CN103049932A - Radial basis function-based plant three-dimensional configuration virtual modeling method - Google Patents

Abstract

The invention provides a radial basis function (RBF)-based plant three-dimensional configuration virtual modeling method, which comprises the following steps of: scanning plant image data by using three-dimensional laser; storing the data in a form of a triangular mesh; and realizing deformation of scanned data by using the radial basis function (RBF) by taking a framework topological structure and an organ topological structure of a specific plant as deformation restraints, thereby realizing construction of a target plant model. The sense of reality of details for plant organ three-dimensional configuration virtual modeling is greatly improved, and meanwhile, the method is easy and convenient and has higher efficiency.

Description

A kind of plant three-dimensional shape virtual modeling method based on radial basis function

Technical field

The present invention relates to the three-dimensional picture treatment technology, especially a kind of plant three-dimensional shape virtual modeling method based on radial basis function.

Background technology

The plant three-dimensional shape realistic modeling is the major issue that faces in computer graphics, virtual reality field, also have broad application prospects in the Agricultural Information technical field, plant three-dimensional shape virtual architecture model so that digitizing, visual mode make up is applied at typical patterns such as landscape design, animation making, virtual agricultural, scene display.Along with developing rapidly of computer hardware technique, plant three-dimensional modeling method all is greatly improved at aspects such as the modeling sense of reality, modeling accuracy and modeling efficiencies, and the theoretical method in the fields such as computer graphics, virtual reality provides strong support for the virtual plant modeling technique.

Fast development along with modern information technologies, there has been further lifting in each field to the sense of reality of plant modeling and the requirement of modeling efficiency, for effective performance problem of the minutia of complicated plant organ, be one of the problem demanding prompt solution in current virtual plant modeling field particularly.In recent years, developing rapidly of three-dimensional laser scanning technique, 3 D laser scanning equipment is progressively popularized, introduce three-dimensional laser scanning technique, realize the three-dimensional modeling of plant minutia, to promoting the sense of reality of plant dummy model, be of great immediate significance, in the deeper application in modern information technologies field and Agricultural Information field huge facilitation is arranged for promoting virtual plant.

Plant organ three-dimensional configuration textural anomaly is complicated, and the plant organ minutia is difficult to describe with effective parametric method and mathematic(al) representation, and traditional modeling method and technique is not enough to accurately show the minutia of plant organ form.Three-dimensional laser scanning technique can effectively catch minutia and the surface three dimension Morphologic Characteristics of plant organ, therefore making up the plant three dimensional virtual models based on three-dimensional laser scanner is effective solution, and because laser scanning operation is comparatively complicated, data volume is also larger, therefore be faced with the problem on deformation of scan-data based on the plant modeling of laser scanning data, thereby need according to the typical diversified plant three-dimensional shape model of plant scan-data model construction.

Summary of the invention

The invention provides a kind of plant three-dimensional shape virtual modeling method based on radial basis function, be used for solving prior art, during to the plant three-dimensional modeling, be not enough to accurately show the minutia of plant organ form, and complicated operation, inefficient problem.

A kind of plant three-dimensional shape virtual modeling method based on radial basis function provided by the invention comprises:

Step S1, the plant of using three-dimensional laser scanner to treat modeling is scanned, with the form storage of scan-data with triangle gridding;

Step S2, according to the phytomorph architectural feature for the treatment of modeling, under the local coordinate system of each organ of the plant for the treatment of modeling, determine the target signature point of the morphosis of described plant, and according to the topological relation between each target signature point with apart from the two-dimensional development planimetric coordinates of determining target signature point;

Step S3, take the triangle grid model of scan-data as modelling data source, on the triangle grid model of the scan-data of correspondence, the Euclidean distance of the two-dimensional development planimetric coordinates by described target signature point and the two-dimensional development planimetric coordinates on triangle gridding summit calculates corresponding grid unique point coordinate;

Step S4, use radial basis function to set up grid unique point among the described step S3 to the mapping relations of target signature point, and the distortion side-play amount of the coordinate of having a few on the calculating triangle grid model, then try to achieve the coordinate of having a few on the triangle grid model according to described distortion side-play amount, realize at last the distortion of grid model.

Further, plant three-dimensional shape virtual modeling method of the present invention, among the described step S1, scan-data comprises: summit two-dimensional development planimetric coordinates and the triangle gridding vertex index of the apex coordinate of triangle gridding, the vertex normal of triangle gridding, triangle gridding.

Further, plant three-dimensional shape virtual modeling method of the present invention, the preparation method of the two-dimensional development planimetric coordinates on the triangle gridding summit of described step S3 is:

Step S301 asks neighborhood point weight;

Step S302 determines frontier point two-dimensional development planimetric coordinates;

Step S303 tries to achieve the new coordinate on triangle gridding summit according to neighborhood point weight and frontier point two-dimensional development planimetric coordinates;

Step S304 behind the iterative, draws the two-dimensional development planimetric coordinates on triangle gridding summit.

Further, plant three-dimensional shape virtual modeling method of the present invention, among the described step S3, the computing method of described grid unique point coordinate are all triangle gridding summits of cycle calculations, with the two-dimensional development planimetric coordinates on the two-dimensional development planimetric coordinates of the target signature point chosen and all triangle gridding summit relatively, the triangle gridding summit of Euclidean distance minimum as the grid unique point corresponding with this target signature point, and then the coordinate of definite grid unique point.

Further, plant three-dimensional shape virtual modeling method of the present invention, described step S4 also comprises:

Step S401 calculates the distortion side-play amount of trying to achieve grid unique point coordinate;

Step S402 calculates the distortion side-play amount of trying to achieve all point coordinate on the triangle grid model;

Step S403 according to the distortion side-play amount of all point coordinate on the triangle grid model, tries to achieve the coordinate of having a few on the triangle grid model;

Step S404 finishes the plant three-dimensional modeling according to all point coordinate on the triangle grid model.

Further, plant three-dimensional shape virtual modeling method of the present invention, among the described step S401, the computing method of the distortion side-play amount of grid unique point coordinate are:

Objective definition unique point sum is counted N, and corresponding grid unique point sum is all N, and the objective definition unique point is counted

The grid unique point is

Be the distortion side-play amount of grid unique point, according to formula:

${\stackrel{\‾}{P}}_i=P_i^A-P_i^M$ (i＝1,2,…,N)

Try to achieve the distortion side-play amount ${\stackrel{\‾}{P}}_i(i=\mathrm{1,2},...,N);$ Wherein $P_i^A(i=\mathrm{1,2},...,N)$ With $P_i^M(i=\mathrm{1,2},...,N)$ Be known quantity, try to achieve according to step S2 and step S3.

Further, plant three-dimensional shape virtual modeling method of the present invention, among the described step S402, the computing method of the distortion side-play amount of all point coordinate are on the triangle grid model:

Use the Interpolation Property of Radial Basis Function method to calculate the distortion side-play amount of all point coordinate on the triangle grid model, Interpolation-Radix-Function is selected many quadratic functions (Multi-Quadric), utilizes the Interpolation Property of Radial Basis Function expression formula:

Can calculate the distortion side-play amount F (P) of triangle gridding summit P (x, y, z) coordinate;

Wherein,

F (P) is for putting the distortion side-play amount of P (x, y, z) coordinate on the triangle grid model;

$r_i=R(P,P_i^M)=R(x,y,z;x_i^M,y_i^M,z_i^M)=\sqrt{{(x-x_i^M)}^2+{(y-y_i^M)}^2+{(z-z_i^M)}^2};$

Be the coordinate of network characterization point, S3 draws according to step;

ω _iFor the grid unique point is the network characterization point Weight factor, be a tri-vector Draw according to following formula:

$\left[\begin{array}{cccccc}{a}_{11}& a_{12}& & a_{1j}& & a_{1N}\\ a_{21}& a_{22}& & ...& & \\ & & ...& & & \\ a_{i1}& ...& & a_{\mathrm{ij}}& & \\ & & & & ...& \\ a_{N1}& & & & & a_{\mathrm{NN}}\end{array}\right]\·\left[\begin{array}{ccc}{\mathrm{\ω}}_1^x& {\mathrm{\ω}}_1^y& {\mathrm{\ω}}_1^z\\ {\mathrm{\ω}}_2^x& {\mathrm{\ω}}_2^y& {\mathrm{\ω}}_2^z\\ ...& ...& ...\\ {\mathrm{\ω}}_i^x& {\mathrm{\ω}}_i^y& {\mathrm{\ω}}_i^z\\ ...& ...& ...\\ {\mathrm{\ω}}_N^x& {\mathrm{\ω}}_N^y& {\mathrm{\ω}}_N^z\end{array}\right]=\left[\begin{array}{ccc}\stackrel{\‾}{P_i^x}& \stackrel{\‾}{P_i^y}& \stackrel{\‾}{P_i^z}\\ \stackrel{\‾}{P_2^x}& \stackrel{\‾}{P_2^y}& \stackrel{\‾}{P_2^z}\\ ...& ...& ...\\ \stackrel{\‾}{P_i^x}& \stackrel{\‾}{P_i^y}& \stackrel{\‾}{P_i^z}\\ ...& ...& ...\\ \stackrel{\‾}{P_N^x}& \stackrel{\‾}{P_N^y}& \stackrel{\‾}{P_N^z}\end{array}\right]=\left[\begin{array}{c}\stackrel{\‾}{P_1}\\ \stackrel{\‾}{P_2}\\ ...\\ \stackrel{\‾}{P_i}\\ ...\\ \stackrel{\‾}{P_N}\end{array}\right]$

a _IjBe coefficient, i, j are the coefficient footnote, i=1,2 ... N, j=1,2 ... N, matrix of coefficients are n rank square formations;

$a_{\mathrm{ij}}=R(P_i^M,P_j^M)=R(x_i^M,y_i^M,z_i^M;x_j^M,y_j^M,z_j^M)=\sqrt{{(x_i^M-x_j^M)}^2+{(y_i^M-y_j^M)}^2+{(z_i^M-z_j^M)}^2+1};$

Be the distortion side-play amount of grid unique point, S401 draws according to step.

Further, plant three-dimensional shape virtual modeling method of the present invention is characterized in that, among the described step S1, the saving format of scan-data is " .OBJ ".

Further, plant three-dimensional shape virtual modeling method of the present invention, described plant comprises corn, cucumber, tobacco, paddy rice, wheat.

Further, plant three-dimensional shape virtual modeling method of the present invention, the organ of described plant comprise blade, stipes, fruit, female fringe, tassel.

The invention provides a kind of plant three-dimensional shape virtual modeling method based on radial basis function, utilize laser scanning data, data are with the form storage of triangle gridding, take the skeleton of specified plant and organ topological structure as deformation constrain, use radial basis interpolating function (RBF), realize the distortion of scan-data, thereby the structure of realize target plant model, significantly improved the details sense of reality of the virtual modeling of plant organ three-dimensional configuration, while method is easy, has higher efficient.

Description of drawings

Fig. 1 is the process flow diagram of the described plant three-dimensional shape virtual modeling method of the embodiment of the invention;

Fig. 2 is the described maize leaf scan-data of embodiment of the invention grid model figure;

Fig. 3 is the described target signature point of the embodiment of the invention and grid unique point synoptic diagram;

Fig. 4 is that the described blade of the embodiment of the invention is based on the deformation effect figure of Interpolation Property of Radial Basis Function method;

Fig. 5 is the described corn sense of reality illustraton of model that makes up based on scan-data of the embodiment of the invention.

Embodiment

In order to understand better the present invention, the invention will be further described below in conjunction with accompanying drawing and embodiment.

The described a kind of plant three-dimensional shape virtual modeling method based on radial basis function of the embodiment of the invention take maize leaf as example, specifically describes how to use method of the present invention for this plant organ of maize leaf.Except blade, can also use modeling method of the present invention to plant organs such as stipes, fruit, female fringe, tassels.Need to choose the unique point of plant organ during application, so that modeling.The unique point of stipes, fruit, female fringe, tassel is chosen on this organ axis, and evenly distributes.The modeling method of the plant organs such as stipes, fruit, female fringe, tassel is identical with the modeling method of blade, therefore repeats no more.

Fig. 1 is the process flow diagram of the described plant three-dimensional shape virtual modeling method of the embodiment of the invention, and as shown in Figure 1, the described plant three-dimensional shape virtual modeling method of the embodiment of the invention comprises:

Step S1, the plant of using three-dimensional laser scanner to treat modeling is scanned, with the form storage of scan-data with triangle gridding;

Maize leaf is carried out 3 D laser scanning, Fig. 2 is the described maize leaf scan-data of embodiment of the invention grid model figure, as shown in Figure 2, the data of three-dimensional laser scanner are stored with the form of triangle gridding (Mesh), data message comprises triangle gridding apex coordinate, vertex normal, summit two-dimensional development planimetric coordinates, triangle gridding vertex index; Wherein summit two-dimensional development planimetric coordinates XY scope is [0,1];

The storage organization of triangle gridding is the grid model of OBJ form;

Step S2, according to the phytomorph architectural feature for the treatment of modeling, under the local coordinate system of each organ of the plant for the treatment of modeling, determine the target signature point of the morphosis of described plant, and according to the topological relation between each target signature point with apart from the two-dimensional development planimetric coordinates of determining target signature point;

Treat the plant organ of modeling for each, the sequence that target signature point forms can effectively be described subject form and the topological structure of this organ.

Take maize leaf as example, maize leaf is the strip structure, choose base portion point, Ye Changsan/a bit, blade peak at corn vein curve, four points of blade tip point are the target signature point, because the target signature point of each selected blade is positioned at the vein curve, namely on the blade center line, therefore the X value of the two-dimensional development planimetric coordinates of each point is 0.5, and Y value is tried to achieve in proportion by the distance between the each point, and maximal value is 1.0; The vein curve of each blade determines that by the milpa plant type structure for the treatment of modeling plant type structure is by internode length, internode radius, leaf length, leaf inclination angle, leaf position angle, the high description of leaf;

Step S3, take the triangle grid model of scan-data as modelling data source, on the triangle grid model of the scan-data of correspondence, the Euclidean distance of the two-dimensional development planimetric coordinates by described target signature point and the two-dimensional development planimetric coordinates on triangle gridding summit calculates corresponding grid unique point coordinate;

The two-dimensional development planimetric coordinates on triangle gridding summit obtains in the following way:

Step S301 asks neighborhood point weight; Summit P for triangle gridding _i(x, y, z), this summit neighborhood number of vertices is n, is respectively P _Ij(j=1,2 ..., n), n is natural number, neighborhood vertex weights ω _jFor: ω _j=1/n;

Step S302 determines frontier point two-dimensional development planimetric coordinates; Determine border vertices two-dimensional development planimetric coordinates according to the organ concrete form, at first choose four angle points, coordinate is respectively [0,0], [1,0], [1,1], [0,1], choosing of four angle points is relevant with the organ concrete form, take maize leaf as example, the coordinate of choosing the blade base both sides of the edge is respectively [0,0], [1,0], the coordinate of point of blade both sides of the edge is respectively [0,1], [1,1], then, the coordinate of other frontier points in the order of connection linear interpolation is tried to achieve, the coordinate of border vertices is distributed in [0 in the order of connection, 0] to the square coordinates regional of [1,1], the coordinate assignment of frontier point is given the X value of the border vertices of corresponding triangle gridding, the Y value component, the component of Z value is 0, can try to achieve frontier point two-dimensional development planimetric coordinates;

Step S303 tries to achieve the new coordinate on triangle gridding summit according to neighborhood point weight and frontier point two-dimensional development planimetric coordinates; According to the two-dimensional development planimetric coordinates numerical value of border vertices, the new coordinate P in summit is tried to achieve on all summits of circulating _i':

Step S304 behind the iterative, draws the two-dimensional development planimetric coordinates on triangle gridding summit; Repeating step S303 is until the maximum offset in all summits

Less than 0.01, X value and Y value take the X value on final triangle gridding summit, Y value component as the two-dimensional development planimetric coordinates on triangle gridding summit, Z value component is 0.

Obtain after the two-dimensional development planimetric coordinates on triangle gridding summit, take the target signature point as reference, the two-dimensional development planimetric coordinates on the two-dimensional development planimetric coordinates of the target signature point that obtains by step S2 and the triangle gridding summit of acquisition calculates the grid unique point of correspondence, computing method are all grid vertexes of cycle calculations, choose the two-dimensional development planimetric coordinates grid unique point corresponding with this target signature point with the triangle gridding summit conduct of certain target signature point two-dimensional development planimetric coordinates Euclidean distance minimum;

Take maize leaf as example, four selected points are that target signature point is as the vein curve, calculating is tried to achieve and is scanned the triangle gridding summit that four corresponding on grid model points are respectively two-dimensional development planimetric coordinates and certain target signature point two-dimensional development planimetric coordinates Euclidean distance minimum, as four the grid unique points corresponding with this target signature point; Fig. 3 is the described target signature point of the embodiment of the invention and grid unique point synoptic diagram, and target signature point and grid unique point are as shown in Figure 3.

Step S4, use radial basis function to set up grid unique point among the described step S3 to the mapping relations of target signature point, and the distortion side-play amount of the coordinate of having a few on the calculating triangle grid model, then try to achieve the coordinate of having a few on the triangle grid model according to described distortion side-play amount, realize at last the distortion of grid model.

Deformation process is take the target signature point as constraint, and the scanning grid model after the distortion and target plant model structure are coincide;

Wherein, step S4 specifically comprises:

Step S401 calculates the distortion side-play amount of trying to achieve grid unique point coordinate;

The computing method of the distortion side-play amount of grid unique point coordinate are:

Objective definition unique point sum is counted N, and corresponding grid unique point sum is all N, and the objective definition unique point is counted

The grid unique point is

Be the distortion side-play amount of grid unique point, according to formula:

${\stackrel{\‾}{P}}_i=P_i^A-P_i^M$ (i＝1,2,…,N)

Try to achieve the distortion side-play amount ${\stackrel{\‾}{P}}_i(i=\mathrm{1,2},...,N);$ Wherein $P_i^A(i=\mathrm{1,2},...,N)$ With $P_i^M(i=\mathrm{1,2},...,N)$ Be known quantity, try to achieve according to step S2 and step S3.The distortion side-play amount is different from Euclidean distance, is to contain directive vector.

Step S402 calculates the distortion side-play amount of trying to achieve all point coordinate on the triangle grid model;

The computing method of the distortion side-play amount of all point coordinate are on the triangle grid model:

Use radial basis function (RBF, Radial basis function) interpolation method calculates the distortion side-play amount of all point coordinate on the triangle grid model, Interpolation-Radix-Function is selected many quadratic functions (Multi-Quadric), utilizes the Interpolation Property of Radial Basis Function expression formula:

Can calculate the distortion side-play amount F (P) of triangle gridding summit P (x, y, z) coordinate;

Wherein,

F (P) is for putting the distortion side-play amount of P (x, y, z) coordinate on the triangle grid model;

$r_i=R(P,P_i^M)=R(x,y,z;x_i^M,y_i^M,z_i^M)=\sqrt{{(x-x_i^M)}^2+{(y-y_i^M)}^2+{(z-z_i^M)}^2};$

Be the coordinate of network characterization point, S3 draws according to step;

ω _iFor the grid unique point is the network characterization point Weight factor, be a tri-vector

Draw according to following formula:

$\left[\begin{array}{cccccc}{a}_{11}& a_{12}& & a_{1j}& & a_{1N}\\ a_{21}& a_{22}& & ...& & \\ & & ...& & & \\ a_{i1}& ...& & a_{\mathrm{ij}}& & \\ & & & & ...& \\ a_{N1}& & & & & a_{\mathrm{NN}}\end{array}\right]\·\left[\begin{array}{ccc}{\mathrm{\ω}}_1^x& {\mathrm{\ω}}_1^y& {\mathrm{\ω}}_1^z\\ {\mathrm{\ω}}_2^x& {\mathrm{\ω}}_2^y& {\mathrm{\ω}}_2^z\\ ...& ...& ...\\ {\mathrm{\ω}}_i^x& {\mathrm{\ω}}_i^y& {\mathrm{\ω}}_i^z\\ ...& ...& ...\\ {\mathrm{\ω}}_N^x& {\mathrm{\ω}}_N^y& {\mathrm{\ω}}_N^z\end{array}\right]=\left[\begin{array}{ccc}\stackrel{\‾}{P_i^x}& \stackrel{\‾}{P_i^y}& \stackrel{\‾}{P_i^z}\\ \stackrel{\‾}{P_2^x}& \stackrel{\‾}{P_2^y}& \stackrel{\‾}{P_2^z}\\ ...& ...& ...\\ \stackrel{\‾}{P_i^x}& \stackrel{\‾}{P_i^y}& \stackrel{\‾}{P_i^z}\\ ...& ...& ...\\ \stackrel{\‾}{P_N^x}& \stackrel{\‾}{P_N^y}& \stackrel{\‾}{P_N^z}\end{array}\right]=\left[\begin{array}{c}\stackrel{\‾}{P_1}\\ \stackrel{\‾}{P_2}\\ ...\\ \stackrel{\‾}{P_i}\\ ...\\ \stackrel{\‾}{P_N}\end{array}\right]$

a _IjBe coefficient, i, j are the coefficient footnote, i=1,2 ... N, j=1,2 ... N, matrix of coefficients are n rank square formations;

$a_{\mathrm{ij}}=R(P_i^M,P_j^M)=R(x_i^M,y_i^M,z_i^M;x_j^M,y_j^M,z_j^M)=\sqrt{{(x_i^M-x_j^M)}^2+{(y_i^M-y_j^M)}^2+{(z_i^M-z_j^M)}^2+1};$

Be the distortion side-play amount of grid unique point, S401 draws according to step.

Take maize leaf as example, choose vein curvilinear characteristic point, N=4, the target signature point of definition is counted

The grid unique point is

Be the distortion side-play amount on feature summit, utilize the target signature point

With the grid unique point The distortion side-play amount on feature summit is tried to achieve in calculating

The Interpolation Property of Radial Basis Function basis function can also be linear interpolation function or thin plate spline function.

Step S403 according to the distortion side-play amount of all point coordinate on the triangle grid model, tries to achieve the coordinate of having a few on the triangle grid model;

Step S404 finishes the plant three-dimensional modeling according to all point coordinate on the triangle grid model.

All organs of plant model for the treatment of modeling are processed by the method for above-mentioned steps, realize the three-dimensional reconstruction based on scan model of all organs, realize on the whole the plant three-dimensional shape virtual modeling method based on scan-data, the plant model that makes up has kept the minutia of scan-data, and has higher counting yield.

Be modeled as example with corn, scan-data based on each typical organ, make up whole strain corn three-dimensional configuration model according to plant type structure, realize the virtual modeling of milpa three-dimensional configuration based on scan-data, effectively kept the minutia information such as fold of maize leaf.Fig. 4 be the described blade of the embodiment of the invention based on the deformation effect figure of Interpolation Property of Radial Basis Function method, the implementation result of maize leaf is as shown in Figure 4.Fig. 5 is the described corn sense of reality illustraton of model that makes up based on scan-data of the embodiment of the invention, and the implementation result of corn as shown in Figure 5.

The method of the invention is carried out three-dimensional reconstruction mainly for blocking less crops, plant such as corn, cucumber, tobacco, paddy rice, wheat, and the organ of plant comprises blade, stipes, fruit, female fringe, tassel etc.

Below only be the preferred embodiments of the present invention; certainly; the present invention can also have other various embodiments; in the situation that does not deviate from spirit of the present invention and essence thereof; those of ordinary skill in the art work as can make according to the present invention various corresponding changes and distortion, but these corresponding changes and distortion all should belong to the protection domain of the appended claim of the present invention.

Claims (10)

1. the plant three-dimensional shape virtual modeling method based on radial basis function is characterized in that, comprising:

Step S1, the plant of using three-dimensional laser scanner to treat modeling is scanned, with the form storage of scan-data with triangle gridding;

Step S2, according to the phytomorph architectural feature for the treatment of modeling, under the local coordinate system of each organ of the plant for the treatment of modeling, determine the target signature point of the morphosis of described plant, and according to the topological relation between each target signature point with apart from the two-dimensional development planimetric coordinates of determining target signature point;

Step S3, take the triangle grid model of scan-data as modelling data source, on the triangle grid model of the scan-data of correspondence, the Euclidean distance of the two-dimensional development planimetric coordinates by described target signature point and the two-dimensional development planimetric coordinates on triangle gridding summit calculates corresponding grid unique point coordinate;

Step S4, use radial basis function to set up grid unique point among the described step S3 to the mapping relations of target signature point, and the distortion side-play amount of the coordinate of having a few on the calculating triangle grid model, then try to achieve the coordinate of having a few on the triangle grid model according to described distortion side-play amount, realize at last the distortion of grid model.

2. plant three-dimensional shape virtual modeling method according to claim 1, it is characterized in that, among the described step S1, scan-data comprises: summit two-dimensional development planimetric coordinates and the triangle gridding vertex index of the apex coordinate of triangle gridding, the vertex normal of triangle gridding, triangle gridding.

3. plant three-dimensional shape virtual modeling method according to claim 2 is characterized in that, the preparation method of the two-dimensional development planimetric coordinates on the triangle gridding summit of described step S3 is:

Step S301 asks neighborhood point weight;

Step S302 determines frontier point two-dimensional development planimetric coordinates;

Step S303 tries to achieve the new coordinate on triangle gridding summit according to neighborhood point weight and frontier point two-dimensional development planimetric coordinates;

Step S304 behind the iterative, draws the two-dimensional development planimetric coordinates on triangle gridding summit.

4. plant three-dimensional shape virtual modeling method according to claim 3, it is characterized in that, among the described step S3, the computing method of described grid unique point coordinate are all triangle gridding summits of cycle calculations, with the two-dimensional development planimetric coordinates on the two-dimensional development planimetric coordinates of the target signature point chosen and all triangle gridding summit relatively, the triangle gridding summit of Euclidean distance minimum as the grid unique point corresponding with this target signature point, and then the coordinate of definite grid unique point.

5. plant three-dimensional shape virtual modeling method according to claim 4 is characterized in that, described step S4 also comprises:

Step S401 calculates the distortion side-play amount of trying to achieve grid unique point coordinate;

Step S402 calculates the distortion side-play amount of trying to achieve all point coordinate on the triangle grid model;

Step S403 according to the distortion side-play amount of all point coordinate on the triangle grid model, tries to achieve the coordinate of having a few on the triangle grid model;

Step S404 finishes the plant three-dimensional modeling according to all point coordinate on the triangle grid model.

6. plant three-dimensional shape virtual modeling method according to claim 5 is characterized in that, among the described step S401, the computing method of the distortion side-play amount of grid unique point coordinate are:

Objective definition unique point sum is counted N, and corresponding grid unique point sum is all N, and the objective definition unique point is counted

The grid unique point is

Be the distortion side-play amount of grid unique point, according to formula:

${\stackrel{\‾}{P}}_i=P_i^A-P_i^M$ (i＝1,2,…,N)

Try to achieve the distortion side-play amount ${\stackrel{\‾}{P}}_i(i=\mathrm{1,2},...,N);$ Wherein $P_i^A(i=\mathrm{1,2},...,N)$ With $P_i^M(i=\mathrm{1,2},...,N)$ Be known quantity, try to achieve according to step S2 and step S3.

7. plant three-dimensional shape virtual modeling method according to claim 6 is characterized in that, among the described step S402, the computing method of the distortion side-play amount of all point coordinate are on the triangle grid model:

Use the Interpolation Property of Radial Basis Function method to calculate the distortion side-play amount of all point coordinate on the triangle grid model, Interpolation-Radix-Function is selected many quadratic functions, utilizes the Interpolation Property of Radial Basis Function expression formula:

Can calculate the distortion side-play amount F (P) of triangle gridding summit P (x, y, z) coordinate;

Wherein,

F (P) is for putting the distortion side-play amount of P (x, y, z) coordinate on the triangle grid model;

$r_i=R(P,P_i^M)=R(x,y,z;x_i^M,y_i^M,z_i^M)=\sqrt{{(x-x_i^M)}^2+{(y-y_i^M)}^2+{(z-z_i^M)}^2};$

Be the coordinate of network characterization point, S3 draws according to step;

ω i is that the grid unique point is the network characterization point

Weight factor, be a tri-vector

Draw according to following formula:

$\left[\begin{array}{cccccc}{a}_{11}& a_{12}& & a_{1j}& & a_{1N}\\ a_{21}& a_{22}& & ...& & \\ & & ...& & & \\ a_{i1}& ...& & a_{\mathrm{ij}}& & \\ & & & & ...& \\ a_{N1}& & & & & a_{\mathrm{NN}}\end{array}\right]\·\left[\begin{array}{ccc}{\mathrm{\ω}}_1^x& {\mathrm{\ω}}_1^y& {\mathrm{\ω}}_1^z\\ {\mathrm{\ω}}_2^x& {\mathrm{\ω}}_2^y& {\mathrm{\ω}}_2^z\\ ...& ...& ...\\ {\mathrm{\ω}}_i^x& {\mathrm{\ω}}_i^y& {\mathrm{\ω}}_i^z\\ ...& ...& ...\\ {\mathrm{\ω}}_N^x& {\mathrm{\ω}}_N^y& {\mathrm{\ω}}_N^z\end{array}\right]=\left[\begin{array}{ccc}\stackrel{\‾}{P_i^x}& \stackrel{\‾}{P_i^y}& \stackrel{\‾}{P_i^z}\\ \stackrel{\‾}{P_2^x}& \stackrel{\‾}{P_2^y}& \stackrel{\‾}{P_2^z}\\ ...& ...& ...\\ \stackrel{\‾}{P_i^x}& \stackrel{\‾}{P_i^y}& \stackrel{\‾}{P_i^z}\\ ...& ...& ...\\ \stackrel{\‾}{P_N^x}& \stackrel{\‾}{P_N^y}& \stackrel{\‾}{P_N^z}\end{array}\right]=\left[\begin{array}{c}\stackrel{\‾}{P_1}\\ \stackrel{\‾}{P_2}\\ ...\\ \stackrel{\‾}{P_i}\\ ...\\ \stackrel{\‾}{P_N}\end{array}\right]$

a _IjBe coefficient, i, j are the coefficient footnote, i=1,2 ... N, j=1,2 ... N, matrix of coefficients are n rank square formations;

$a_{\mathrm{ij}}=R(P_i^M,P_j^M)=R(x_i^M,y_i^M,z_i^M;x_j^M,y_j^M,z_j^M)=\sqrt{{(x_i^M-x_j^M)}^2+{(y_i^M-y_j^M)}^2+{(z_i^M-z_j^M)}^2+1};$

Be the distortion side-play amount of grid unique point, S401 draws according to step.

8. according to claim 1 to the described plant three-dimensional shape virtual modeling method of 7 any one, it is characterized in that among the described step S1, the saving format of scan-data is " .OBJ ".

9. plant three-dimensional shape virtual modeling method according to claim 8 is characterized in that, described plant comprises corn, cucumber, tobacco, paddy rice, wheat.

10. plant three-dimensional shape virtual modeling method according to claim 9 is characterized in that, the organ of described plant comprises blade, stipes, fruit, female fringe, tassel.

Images