CN114494641B - Three-dimensional model light weight method and device - Google Patents

Abstract

The invention discloses a three-dimensional model light-weight method and device, and relates to the technical field of image processing. The method comprises the following steps: reconstructing the three-dimensional model according to the tile data of the three-dimensional model to obtain a topological continuous triangular grid model; performing surface segmentation according to the plane characteristics of the triangular mesh model to obtain segmented flat surface areas after the whole three-dimensional model is segmented; carrying out plane difference degree calculation on adjacent division surfaces according to the division flat surface areas to obtain a combined block; screening geometric feature points in boundary line nodes according to boundary lines of the combined blocks, and taking a space surface formed by the geometric feature points as a lightweight block surface of the three-dimensional model; and generating new textures by adopting a space orthographic projection method according to the space region range covered by the light-weight blocking surface to obtain a light-weight three-dimensional model. The invention can improve the simplification efficiency and the simplification quality of the three-dimensional model, and realize the optimized storage and the high-efficiency loading of the three-dimensional model.

Description

A three-dimensional model lightweight method and device

Technical Field

The present invention relates to the field of image processing technology, and in particular to a three-dimensional model lightweight method and device.

Background Art

With the development of network technology and computer technology, three-dimensional visualization technology has also developed rapidly and has been widely used. For example, urban three-dimensional models are an important part of urban digital infrastructure, including urban planning, environmental monitoring, spatial information analysis, etc. With the continuous development of digital cities and new surveying and mapping technologies, the accuracy of the obtained three-dimensional models is getting higher and higher. The display of these three-dimensional models has high requirements for hardware performance such as computing performance, physical storage space, memory space, and GPU rendering capabilities. For a long time, it can only rely on personal computers with better hardware conditions. In addition, due to the large data size of the three-dimensional model and the long transmission time, the display of the three-dimensional model in the visualization interface is stuck, so the three-dimensional model needs to be lightweight.

Summary of the invention

The object of the present invention is to provide a three-dimensional model lightweight method to improve the simplification efficiency and simplification quality of the three-dimensional model and realize the optimized storage and efficient loading of the three-dimensional model.

To achieve the above object, an embodiment of the present invention provides a three-dimensional model lightweight method, comprising:

Reconstruct the three-dimensional model according to the tile data of the three-dimensional model to obtain a topologically continuous triangular grid model;

Performing face segmentation according to the plane features of the triangular mesh model to obtain segmented flat face domains after the overall three-dimensional model is segmented;

Calculating the plane difference of adjacent segmented surfaces according to the block flattened surface domain, merging adjacent segmented surfaces whose plane difference is within a preset threshold range to obtain a merged block;

According to the boundary line of the merged block, the geometric feature points in the boundary line nodes are screened, and the spatial surface composed of the geometric feature points is used as the lightweight block surface of the three-dimensional model;

According to the spatial area covered by the lightweight block surface, a new texture is generated by using a spatial orthographic projection method to obtain a lightweight three-dimensional model.

Preferably, the three-dimensional model is reconstructed according to the tile data of the three-dimensional model to obtain a topologically continuous triangular mesh model, including:

The tile data in the selected area of the three-dimensional model is obtained, and the tiles of the specified accuracy level in the tile data are reconstructed by using a tile merging method to generate a topologically continuous triangular mesh model.

Preferably, the surface segmentation is performed according to the plane features of the triangular mesh model to obtain the segmented flat surface areas of the overall three-dimensional model, including:

The local fitting plane of the triangular grid model is determined based on the K-means clustering algorithm, and the model grid vertices are classified to obtain the block flat surface domain after the overall three-dimensional model is segmented.

Preferably, the calculating of the plane difference of adjacent segmented surfaces according to the block flattened surface domain, merging adjacent segmented surfaces whose plane difference is within a preset threshold range to obtain a merged block includes:

According to the curvature of the boundary connection of the 3D model blocks, the angle between the normal vectors of the triangular grid at the segmentation of the 3D model is calculated to obtain the plane difference;

Merge adjacent segmented surfaces with smaller plane differences to obtain merged blocks.

Preferably, the step of screening geometric feature points in the boundary line nodes according to the boundary line of the merged block and using the spatial surface composed of the geometric feature points as the lightweight block surface of the three-dimensional model includes:

Based on the moving screening method, the nodes in the block boundary are selected and judged, the spatial fitting straight lines of some nodes on the model boundary line are calculated, the geometric feature points in the boundary line nodes are screened, and the spatial surface composed of the geometric feature points is used as the lightweight block surface of the model.

The embodiment of the present invention further provides a three-dimensional model lightweight device, comprising:

A tile merging module is used to reconstruct the three-dimensional model according to the tile data of the three-dimensional model to obtain a topologically continuous triangular mesh model;

A plane segmentation module is used to perform surface segmentation according to the plane features of the triangular mesh model to obtain a flat surface area after segmentation of the overall three-dimensional model;

A block merging module is used to calculate the plane difference of adjacent segmented surfaces according to the block flattened surface domain, and merge adjacent segmented surfaces whose plane difference is within a preset threshold range to obtain a merged block;

A feature screening module, used to screen geometric feature points in the boundary line nodes according to the boundary lines of the merged blocks, and use the spatial surface composed of the geometric feature points as the lightweight block surface of the three-dimensional model;

The space projection module is used to generate a new texture by using a space orthographic projection method according to the space area covered by the lightweight block surface, so as to obtain a lightweight three-dimensional model.

Preferably, the tile merging module is further used for:

The tile data in the selected area of the three-dimensional model is obtained, and the tiles of the specified accuracy level in the tile data are reconstructed by using a tile merging method to generate a topologically continuous triangular mesh model.

Preferably, the plane segmentation module is further used for:

The local fitting plane of the triangular grid model is determined based on the K-means clustering algorithm, and the model grid vertices are classified to obtain the block flat surface domain after the overall three-dimensional model is segmented.

Preferably, the block merging module is further used for:

According to the curvature of the boundary connection of the 3D model blocks, the angle between the normal vectors of the triangular grid at the segmentation of the 3D model is calculated to obtain the plane difference;

Merge adjacent segmented surfaces with smaller plane differences to obtain merged blocks.

Preferably, the feature screening module is further used for:

Based on the moving screening method, the nodes in the block boundary are selected and judged, the spatial fitting straight lines of some nodes on the model boundary line are calculated, the geometric feature points in the boundary line nodes are screened, and the spatial surface composed of the geometric feature points is used as the lightweight block surface of the model.

Compared with the prior art, the present invention has the following beneficial effects:

The three-dimensional model lightweight method provided by the present invention reconstructs the three-dimensional model through the tile data of the three-dimensional model, performs face segmentation according to the plane features of the triangular mesh model, calculates the plane difference between adjacent segmented faces, selects the geometric feature points in the boundary line nodes, uses the space surface composed of the geometric feature points as the lightweight block surface of the three-dimensional model, generates a new texture by using the space orthographic projection method, and realizes texture mapping from the texture image to the simplified three-dimensional model. The present invention can improve the simplification efficiency and simplification quality of the three-dimensional model, and realize the optimized storage and efficient loading of the three-dimensional model.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

FIG1 is a schematic diagram of a process of a three-dimensional model lightweight method provided by an embodiment of the present invention;

FIG2 is a schematic diagram of extracting adjacent boundaries before merging triangular tiles provided by an embodiment of the present invention;

FIG3 is a schematic diagram of triangular tiles after merging according to an embodiment of the present invention;

FIG4 is a schematic structural diagram of a three-dimensional model lightweight device provided by an embodiment of the present invention;

FIG. 5 is a schematic diagram of the structure of a computer terminal device provided in an embodiment of the present invention.

DETAILED DESCRIPTION

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.

It should be understood that the step numbers used in this article are only for the convenience of description and are not intended to limit the order in which the steps are executed.

It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and/or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or combinations thereof.

The term "and/or" means and includes any and all possible combinations of one or more of the associated listed items.

Please refer to Figure 1, which is a schematic diagram of a process flow of a three-dimensional model lightweight method provided by an embodiment of the present invention. In this embodiment, the three-dimensional model lightweight method includes the following steps:

S110, reconstructing the three-dimensional model according to the tile data of the three-dimensional model to obtain a topologically continuous triangular mesh model;

S120, performing surface segmentation according to the plane features of the triangular mesh model to obtain segmented flat surface regions of the overall three-dimensional model;

S130, calculating the plane difference of adjacent segmented surfaces according to the block flattened surface domain, and merging adjacent segmented surfaces whose plane difference is within a preset threshold range to obtain a merged block;

S140, according to the boundary line of the merged block, screening geometric feature points in the boundary line nodes, and using the spatial surface composed of the geometric feature points as the lightweight block surface of the three-dimensional model;

S150, generating a new texture by using a spatial orthographic projection method according to the spatial area covered by the lightweight block surface, to obtain a lightweight three-dimensional model.

In a certain embodiment, step S110, reconstructing the three-dimensional model according to the tile data of the three-dimensional model to obtain a topologically continuous triangular mesh model, includes: obtaining tile data in a selected area of the three-dimensional model, using a tile merging method to reconstruct a model of tiles of a specified accuracy level in the tile data, and generating a topologically continuous triangular mesh model.

Specifically, in some embodiments, based on the tile pyramid model of the three-dimensional model measured by oblique photogrammetry, the tile pyramid level of the required accuracy is first determined; then grouping is performed according to the spatial position relationship, and spatially adjacent sub-tiles are grouped into one group, and the adjacent edges of each group of sub-tiles are extracted, see Figure 2. The number of vertices in the adjacent edges of adjacent sub-tiles is compared. When the number of vertices is different, the edge folding operation is performed on the edge with a larger number to reduce the number of vertices in its edge until the number of vertices on the two adjacent edges is equal. Finally, the adjacent vertices are alternately selected in spatial order as the connecting triangular mesh of the triangular tiles, and all the vertex coordinates, normal vectors and texture information in the two tiles are merged into one file to complete the merging of the two tiles, see Figure 3. Similarly, the above process is iterated until all tiles are merged into an integral triangular mesh.

In a certain embodiment, step S120, performing surface segmentation according to the plane features of the triangular mesh model to obtain the block flat surface domain after the overall three-dimensional model is segmented, including: determining the local fitting plane of the triangular mesh model based on the K-means clustering algorithm, classifying the model mesh vertices, and obtaining the block flat surface domain after the overall three-dimensional model is segmented.

Specifically, in some embodiments, the triangular mesh vertices are first classified using the K-means clustering method, and then the overall three-dimensional model is fitted using k planes, and each fitting plane is determined by the three-dimensional spatial position of the three-dimensional model vertices, so that the distance between each vertex and the fitting plane to which it belongs is close enough. The specific method for determining the fitting plane of the triangular mesh vertices is as follows:

Assume that the objective function of fitting the plane of the triangular grid vertices in a certain area is:

ax+by+cz=d

Wherein, ω＝(a, b, c) is a vector and satisfies: ||ω|| ₂ ＝1.

Center the set of vertices:

The sum of the squares of the distances from each model vertex in the cluster to the fitting plane is:

x _i , y _i , _zi are the spatial coordinates of the model vertices. According to the plane fitting goal, the minimum value of the above formula should be taken. Using the Lagrange multiplier method, we can get:

Taking partial derivatives we get:

Right now:

in,

为拟合平面上的一点。Perform eigenvalue decomposition on the matrix X ^T X, obtain the eigenvalues λ ₁ ≥λ ₂ ≥λ ₃ , and obtain the eigenvectors v ₁ , v ₂ , v ₃ . Take v ₃ as the normal vector of the fitting plane,

is a point on the fitting plane.

The equation of the fitted plane is:

Define the distance function:

When the number of vertices in the current cluster _Cj is less than 3, the distance is measured using the Euclidean distance.

dist _ed (p _i , u _j )=||p _i -u _j || ₂

When the number of vertices in the current cluster _Cj is ≥ 3, the distance is measured as "weighted distance", the distance dist _pa from the model vertex to the fitting plane is calculated, and then the weighted distance is calculated:

dist _wd (p _i , u _j )=w ₁ ·dist _pa +w ₂ ·||p _i -u _j || ₂

w₁和w₂为加权系数，w₁≥0，w₂≥0，w₁+w₂＝1。in

w ₁ and w ₂ are weighting coefficients, w ₁ ≥0, w ₂ ≥0, w ₁ +w ₂ =1.

For a given noise grid model D = {p ₁ , p ₂ , ... , p _m }, k model vertices {u ₁ , u ₂ , ... , _uk } are selected from the grid model as initial clustering seed points, and the final clustering result C = {C ₁ , C ₂ , ... , C _k } is obtained through the following clustering algorithm.

The specific process includes:

In a certain embodiment, step S130, according to the block flat surface domain, the plane difference of adjacent segmented surfaces is calculated, and the adjacent segmented surfaces whose plane difference is within a preset threshold range are merged to obtain a merged block, including: according to the curvature of the boundary connection of the three-dimensional model block, the angle between the normal vectors of the triangular grid at the segmentation of the three-dimensional model is calculated to obtain the plane difference; adjacent segmented surfaces with smaller plane difference are merged to obtain a merged block.

In this embodiment, the calculation of the plane difference includes: calculating the angle θ between the normal vectors of the triangular grid at the model segmentation point, setting the threshold angle δ (δ = 15°), if θ<δ, the angles of the blocks on both sides are within a reasonable threshold range, recorded as small curvature edges, otherwise recorded as large curvature edges. Finally, the proportion u of large curvature edges between adjacent triangular tiles at the block boundary connection is counted, and u is the plane difference. Generally speaking, in order to obtain the merged block, adjacent segmentation surfaces with smaller plane differences are selected for merging. As a preferred solution, if u is less than 75%, the two blocks are merged, and if it is greater than 75%, they are considered to be two independent areas.

Specifically, in some embodiments, through step S120, the model grid clustering segmentation method divides the oblique photogrammetry 3D model D into k block grids, namely C _segm ={C ₁ ,C ₂ ,…,C _k }. To ensure the rationality of model segmentation, when selecting model grid clustering seed points, the sampling density of seed points is relatively large, which will cause too many model blocks after segmentation and the phenomenon of a large number of fragmented model blocks. It is necessary to merge the segmented model blocks to reduce the segmented blocks.

Taking into account the plane properties and surface properties of the three-dimensional model, the model block is measured by the curvature of the model block boundary connection. Calculate the angle θ between the normal vectors of the triangular grid at the model segmentation point, set the threshold δ, if θ<δ, the angle between the blocks on both sides is within a reasonable threshold range, indicating that the curvature is small, otherwise the curvature difference is large. Finally, count the proportion of angles between adjacent triangular tiles less than the threshold δ at all block boundary connections. If it is less than 75%, the two blocks are merged. If it is greater than 75%, they are considered to be two independent regions. The specific implementation of the region merging algorithm includes:

In a certain embodiment, step S140, according to the boundary line of the merged block, screens the geometric feature points in the boundary line nodes, and uses the spatial surface composed of the geometric feature points as the lightweight block surface of the three-dimensional model, including: selecting and judging the nodes in the block boundary based on the moving screening method, calculating the spatial fitting straight line of some nodes of the model boundary line, screening the geometric feature points in the boundary line nodes, and using the spatial surface composed of the geometric feature points as the lightweight block surface of the model.

Specifically, in some embodiments, after the block models are merged, there are a large number of nodes at the block boundaries, most of which are redundant boundary points. The boundary points need to be screened to retain their geometric feature points. Considering the spatial feature attributes of the triangular grid, the boundary lines of the triangular grid are first extracted, and then the moving screening method is used to select the geometric feature points of each point in the boundary, and finally only the selected geometric feature points are retained.

When using the moving screening method to determine whether the current vertex is a geometric feature point, based on the spatial coordinates of the current vertex and the first two topologically adjacent points and the last two topologically adjacent points on the boundary line, a total of five points, it is necessary to first determine a spatial straight line that is closest to them.

Assume that a straight line in the triangular grid space passes through a point (x ₀ , y ₀ , z ₀ ), the standard equation of the space straight line is:

After finishing, we can get:

in:

The above formula is the equation of two spatial planes. The intersection of the spatial planes determines a unique straight line. Determining the spatial fitting straight line can be transformed into determining two spatial fitting planes. The solution goal after the transformation is to determine the parameters a, b, c, and d so that the sum of the Euclidean distances from each discrete point in the grid space to the two spatial planes is minimized.

It can be expressed in matrix form as:

将上式转为误差矩阵方程为：Assumptions

Convert the above equation into the error matrix equation:

That is: V = ωX-b

in:

Because there are increments in the three directions of x, y, and z at the grid boundary points, the matrix ω contains the coordinate variable z, which contains random errors, so the error matrix equation V is an equation whose coefficient matrix contains errors.

Based on formula (1), the adjustment criterion is introduced:

Substituting equation (1) into the above equation, we can differentiate each element in the matrix ω and the parameter vector X, and obtain the iterative equation by dividing the equation into two categories:

in,

The specific solution process includes:

After obtaining the spatial linear fitting equation of the selected vertex, determine whether the vertex is a feature point of the block grid model based on the relative distance from the vertex to the spatial straight line. Finally, use the point set of all feature points to form a simplified model block surface.

When selecting the feature points of each model segmentation surface using spatial linear fitting boundary contour points, in order to avoid breaks in the simplified model, after being selected as a model feature point in one model segmentation surface, it also becomes a feature point of the adjacent segmentation surface.

Specific implementation methods include:

In one embodiment, step S150 generates a new texture by using a spatial orthographic projection method according to the spatial area covered by the lightweight block surface of the model, thereby realizing texture mapping after the model triangular mesh is simplified.

Specifically, in some embodiments, texture mapping is to calculate the texture coordinates (u, v) of the three-dimensional model on the texture image through a function according to the spatial coordinates (x, y, z) of the three-dimensional model, and use the texture coordinates to extract the corresponding texture values and render them into the three-dimensional model. Therefore, it is necessary to determine the mapping function F so that F(x, y, z)→(u, v).

In the original 3D model, each model space point (xi _{, yi} , _zi ) has a unique corresponding texture coordinate ( _ui , _vi ). On the basis of maintaining the original model texture, according to the spatial range covered by the characteristic contour surface (based on the normal projection direction of the characteristic contour line fitting plane), the points within the range of the characteristic contour surface are orthographically projected, and the corresponding texture coordinates remain unchanged. By projecting the original model vertices onto the characteristic contour surface, the texture coordinates corresponding to the simplified 3D model are obtained, and the texture mapping from the texture image to the simplified 3D model is realized. That is: _Fpro (x, y, z)→(x′, y′, z′), F(x′, y′, z′)→(u, v).

Please refer to Figure 4, which is a schematic diagram of the structure of a three-dimensional model lightweight device provided by an embodiment of the present invention. In this embodiment, the three-dimensional model lightweight device includes:

The tile merging module 210 is used to reconstruct the three-dimensional model according to the tile data of the three-dimensional model to obtain a topologically continuous triangular mesh model;

A plane segmentation module 220 is used to perform surface segmentation according to the plane features of the triangular mesh model to obtain segmented flat surface areas of the overall three-dimensional model;

The block merging module 230 is used to calculate the plane difference of adjacent segmented surfaces according to the block flattened surface domain, and merge the adjacent segmented surfaces whose plane difference is within a preset threshold range to obtain a merged block;

A feature screening module 240 is used to screen geometric feature points in the boundary line nodes according to the boundary lines of the merged blocks, and use the spatial surface composed of the geometric feature points as the lightweight block surface of the three-dimensional model;

The spatial projection module 250 is used to generate a new texture by using a spatial orthographic projection method according to the spatial area covered by the lightweight block surface to obtain a lightweight three-dimensional model.

In one embodiment, the tile merging module 210 is further used to: obtain tile data in a selected area of the three-dimensional model, reconstruct a model of tiles of a specified accuracy level in the tile data using a tile merging method, and generate a topologically continuous triangular mesh model.

In one embodiment, the plane segmentation module 220 is further used to: determine the local fitting plane of the triangular grid model based on the K-means clustering algorithm, classify the model grid vertices, and obtain the block flat surface area after the overall three-dimensional model is segmented.

In a certain embodiment, the block merging module 230 is also used to: calculate the angle between the normal vectors of the triangular grid at the three-dimensional model segmentation according to the curvature at the boundary connection of the three-dimensional model block to obtain the plane difference; merge adjacent segmentation surfaces with smaller plane differences to obtain a merged block.

In a certain embodiment, the feature screening module 240 is also used to: select and identify nodes in the block boundary based on the moving screening method, calculate the spatial fitting straight lines of some nodes of the model boundary line, screen the geometric feature points in the boundary line nodes, and use the spatial surface composed of the geometric feature points as the lightweight block surface of the model.

For the specific definition of the three-dimensional model lightweight device, please refer to the definition of the three-dimensional model lightweight method mentioned above, which will not be repeated here. Each module in the above-mentioned three-dimensional model lightweight device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

Referring to Fig. 5, an embodiment of the present invention provides a computer terminal device, comprising one or more processors and a memory. The memory is coupled to the processor and is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the three-dimensional model lightweight method in any of the above embodiments.

The processor is used to control the overall operation of the computer terminal device to complete all or part of the steps of the above-mentioned three-dimensional model lightweight method. The memory is used to store various types of data to support the operation of the computer terminal device, and these data may include, for example, instructions for any application or method used to operate on the computer terminal device, and data related to the application. The memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk.

In an exemplary embodiment, the computer terminal device can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned three-dimensional model lightweight method and achieve the same technical effect as the above-mentioned method.

In another exemplary embodiment, a computer-readable storage medium including a computer program is also provided, and when the computer program is executed by a processor, the steps of the three-dimensional model lightweight method in any of the above embodiments are implemented. For example, the computer-readable storage medium can be the above-mentioned memory including the computer program, and the above-mentioned computer program can be executed by a processor of a computer terminal device to complete the above-mentioned three-dimensional model lightweight method and achieve the same technical effect as the above-mentioned method.

The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims (9)

1. A method for lightening a three-dimensional model, comprising:

reconstructing the three-dimensional model according to the tile data of the three-dimensional model to obtain a topological continuous triangular grid model;

performing surface segmentation according to the plane characteristics of the triangular mesh model to obtain segmented flat surface areas of the whole three-dimensional model after segmentation;

performing plane difference calculation on the adjacent division surfaces according to the block flattening surface areas, and combining the adjacent division surfaces with the plane difference within a preset threshold range to obtain a combined block;

and screening geometrical feature points in boundary line nodes according to boundary lines of the combined blocks, and taking a space surface formed by the geometrical feature points as a lightweight block surface of a three-dimensional model, wherein the method comprises the following steps of: selecting and judging nodes in the block boundary based on a mobile screening method, calculating a space fitting straight line of part of nodes of the boundary line of the model, screening geometric feature points in the boundary line nodes, and taking a space plane formed by the geometric feature points as a lightweight block plane of the model;

and generating new textures by adopting a space orthographic projection method according to the space region range covered by the light-weight blocking surface to obtain a light-weight three-dimensional model.

2. The method for lightening a three-dimensional model according to claim 1, wherein reconstructing the three-dimensional model according to tile data of the three-dimensional model to obtain a topologically continuous triangular mesh model comprises:

and acquiring tile data in a selected area of the three-dimensional model, and performing model reconstruction on tiles with specified precision levels in the tile data by using a tile merging method to generate a triangular grid model with continuous topology.

3. The method for lightening a three-dimensional model according to claim 1, wherein the performing surface segmentation according to the planar features of the triangular mesh model to obtain segmented flat surface areas after the whole three-dimensional model segmentation comprises:

and determining a local fitting plane of the triangular mesh model based on a K-means clustering algorithm, and classifying the vertices of the triangular mesh model to obtain a segmented flat area after the whole three-dimensional model is segmented.

4. The method for lightening a three-dimensional model according to claim 1, wherein the calculating the plane difference of the adjacent divided surfaces according to the block flattening area, and combining the adjacent divided surfaces with the plane difference within a preset threshold range to obtain a combined block comprises:

calculating the included angle between the triangular grid normal vectors of the three-dimensional model segmentation part according to the curvature of the boundary connection part of the three-dimensional model segmentation part to obtain the plane difference degree;

and merging the adjacent dividing surfaces with smaller plane difference degree to obtain a merged block.

5. A three-dimensional model lightweight device, comprising:

the tile merging module is used for reconstructing the three-dimensional model according to tile data of the three-dimensional model to obtain a triangular grid model with continuous topology;

the plane segmentation module is used for carrying out plane segmentation according to the plane characteristics of the triangular mesh model to obtain a segmented flat area after the whole three-dimensional model is segmented;

the block merging module is used for carrying out plane difference degree calculation on the adjacent division surfaces according to the block flattening surface areas, merging the adjacent division surfaces with the plane difference degree within a preset threshold range, and obtaining a merged block;

the feature screening module is used for screening geometric feature points in boundary line nodes according to the boundary lines of the combined blocks, and taking a space surface formed by the geometric feature points as a lightweight block surface of the three-dimensional model;

and the space projection module is used for generating new textures by adopting a space orthographic projection method according to the space region range covered by the light-weight block surface to obtain a light-weight three-dimensional model.

6. The three-dimensional model lightweight device of claim 5, wherein the tile merge module is further configured to:

and acquiring tile data in a selected area of the three-dimensional model, and performing model reconstruction on tiles with specified precision levels in the tile data by using a tile merging method to generate a triangular grid model with continuous topology.

7. The three-dimensional model lightweight device as defined in claim 5, wherein the planar segmentation module is further configured to:

and determining a local fitting plane of the triangular mesh model based on a K-means clustering algorithm, and classifying the vertices of the triangular mesh model to obtain a segmented flat area after the whole three-dimensional model is segmented.

8. The three-dimensional model lightweight device according to claim 5, wherein the block merging module is further configured to:

calculating the included angle between the triangular grid normal vectors of the three-dimensional model segmentation part according to the curvature of the boundary connection part of the three-dimensional model segmentation part to obtain the plane difference degree;

and merging the adjacent dividing surfaces with smaller plane difference degree to obtain a merged block.

9. The three-dimensional model lightweight device of claim 5, wherein the feature screening module is further configured to:

and selecting and judging nodes in the block boundary based on a mobile screening method, calculating a space fitting straight line of part of nodes of the boundary line of the model, screening geometrical characteristic points in the boundary line nodes, and taking a space plane formed by the geometrical characteristic points as a lightweight block plane of the model.

Images