073]
[0074]
[0075] The sequence of triangle vertices in the projected ndc coordinate system is LT_a'
[0076] LT_a′=RPO*LOOKAT*LT_a
[0077]
[0078] In this example, the first three dimensions of LT_a' are {(0.5,-0.5,1),(-0.5,-0.5,1),(-0.5,0.5,1),
[0079] (-0,-0,1),(-0.666667,-0,1),(-0.666667,-0.333333,1),
[0080] (-0.5,-0.25,1),(-0.5,0.5,1),(-0.25,0.5,1),
[0081] (0.2,0.2,1),(0.2,-0.4,1),(0.6,0.2,1),
[0082] (0.142857,0.571429,1),(0.571429,-0,1),(0.142857,-0,1),
[0083] (-0.375,0.5,1),(-0.375,-0.5,1),(0.375,-0.5,1),
[0084] (-0.333333,0.444444,1),(-0.333333,0.111111,1),(-0,0.111111,1),
[0085] (0.1,-0.3,1),(-0.2,-0.3,1),(-0.2,-0.2,1),
[0086] (0.363636,0.363636,1),(0.272727,0.363636,1),(0.363636,0.272727,1),
[0087] (-1.42857,-0,1),(-1.71429,-0,1),(-1.42857,0.285714,1)}.
[0088] Step 3: Perform view volume clipping. This embodiment proposes the following solution:
[0089] 1. Traverse the vector LT_a', construct the coordinate range of each triangle, and extract the triangle LT_a'1 where the coordinate range of the triangle patch intersects with the coordinate range of the view volume.
[0090] 2. Get the vertex coordinate vector LT_a'1 of the triangle face primitive that intersects with the coordinate range of the view volume, traverse and determine the positional relationship between the view volume plane and each triangle face in the container LT_a'1. If the triangle face is completely within the view volume, it is retained; if it is partially within the view volume, it is intersected with the part of the triangle face plane within the view volume and the obtained intersection part is retained; otherwise, it is discarded.
[0091] 3. For the intersection of the triangular planes in the view volume, convert them into triangular faces. Get the projected triangular face sequence LT in the view volume i :
[0092] {(0.5,-0.5,1),(-0.5,-0.5,1),(-0.5,0.5,1),
[0093] (-0,-0,1),(-0.666667,-0,1),(-0.666667,-0.333333,1),
[0094] (-0.5,-0.25,1),(-0.5,0.5,1),(-0.25,0.5,1),
[0095] (0.2,0.2,1),(0.2,-0.4,1),(0.6,0.2,1),
[0096] (0.142857,0.571429,1),(0.571429,-0,1),(0.142857,-0,1),
[0097] (-0.375,0.5,1),(-0.375,-0.5,1),(0.375,-0.5,1),
[0098] (-0.333333,0.444444,1),(-0.333333,0.111111,1),(-0,0.111111,1),
[0099] (0.1,-0.3,1),(-0.2,-0.3,1),(-0.2,-0.2,1),
[0100] (0.363636,0.363636,1),(0.272727,0.363636,1),(0.363636,0.272727,1)}.
[0101] The triangles after clipping the view volume are as follows Figure 2 shown.
[0102] Step 4: Perform visual surface judgment. This embodiment proposes the following solution:
[0103] (1) Read the triangle vertex sequence set LT inside the view body after clipping i , LT i The vertices of the middle triangle are orthogonally projected onto the near section of the view volume, and the x and y coordinates of each triangle vertex are taken to convert the three-dimensional coordinates into two-dimensional coordinates.
[0104] (2) Calculate the projection area conversion coefficient and construct the mapping relationship between three-dimensional space and two-dimensional space. Calculate the area S of each triangle after orthogonal projection i _p and the unprojected triangular surface area S i_ a, and obtain the projection area conversion coefficient λ i Here, the area of the triangle is solved using Heron's formula. In this example, λ i The sequence is:
[0105] λ i=(0.0025, 0.00111111, 0.000625, 0.0004, 0.000204082, 0.00015625, 0.000123457, 0.0001, 0.0000826447, 0.000204082)
[0106] (3) Select and create the initial polygon. i For all triangles in LT i All triangular faces in the array are sorted from small to large by the minimum z coordinate value of the three vertices of the triangular face, that is, the triangular faces are arranged from front to back, and the triangle determined in the x and y coordinate space of the triangular face LT0 with the smallest z coordinate value is used as the initial polygon face poly0 and stored in the structure of polygon ring chain. And create the initial polygon ring chain structure
[0107] POLY
[0108] {int num_of_link=1; / / Number of links
[0109] Linkset
[0110] {int pointset = 3; / / chain vertex
[0111] Vector <vec2>direct_of_ply=((1,0),(1,1),(0,-1)); / / Edges and directions of the chain
[0112] int attribute = true; / / each link attribute}
[0113] / / chain collection}
[0114] (4) Poly0 is used as the initial polygon, and the area of the visible area after the current projection Sp is calculated, that is, the area of the current polygon poly0 and the area of the model visible area Sm i .
[0115] Sm0=λ0Sp=200*0.0025=0.5
[0116] Extract LT i The x and y coordinates of the vertices of the middle triangle surface are used to obtain the current triangle sequence set LTO to be judged i . Judge from back to front. Here the polygon area is solved using the following formula:
[0117]
[0118] Here O is the origin of coordinates, P i are the vertices of the polygon and s is the area.
[0119] Step 1: As attached Figure 3 As shown, LTO1 partially intersects with poly0, then Sm1=λ0Sp+λ1(S1-Sover), where Sover=25, S1=100, update the union and obtain poly1.
[0120] Step2: As attached Figure 4 As shown, LTO2 partially intersects with poly1, then Sm2=Sm1+λ2(S2-Sover), where Sover=16.67, S1=150, update the union and obtain poly2.
[0121] Step 3: As attached Figure 5 As shown, LTO3 is an independent triangle, then Sm3=Sm2+λ3S3, LTO3 is stored in poly2 as an outer ring and poly2 is updated to poly3.
[0122] Step 4: As attached Figure 6 As shown, LTO4 partially intersects with poly3, then Sm4=Sm3+λ4(S4-Sover), update the union and obtain poly4.
[0123] Step5: As attached Figure 7 As shown, LTO5 partially intersects with poly4, then Sm5=Sm4+λ5(S5-Sover), update the union and obtain poly5.
[0124] Step 6: As attached Figure 8 As shown, LTO6 partially intersects with poly5, then Sm6=Sm5+λ6(S6-Sover), update the union and obtain poly6.
[0125] Step7: As attached Fig. 9 As shown, LTO7 is completely included in poly6, then Sm7=Sm6, and poly6 is updated to poly7.
[0126] Step8: As attached Fig.10 As shown, LTO8 is an independent triangle, then Sm8 = Sm7 + λ8S8, update and calculate, and obtain poly8, LT i All triangles in the image are traversed, and Sm = 1.320196 is obtained. When restored to the world coordinate system, Sm = 3300.49.
[0127] The present invention discloses a method for extracting the visible area of a triangular face model based on a layered superposition method. By projecting other triangular faces onto a two-dimensional plane, the invisible area is layered and superimposed to obtain the total visible area and area of the model, thereby improving the extraction efficiency compared to traditional three-dimensional model visible area extraction.
[0128] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims
1. A method for extracting visible area of a three-dimensional model group based on layered superposition and region growing, characterized in that: The method for extracting the visible area of a three-dimensional model group includes the following steps: Step 1: Import the 3D model of the triangular surface primitive, read the vertices of each triangular surface, and construct the vertex coordinate sequence V{(x0, y0, z0), (x1, y1, z1)...(x n ,y n ,z n )}; Step 2: Model view transformation: Perform coordinate transformation and view volume clipping on the fixed-point coordinate sequence obtained in step 1, find all the triangles in the view volume, and clip out the triangle sequence LT of the part in the view volume i ; Step 3: Perform projection transformation of the 3D model: transform the triangle face sequence LT of the part within the view volume cut in step 2 i Orthogonally project the fixed points of the triangular face in the image onto the near cross section of the view volume, and take the x and y coordinates of the vertices of each triangular face to convert the three-dimensional coordinates into two-dimensional coordinates; Step 4: Calculate the projection area conversion coefficient and construct a mapping relationship between three-dimensional space and two-dimensional space. The mapping relationship refers to the ratio λ0 of the area of the triangle in three-dimensional space to the projection area of the triangle after projection deformation on the far plane of the view volume; Step 5: Select and create the initial polygon poly0, and create the initial polygon chain structure POLY; Step 6: Calculate the visible area of the projected 3D model, i.e. the area of polygon poly0; Step 7: Repeat step 6 until LTO is traversed i All triangles in the 3D model group are obtained to obtain the visible area Sm n ; Wherein, the step 2 specifically includes the following steps: Step 2-1: Initialize camera parameters, including camera position P(x p ,y p ,z p ), the z coordinate of the far section of the view volume f, the z coordinate of the near section of the view volume n, the y coordinate of the top plane of the near section of the view volume t, the y coordinate of the bottom plane of the near section of the view volume b, the x coordinate of the left plane of the near section of the view volume l, the x coordinate of the right plane of the near section of the view volume r, the right vector R of the current camera observation direction, the vector U of the camera observation direction, the direction vector D of the camera observation direction, construct the projection matrix PRO, the observation matrix LOOKAT and the view volume surface equation, where: PRO is: LOOKAT is: The normalized horizon function is: The coordinates in V are transformed into perspective projection by the projection matrix PRO according to the following formula, and normalized, and the coordinates after projection transformation are converted into ndc coordinates; V=V / Vw V′=FOR T *LOOKAT T *IN; Step 2-2: Traverse the vertex coordinate sequence V_ndc of each triangular face of the transformed triangular face model, calculate the situation of each triangular face in V_ndc inside and outside the hexahedron surrounded by each face of the view volume, extract the vertex sequence of the triangular face that is partially inside the view volume and the vertex sequence of the triangular face that is completely inside the view volume, which are V1 and V2 respectively; Step 2-3: For the vertex coordinate vector V1 of the triangular face primitives in the view volume, traverse and determine the intersection of the view volume plane and each triangular face in the container V1. If they intersect, find the intersection point and reassemble them into new triangular faces, split the triangular faces, and merge the split triangular face set with V2 to obtain the triangular face vertex sequence set LT in the view volume. i .
2. The method for extracting visible area of a three-dimensional model group based on layered superposition and region growing according to claim 1, characterized in that: The step 5 is specifically as follows: traverse the triangular face sequence LT i All triangles in the triangle sequence LT i All triangular faces in the are sorted from small to large according to the minimum z coordinate value of the three vertices of the triangular face, that is, the triangular faces are arranged from front to back, and the triangle determined in the x and y coordinate space of the triangular face LT0 with the smallest z coordinate value is used as the initial polygon face poly0 and is stored in the structure of a polygon ring chain.
3. The method for extracting visible area of a three-dimensional model group based on layered superposition and region growing according to claim 1, characterized in that: Specifically, the step 6 is as follows: the area of the visible region of the projected three-dimensional model, i.e., the area of the polygon poly0, is calculated as: Among them, Sm is the visible area, Sm i is the visible area of each triangle during the iteration, λ i is the current projection area conversion coefficient.
4. The method for extracting visible area of a three-dimensional model group based on layered superposition and region growing according to claim 3, characterized in that: The visible area Sm of each triangle during the iteration process i The calculation process is: extract LT i The x and y coordinates of the vertices of the middle triangle surface are used to obtain the current triangle sequence set LTO to be judged i , judge from back to front, and the judgment results are: 5-1. If the three vertices of triangle LTO1 are completely inside poly0, then i and LT i Eliminate LTO1 and the corresponding LT1, and update poly0 to poly1; 5-2. If the triangle LTO1 intersects with poly0, find the intersection of LTO1 area S1 and poly0 OVER and its area Sover. Sm1=λ0Sp+λ1(S1-Sover) And from LTO i and LT i Eliminate LTO1 and the corresponding LT1, and find the union of LTO1 and poly0 as poly1; 5-3. The triangle LTO1 is separated from poly0. LTO1 is included in poly0 as the outer ring of the complex polygon contour, and poly0 is updated to poly1. Then Sm1=λ0Sp+λ1S1 And from LTO i and LT i LTO1 and the corresponding LT1 are excluded.
5. The method for extracting visible area of a three-dimensional model group based on layered superposition and region growing according to claim 1, characterized in that: In step 4, the projection area conversion coefficient calculation formula is: l i =S i _p / S i _a Among them, λ i is the current projection area conversion coefficient; S i _p is the area of each triangle after orthogonal projection, S i _a is the area of each triangle face without projection.
