CN106204740B - Three-dimensional defect face model reconstruction method based on slice data symmetry analysis - Google Patents

Three-dimensional defect face model reconstruction method based on slice data symmetry analysis Download PDF

Info

Publication number
CN106204740B
CN106204740B CN201610596173.3A CN201610596173A CN106204740B CN 106204740 B CN106204740 B CN 106204740B CN 201610596173 A CN201610596173 A CN 201610596173A CN 106204740 B CN106204740 B CN 106204740B
Authority
CN
China
Prior art keywords
dimensional
model
layer
point
axis
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201610596173.3A
Other languages
Chinese (zh)
Other versions
CN106204740A (en
Inventor
孙进
丁静
黄则栋
刘远
陶晔
曾励
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yangzhou University
Original Assignee
Yangzhou University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yangzhou University filed Critical Yangzhou University
Priority to CN201610596173.3A priority Critical patent/CN106204740B/en
Publication of CN106204740A publication Critical patent/CN106204740A/en
Application granted granted Critical
Publication of CN106204740B publication Critical patent/CN106204740B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G06T2207/30201—Face

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Graphics (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Architecture (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Image Processing (AREA)

Abstract

A three-dimensional defected facial model reconstruction method based on slice data symmetry analysis belongs to the field of computer measurement data modeling. The invention comprises the following steps: 1) inputting a three-dimensional face defect model, and slicing the three-dimensional model: after the three-dimensional model is input, equally-spaced slicing and layering are carried out on the three-dimensional model perpendicular to a set axial direction, and the layers are further divided into sectors; 2) carrying out symmetry analysis on the slice model obtained in the step 1) to obtain an optimal symmetry plane; 3) reconstructing a three-dimensional grid of the defect area based on the optimal symmetry plane obtained in the step 2) and then outputting a three-dimensional model. The invention has larger flexibility and expandability and can meet the requirement of higher precision.

Description

Three-dimensional defect face model reconstruction method based on slice data symmetry analysis
Technical Field
The invention relates to a three-dimensional defective facial model reconstruction method, in particular to a three-dimensional defective facial model reconstruction method based on slice data symmetry analysis, and belongs to the field of computer measurement data modeling.
Background
With the development of the geometric modeling technology of the measured data, the requirement of people on the three-dimensional reconstruction of the defect model is higher and higher, especially under the conditions of change of human facial expressions, different illumination environments and larger defect data. However, because the human face has symmetry, for defective data with symmetric characteristics, the data of the intact region can be accurately mirrored to the defective region through the symmetry plane, so that the human face is modeled.
For the symmetry analysis of a three-dimensional model, the following symmetric classifications, global symmetry and local symmetry, and extrinsic symmetry and intrinsic symmetry, are generally given attention to various different types of geometric symmetry in the model. The extrinsic symmetry existing in the three-dimensional Euclidean space comprises reflection, rotation, translation, sliding reflection, rotation reflection and spiral symmetry, and compared with global symmetry detection, the local symmetry detection technology has wider applicability. Currently, there is much research on the detection of symmetry planes with complete data of locally symmetric or approximately symmetric features, and in 2006 Gal and Cohen et al, "clinical geometry for partial shape matching and similarity" (ACMTransections on Graphics (TOG),2006,25(1):130-150.) propose a partial symmetry method capable of detecting self-similarity. The method comprises the steps of realizing local automatic segmentation of a model according to the curvature of the surface of the model, using the salient features as three-dimensional model shape descriptors, using a very large hash table to store the features and quickly eliminating unnecessary searches to gain efficiency, and measuring the overall similarity of the model by using the local similarity.
Podolak et al used the proposed Plane Reflection Symmetry Transformation (PRST) to define the principal symmetry axis and the symmetry center of a three-dimensional model in "A planar-reflective symmetry transform for 3 Dmaps" (InACM Transformations On Graphics (TOG) [ C ], 2006; 549-) -559.) in 2006, to replace the principal axis and the shape center obtained by principal component analysis to complete the alignment operation on the input model, achieving more accurate and stable effects. While this method is less noisy when used to reconstruct voxel (3D pixel) models, they are sensitive to high amounts of noise and are susceptible to impact. And does not provide accurate assurance of reconstruction of the polygonal mesh model. Martinet et al, in the same year, "Accurate detection of symmetry in 3D shapes" (ACM Transactions of Graphics,2006, vol.25, No.1, pp.439-464), proposed a method of automatically finding three-dimensional shape symmetry, using the symmetry of a generalized matrix and a transition condition to check the spherical harmonic coefficients and restore the shape symmetry parameters. For the fusion of large objects, the method firstly obtains local symmetry and then recovers the overall shape by using an incremental algorithm, and the method does not depend on the subdivision of the model surface and has small noise effect.
In 2011 Xin Li, Zhao Yin et al, in "Symmetry and template-adjusted complex formed bones" Computers & Graphics,2011,35(4): 885-893), studied human skull with symmetric features, using symmetric regions on the skull for repairing the defect, they proposed the principle curvature (encoding of the curved surface) and the diameter function of the shape (model of the encoding volume) of the local shape feature to obtain a reliable Symmetry check of the incomplete model, and in case of large defects, they proposed a template-based method, first mapping the template-damaged skull, then transplanting the template to fill the defect region, and then repeating the template-based Symmetry repair using the proposed complete framework.
Therefore, a three-dimensional defective face model reconstruction method based on slice data symmetry analysis is provided.
Disclosure of Invention
The invention aims to provide a three-dimensional defective face model reconstruction method based on slice data symmetry analysis.
The invention is realized by the following technical scheme, and the three-dimensional defected face model reconstruction method based on slice data symmetry analysis is characterized by comprising the following steps:
1) inputting a three-dimensional face defect model, and slicing the three-dimensional model: after the three-dimensional model is input, equally-spaced slicing and layering are carried out on the three-dimensional model perpendicular to a set axial direction, and the layers are further divided into sectors;
2) carrying out symmetry analysis on the slice model obtained in the step 1) to obtain an optimal symmetry plane;
3) reconstructing a three-dimensional grid of the defect area based on the optimal symmetry plane obtained in the step 2) and then outputting a three-dimensional model;
the step 1) is specifically as follows:
1_1), initial setting:
setting an input three-dimensional face model as M, and a slice model obtained by slicing as M1(ii) a Setting the axis as a Z axis; layering the three-dimensional facial model M by equidistant slicing perpendicular to the Z axis to obtain N layerslSetting the height of the layer to be DeltaZSetting the Z-axis direction coordinate of the layer center layer as L; dividing each layer into equal angle sectors by taking the intersection point of the Z axis and each layer as the center, and setting the number of the obtained equal angle sectors of each layer as Ns(ii) a Setting A as a point array formed by three-dimensional points contained in the sector; d is a distance value between the three-dimensional point and the Z axis;
1_2), layering:
setting Z-axis direction coordinates of the bottom layer and the top layer of the layer as Zh＝L+ΔZ(iii) 2 and Zl＝L-ΔZ/2；
Any three-dimensional point coordinate in the model is p ═ x, y, Z, if Z ∈ (Z) of the three-dimensional point pl,Zh) If yes, the point p belongs to the setting layer;
1_3) sectorization:
each sector having an angle of deltaa＝2π/Ns(ii) a For a three-dimensional point p, let
Figure GDA0002011298680000031
α ═ arccos (x/r), if y of three-dimensional point p>0, the sector number k of the three-dimensional point p is [ α/delta ]a]Otherwise k is equal to Ns-1-[α/Δa]Finally, the three-dimensional point p is taken as the kth value A of the A arraykInserting into an A array; definition DkDefining D as the maximum value of the distance between the three-dimensional point and the Z axiskIs the maximum value of D if r>DkThen r is assigned to Dk；
1_4) calculating all points on the triangular mesh, only leaving the outermost points, and obtaining a point array A formed by three-dimensional points contained in each sector, namely a slice model M1。
The step 2) is to perform symmetry analysis on the slice model, and the method for obtaining the optimal symmetry plane comprises the following steps:
2_1) after slicing, determining the symmetry type of the model and adjusting the numerical parameters so that the three-dimensional model is adjusted along the Z-axis to be closer to the actual axis (or through the plane of symmetry);
a. initializing Nmax＝0，rbest(x,y,z)＝null，Nsym0, r for each rotationjJ 1 … … R, i 1 … … N for each layer index il；
b. Combined output L ═ Z + i ΔzTo calculate the arrays A, D, calculate DaAs a stored value for the non-zero distance in D, the relative difference of the average distance is calculated, the relative difference being Max (| D)k-Da|)/Da,k∈(1,2……Ns)；
c. If the relative difference is poor<εDThen let Nsym＝Nsym+1, otherwise, searching the next layer; if N is presentsym>NmaxThen let Nmax＝Nsym，rbest(x,y,z)＝rj(x, y, z), otherwise, performing the next rotation to obtain the rotation coordinate r of the rotation model Mbest(x,y,z)；
rj(x, y, z) is a rotation setting model calibration test, where j 1.. R; epsilonDA threshold value, N, of the difference between the distance of the outermost point to the z-axis and the average distance of the points on the layer from the z-axissThe number of layers for each rotational symmetry;
2_2) detecting the symmetry of the alignment model and collecting information of asymmetric or missing parts in the layer-layer base model, wherein the missing parts are stored in the polyline pairs to be used for generating the three-dimensional grid;
a. for each layer index l, l 1lCombined output L ═ Z + L ΔzTo calculate arrays a 'and D'; an array of points P is created, the elements initialized as follows: if A'kNot equal to null, then Pk＝A'k(ii) a Otherwise Pk＝(0,0,L')；
b. Let Pl＝0，D'max0, when k is 1sIf D'i>D'maxLet D'max＝D'i(ii) a When j is 1sWhen it is used, orderP'l(0,j)Let θ be j × 2 pi/NsIf D'i<γ×D'max(γ is a threshold value for the distance of the two side points of the shaft), then P'l(1,j)＝(D'max×cosθ,D'maxX sin θ, L '), otherwise P'l(1,j)＝(D'j×cosθ,D'jX sin θ, L ') and then gives the array P'l(i,j)；
2_3) execution, the model is sliced, and in each layer, the point of greatest distance to the Z axis is recorded at D'maxOf each point j on the layer and a distance D'jIs re-encoded when the array polyline angle is θ, and the output of the algorithm is P'l(i,j)And the grid is used for the next grid generation.
The method for reconstructing the three-dimensional grid of the defect area based on the optimal symmetry plane and then outputting the three-dimensional model in the step 3) comprises the following steps:
3_1), distinguishing array broken lines by using empty broken lines, and finding out non-intersected defect parts to carry out reconstruction processing;
3_2), connecting the corresponding points of the middle sections of the two adjacent layers corresponding to the two non-empty broken line pairs, and adding the generated triangle into the triangle array on the basis of the connecting points;
3_3), adding a new point into the existing triangle array, and searching all circumscribed circles of the triangle containing the new addition point;
3_4). after the triangle itself containing the new joining point is found, its adjoining triangles and the adjoining triangles of the adjoining triangles are searched for by iterative nature until no more circumcircles enveloping the triangle of the new joining point can be found; drawing each triangle in the triangle array to realize the re-modeling of the three-dimensional mesh of the defect area; and then fitting by adopting a Bezier curved surface on the triangular mesh, and outputting a smooth three-dimensional model after smoothing treatment.
The three-dimensional defective facial model reconstruction method based on slice data symmetry analysis has the advantages of being high in efficiency, high in flexibility and expandability and capable of meeting the requirement of high precision.
Drawings
FIG. 1 is a flow chart of the steps of a three-dimensional defect model reconstruction method;
fig. 2 is a flowchart of the model slicing processing steps.
Detailed Description
The three-dimensional defect face model reconstruction method based on slice data symmetry analysis is combined with the attached drawings. Specific implementations of the present invention are further described.
The invention comprises the following steps:
1) inputting a three-dimensional face defect model, and slicing the three-dimensional model: after the three-dimensional model is input, equally-spaced slicing and layering are carried out on the three-dimensional model perpendicular to a set axial direction, and the layers are further divided into sectors;
1_1. initial setting:
setting an input three-dimensional face model as M, and a slice model obtained by slicing as M1. Setting the axis as a Z axis; layering the three-dimensional facial model M by equidistant slicing perpendicular to the Z axis to obtain N layerslSetting the height of the layer to be DeltaZSetting the Z-axis direction coordinate of the layer center layer as L; dividing each layer into equal angle sectors by taking the intersection point of the Z axis and each layer as the center, and setting the number of the obtained equal angle sectors of each layer as Ns(ii) a Setting A as a point array formed by three-dimensional points contained in the sector; d is the distance value between the three-dimensional point and the Z axis.
1_2. layering:
setting Z-axis direction coordinates of the bottom layer and the top layer of the layer as Zh＝L+ΔZ(iii) 2 and Zl＝L-ΔZ/2. Any three-dimensional point coordinate in the model is p ═ x, y, Z, if Z ∈ (Z) of the three-dimensional point pl,Zh) If yes, the point p belongs to the setting layer;
1_3. sectorization:
each sector having an angle of deltaa＝2π/Ns(ii) a For a three-dimensional point p, let
Figure GDA0002011298680000051
α ═ arccos (x/r), if y of three-dimensional point p>0, the sector number k of the three-dimensional point p＝[α/Δa]Otherwise k is equal to Ns-1-[α/Δa]Finally, the three-dimensional point p is taken as the kth value A of the A arraykInserting into an A array; definition DkDefining D as the maximum value of the distance between the three-dimensional point and the Z axiskIs the maximum value of D if r>DkThen r is assigned to Dk；
1_4, calculating all points on the triangular mesh, only leaving the outermost points, and obtaining a point array A formed by three-dimensional points contained in each sector, namely a slice model M1。
2) Analyzing the symmetry of the slice model to obtain an optimal symmetry plane;
2_1) after slicing, determining the symmetry type of the model and adjusting the numerical parameters so that the three-dimensional model is adjusted along the Z-axis to be closer to the actual axis (or through the plane of symmetry);
a. initializing Nmax＝0，rbest(x,y,z)＝null，Nsym0, r for each rotationjJ 1 … … R, i 1 … … N for each layer index il；
b. Combined output L ═ Z + i ΔzTo calculate the arrays A, D, calculate DaAs a stored value for the non-zero distance in D, the relative difference of the average distance is calculated, the relative difference being Max (| D)k-Da|)/Da,k∈(1,2……Ns)；
c. If the relative difference is poor<εDThen let Nsym＝Nsym+1, otherwise, searching the next layer; if N is presentsym>NmaxThen let Nmax＝Nsym，rbest(x,y,z)＝rj(x, y, z), otherwise, performing the next rotation to obtain the rotation coordinate r of the rotation model Mbest(x,y,z)；
rj(x, y, z) is a rotation setting model calibration test, where j 1.. R; epsilonDA threshold value, N, of the difference between the distance of the outermost point to the z-axis and the average distance of the points on the layer from the z-axissThe number of layers for each rotational symmetry;
2_2) detecting the symmetry of the alignment model and collecting information of asymmetric or missing parts in the layer-layer base model, wherein the missing parts are stored in the polyline pairs to be used for generating the three-dimensional grid;
a. for each layer index l, l 1lCombined output L ═ Z + L ΔzTo calculate arrays a 'and D'; an array of points P is created, the elements initialized as follows: if A'kNot equal to null, then Pk＝A'k(ii) a Otherwise Pk＝(0,0,L')；
b. Let Pl＝0，D'max0, when k is 1sIf D'i>D'maxLet D'max＝D'i(ii) a When j is 1sThen, let P'l(0,j)Let θ be j × 2 pi/NsIf D'i<γ×D'max(γ is a threshold value for the distance of the two side points of the shaft), then P'l(1,j)＝(D'max×cosθ,D'maxX sin θ, L '), otherwise P'l(1,j)＝(D'j×cosθ,D'jX sin θ, L ') and then gives the array P'l(i,j)；
2_3) execution, the model is sliced, and in each layer, the point of greatest distance to the Z axis is recorded at D'maxOf each point j on the layer and a distance D'jIs re-encoded when the array polyline angle is θ, and the output of the algorithm is P'l(i,j)And the grid is used for the next grid generation.
3) And reconstructing a three-dimensional grid of the defect area based on the optimal symmetry plane and then outputting a three-dimensional model.
3_1, distinguishing array broken lines by using empty broken lines, and finding out non-intersected defect parts to carry out reconstruction processing;
3_2, connecting the corresponding points of the middle sections of the two adjacent layers corresponding to the two non-empty broken line pairs, and adding the generated triangle into the triangle array on the basis of the connecting points;
3_3, adding a new point in the existing triangle array, and searching all circumscribed circles of the triangle containing the new addition point;
3_4. after finding the triangle itself containing the new joining point, then searching its adjacent triangle and the adjacent triangle of the adjacent triangle in an iterative nature until no more circumcircles can be found to envelope the triangle of the new joining point; drawing each triangle in the triangle array to realize the re-modeling of the three-dimensional mesh of the defect area; and then fitting by adopting a Bezier curved surface on the triangular mesh, and outputting a smooth three-dimensional model after smoothing treatment.

Claims (3)

1. The three-dimensional defected face model reconstruction method based on slice data symmetry analysis is characterized by comprising the following steps of:
1) inputting a three-dimensional face defect model, and slicing the three-dimensional model: after the three-dimensional model is input, equally-spaced slicing and layering are carried out on the three-dimensional model perpendicular to a set axial direction, and the layers are further divided into sectors;
2) carrying out symmetry analysis on the slice model obtained in the step 1) to obtain an optimal symmetry plane;
3) reconstructing a three-dimensional grid of the defect area based on the optimal symmetry plane obtained in the step 2) and then outputting a three-dimensional model;
the step 1) is specifically as follows:
1_1), initial setting:
setting an input three-dimensional face model as M, and a slice model obtained by slicing as M1(ii) a Setting the axis as a Z axis; layering the three-dimensional facial model M by equidistant slicing perpendicular to the Z axis to obtain N layerslSetting the height of the layer to be DeltaZSetting the Z-axis direction coordinate of the layer center layer as L; dividing each layer into equal angle sectors by taking the intersection point of the Z axis and each layer as the center, and setting the number of the obtained equal angle sectors of each layer as Ns(ii) a Setting A as a point array formed by three-dimensional points contained in the sector; d is a distance value between the three-dimensional point and the Z axis;
1_2), layering:
setting Z-axis direction coordinates of the bottom layer and the top layer of the layer as Zh＝L+ΔZ(iii) 2 and Zl＝L-ΔZ/2；
In the modelAny three-dimensional point coordinate is p ═ x, y, Z, if Z e (Z) of the three-dimensional point pl,Zh) If yes, the point p belongs to the setting layer;
1_3) sectorization:
each sector having an angle of deltaa＝2π/Ns(ii) a For a three-dimensional point p, let
Figure FDA0002196860630000011
α ═ arccos (x/r), if y of three-dimensional point p>0, the sector number k of the three-dimensional point p is [ α/delta ]a]Otherwise k is equal to Ns-1-[α/Δa]Finally, the three-dimensional point p is taken as the kth value A of the A arraykInserting into an A array;
definition DkDefining D as the maximum value of the distance between the three-dimensional point and the Z axiskIs the maximum value of D if r>DkThen r is assigned to Dk；
1_4) calculating all points on the triangular mesh, only leaving the outermost points, and obtaining a point array A formed by three-dimensional points contained in each sector, namely a slice model M1。
2. The method for reconstructing the three-dimensional defective face model based on slice data symmetry analysis as claimed in claim 1, wherein the step 2) is to perform symmetry analysis on the slice model to obtain the optimal symmetry plane by:
2_1) after slicing, determining the symmetry type of the model and adjusting the digital parameters to enable the three-dimensional model to be adjusted along the Z axis to be closer to the actual axis or pass through a symmetry plane;
a. initializing Nmax＝0，rbest(x,y,z)＝null，NsymR for each rotation of 0j(x, y, z), j 1 … … R, index i, i 1 … … N for each layerl；
b. Combined output L ═ Z + i ΔzTo calculate the arrays A, D, calculate DaAs a stored value for the non-zero distance in D, the relative difference of the average distance is calculated, the relative difference being Max (| D)k-Da|)/Da,k∈(1,2……Ns)；
c. If the relative difference is poor<εDThen let Nsym＝Nsym+1, otherwise, searching the next layer; if N is presentsym>NmaxThen let Nmax＝Nsym，rbest(x,y,z)＝rj(x, y, z), otherwise, performing the next rotation to obtain a rotation coordinate rbest (x, y, z) of the rotation model;
rj(x, y, z) is a rotation matrix for the test of model alignment, where j 1.. R; epsilonDThe threshold value set for the difference between the distance from the outermost point to the z-axis and the average distance from the point on the layer on the z-axis, i.e., the distance value from the outermost point to the z-axis is defined as a subtrahend and the average distance from all points on the layer to the z-axis is defined as a subtrahend, is called εD，N1The number of layers for each rotational symmetry;
2_2) detecting the symmetry of the alignment model and collecting information of asymmetric or missing parts in the layer-layer base model, wherein the missing parts are stored in the broken line pairs to generate the three-dimensional grid;
a. for each layer index l, l 1lCombined output L' ═ Z + L ΔzTo calculate arrays a 'and D'; an array of points P is created, the elements initialized as follows: if A'kNot equal to null, then Pk＝A'k(ii) a Otherwise Pk＝(0,0,L')；
b. Let Pl＝0，D'max0, when i is 1sIf D'i>D'maxLet D'max＝D'i(ii) a When j is 1sThen, let P'l(0,j)Let θ be j × 2 pi/NsIf D'i<γ×D'maxγ is a threshold value for the distance of two side points of the shaft, then P'l(1,j)＝(D'max×cosθ,D'maxX sin θ, L '), otherwise P'l(1,j)＝(D'j×cosθ,D'jX sin θ, L ') and then gives the array P'l(i,j)；
2_3) execution, the model is sliced, and in each layer, toThe point with the largest Z-axis distance is recorded as D'maxWhere the coordinates and distance of each point j on the layer are re-encoded when the array polyline angle is θ, the output is P'l(i,j)And the grid is used for the next grid generation.
3. The method for reconstructing a three-dimensional defective face model based on slice data symmetry analysis as claimed in claim 2, wherein the method for reconstructing a three-dimensional mesh of a defective region based on an optimal symmetry plane and then outputting a three-dimensional model in step 3) comprises:
3_1), distinguishing array broken lines by using empty broken lines, and finding out non-intersected defect parts to carry out reconstruction processing;
3_2), connecting the corresponding points of the middle sections of the two adjacent layers corresponding to the two non-empty broken line pairs, and adding the generated triangle into the triangle array on the basis of the connecting points;
3_3), adding a new point into the existing triangle array, and searching all circumscribed circles of the triangle containing the new addition point;
3_4). after the triangle itself containing the new joining point is found, its adjoining triangles and the adjoining triangles of the adjoining triangles are searched for by iterative nature until no more circumcircles enveloping the triangle of the new joining point can be found; drawing each triangle in the triangle array to realize the re-modeling of the three-dimensional mesh of the defect area; and then fitting the three-dimensional grid by adopting a Bezier curved surface, and outputting a smooth three-dimensional model after smoothing treatment.
CN201610596173.3A 2016-07-27 2016-07-27 Three-dimensional defect face model reconstruction method based on slice data symmetry analysis Active CN106204740B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610596173.3A CN106204740B (en) 2016-07-27 2016-07-27 Three-dimensional defect face model reconstruction method based on slice data symmetry analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610596173.3A CN106204740B (en) 2016-07-27 2016-07-27 Three-dimensional defect face model reconstruction method based on slice data symmetry analysis

Publications (2)

Publication Number Publication Date
CN106204740A CN106204740A (en) 2016-12-07
CN106204740B true CN106204740B (en) 2020-02-07

Family

ID=57495405

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610596173.3A Active CN106204740B (en) 2016-07-27 2016-07-27 Three-dimensional defect face model reconstruction method based on slice data symmetry analysis

Country Status (1)

Country Link
CN (1) CN106204740B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111975781B (en) * 2020-08-25 2021-10-26 北京华航唯实机器人科技股份有限公司 Robot correction method and device and robot
CN113012294B (en) * 2021-03-30 2022-12-13 上海汇像信息技术有限公司 Symmetrical reconstruction method of three-dimensional model

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103729510A (en) * 2013-12-25 2014-04-16 合肥工业大学 Method for computing accurate mirror symmetry of three-dimensional complex model on basis of internal implication transformation
CN104834919A (en) * 2015-05-20 2015-08-12 东南大学 Contour line based three-dimensional human face iteration preprocessing and feature point extracting method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI288894B (en) * 2005-10-17 2007-10-21 Jing-Jing Fang Quantifiable symmetry criterion method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103729510A (en) * 2013-12-25 2014-04-16 合肥工业大学 Method for computing accurate mirror symmetry of three-dimensional complex model on basis of internal implication transformation
CN104834919A (en) * 2015-05-20 2015-08-12 东南大学 Contour line based three-dimensional human face iteration preprocessing and feature point extracting method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Modelling cloud data using an adaptive slicing approach;Y.F. Wu et al;《Computer-Aided Design》;20041231;第231-240页 *
用于头面部装备设计的3D头部模型的标准化处理;郭祖华 等;《计算机辅助设计与图形学学报》;20050731;第17卷(第7期);第1549-1555页 *

Also Published As

Publication number Publication date
CN106204740A (en) 2016-12-07

Similar Documents

Publication Publication Date Title
CN107767442B (en) Foot type three-dimensional reconstruction and measurement method based on Kinect and binocular vision
Ju Fixing geometric errors on polygonal models: a survey
CN102663820B (en) Three-dimensional head model reconstruction method
Huang et al. Manifoldplus: A robust and scalable watertight manifold surface generation method for triangle soups
Attene et al. Polygon mesh repairing: An application perspective
Chang et al. 3D shape engineering and design parameterization
US7640137B2 (en) Shape model generation method and shape model generation system
Khan et al. Surface remeshing: A systematic literature review of methods and research directions
CA3062802C (en) System and methods for 3d model evaluation
CN106504327A (en) A kind of side slope point cloud surface is rebuild and deformation information extracting method
Sun et al. Medial meshes–a compact and accurate representation of medial axis transform
Rasheed et al. A survey of computer methods in reconstruction of 3D archaeological pottery objects
Birdal et al. A point sampling algorithm for 3d matching of irregular geometries
CN106204740B (en) Three-dimensional defect face model reconstruction method based on slice data symmetry analysis
Li et al. 3D modeling of laser-scanned trees based on skeleton refined extraction
Marinov et al. A robust two‐step procedure for quad‐dominant remeshing
Chen et al. Multiscale feature line extraction from raw point clouds based on local surface variation and anisotropic contraction
Yoshizawa et al. Fast and faithful geometric algorithm for detecting crest lines on meshes
Mao et al. Robust surface reconstruction of teeth from raw pointsets
Srivastava et al. Drought stress classification using 3D plant models
Barceló Visual analysis in archaeology. An artificial intelligence approach
Khameneifar Section-specific geometric error evaluation of airfoil blades based on digitized surface data
Shi et al. An integration method for scanned multi-view range images (MRIs) based on local weighted least squares (LWLS) surface fitting
Xu et al. An efficient surface remeshing algorithm based on centroidal power diagram
Ma Surface reconstruction from unorganized point cloud data via progressive local mesh matching

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant