CN111275710B - Dental model segmentation method and device - Google Patents
Dental model segmentation method and device Download PDFInfo
- Publication number
- CN111275710B CN111275710B CN201911244466.5A CN201911244466A CN111275710B CN 111275710 B CN111275710 B CN 111275710B CN 201911244466 A CN201911244466 A CN 201911244466A CN 111275710 B CN111275710 B CN 111275710B
- Authority
- CN
- China
- Prior art keywords
- dental
- model
- dental model
- matrix
- triangular patches
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 230000011218 segmentation Effects 0.000 title claims abstract description 45
- 230000003595 spectral effect Effects 0.000 claims abstract description 15
- 210000004195 gingiva Anatomy 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 75
- 239000013598 vector Substances 0.000 claims description 22
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000005295 random walk Methods 0.000 claims description 4
- 238000003064 k means clustering Methods 0.000 claims description 3
- 210000000214 mouth Anatomy 0.000 description 7
- 238000012937 correction Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000011505 plaster Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 229910052602 gypsum Inorganic materials 0.000 description 1
- 239000010440 gypsum Substances 0.000 description 1
- 210000004400 mucous membrane Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
- G06T2207/30004—Biomedical image processing
- G06T2207/30036—Dental; Teeth
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Dental Tools And Instruments Or Auxiliary Dental Instruments (AREA)
Abstract
The invention discloses a dental model segmentation method, which comprises the following steps. Step S110: a digitized dental model is created having surface color information and geometric feature information. Step S120: the dental model is divided into teeth and gingiva by utilizing a spectral clustering (spectral clustering) method and combining surface color information and geometric characteristic information of the digital dental model. The invention can accurately divide teeth and gingiva of the dental model by referring to the surface color information of the dental model.
Description
Technical Field
The invention relates to a grid segmentation technology in the field of computer image processing, in particular to a method for automatically segmenting triangular patches (triangulated patches) representing teeth and gums from a dental triangular grid (triangulated mesh) model.
Background
The invisible tooth correction technology adopts the technical means that firstly, a digital tooth model of a patient is established, and data support is provided for the design of a virtual correction scheme. In order to build a precise digital tooth model, a dental plaster model consistent with the teeth of a patient needs to be scanned by an optical method to obtain actual dental three-dimensional data of the patient, and then the digital tooth and gum model is segmented by using a computer image processing technology. When the scanned data of the dental plaster model is used for segmentation, only the geometrical characteristics of the dental model are used as segmentation basis, and when the dental model is scanned, the scanning result is a digital model with uniform color, when the digital dental model and the gingival model are segmented, when the gingival and the tooth boundaries are fused together on the dental triangular mesh model, when the boundaries of the gingival and the tooth boundaries are not obvious, the segmentation has errors, and the resolution of the digital model and the mesh reconstruction precision obtained by the scanning mode have limitations, so that the automatic segmentation becomes extremely difficult, and the wearing treatment result of the invisible dental appliance is affected partially.
An intraoral scanning (simply called intraoral scanning) instrument is characterized in that a small-sized probing optical scanning head is used for directly acquiring three-dimensional morphology and color texture information of soft and hard tissue surfaces such as teeth, gums, mucous membranes and the like in the oral cavity of a human body, and generating a vivid color dental model. The mouth scanning data are added with surface color information on the basis of the tooth jaw gypsum model scanning data, so that the boundaries of teeth and gingiva can be distinguished more conveniently.
Therefore, research on a method for better dividing a digitized tooth model and a digitized gum model in the tooth dividing process has important significance.
Disclosure of Invention
The invention aims to solve the technical problem that the acquired surface color information of the digital dental model is utilized to assist in dividing the digital dental model and the digital gingival model, so as to provide more accurate data for the design of a subsequent correction scheme.
In order to solve the technical problems, the invention provides a first dental model segmentation method, which comprises the following steps. Step S110: a digitized dental model is created having surface color information and geometric feature information. Step S120: the dental model is divided into teeth and gingiva by utilizing a spectral clustering (spectral clustering) method and combining surface color information and geometric characteristic information of the digital dental model. The above method is an embodiment of the present invention for dividing the entire teeth and the entire gums of the dental model.
Further, in the step S110, intraoral data is acquired in the oral cavity of the human body by using an intraoral scanner, and a digital dental model is built based on the intraoral data. The oral scan data includes surface color information of the dental jaw, and is suitable for establishing a digital dental model with the surface color information and the geometric characteristic information.
Further, in the step S110, the digitized dental model is a digital model using triangular meshes, and each triangular mesh is called a triangular patch. This is a common implementation of a digitized dental model.
Further, the step S120 further includes the following steps; step S122: calculating a similarity matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches; step S124: calculating a normalized Laplace matrix according to the similarity matrix; step S126: calculating two eigenvectors corresponding to the minimum eigenvalue and the next smallest eigenvalue of the normalized Laplace matrix; step S128: the elements in the two eigenvectors are divided into two classes by using a k-means clustering method or a maximum inter-class variance method, namely, triangular patches in a dental triangle network model are divided into two classes, wherein one class represents teeth and the other class represents gums. This is a specific implementation of step S120.
Further, in the step S122, a calculation formula of the distance Dist (i, j) between the adjacent triangular patches i, j in the dental triangular mesh model is as follows;
where Col_Dist (i, j) is the color function between adjacent triangular patches i, j, ang_Dist (α) ij ) Is the angular function between adjacent triangular patches i, j, and avg represents the average. Col_dist (i, j) =col (i) -Col (j); where Col (i) represents the color of triangular patch i. Ang_Dist (alpha) ij )＝η(1-cosα ij ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein alpha is ij Is the dihedral angle between adjacent triangular patches i, j; η represents the concave-convex state between the adjacent triangular patches i, j. A method of calculating the distance between adjacent triangular patches is presented herein.
Further, in the step S122, the similarity matrix W is an n×n matrix, where n represents the number of triangular plates of the dental model; the calculation formula of each element in the similarity matrix W is as follows;
wherein,a method of computing a similarity matrix is presented herein.
Further, in the step S124, the normalized laplace matrix uses a symmetric normalized laplace matrix L sym 。L sym ＝D- 1/2 LD- 1/2 ＝I–D- 1/2 WD- 1/2 . Wherein D is a diagonal matrix with elements on the diagonal of D 1 ，d 2 ，…，d n ；I is an identity matrix, and elements on diagonal lines of the identity matrix are all 1. A first calculation of the normalized laplacian matrix is presented herein.
Further, in the step S124, the normalized laplace matrix uses a random walk normalized laplace matrix L rw 。L rw ＝D- 1 L＝I–D- 1 W. Wherein D is a diagonal matrix with elements on the diagonal of D 1 ，d 2 ，…，d n ；I is an identity matrix, and elements on diagonal lines of the identity matrix are all 1. A second calculation of the normalized laplacian matrix is presented herein.
Further, in the step S126, each feature vector corresponds to a feature attribute; the two feature vectors respectively represent different feature attributes; the number of elements contained in each feature vector is the same as the number of vertexes in the dental triangle mesh model, and each element represents one feature attribute of one vertex. This is a specific description of the feature vectors and feature values.
The invention also provides a second dental model segmentation method, which comprises the following steps. Step S110: a digitized dental model is created having surface color information and geometric feature information. Step S130: and (3) carrying out segmentation between each tooth and between the teeth and gingiva on the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the dental model. The above-described method is a second embodiment of the present invention for dividing the entire gingiva and each individual tooth of the dental model.
Further, in the step S110, the digitized dental model is a digital model using triangular meshes, and each triangular mesh is called a triangular patch. This is a common implementation of a digitized dental model.
Further, the step S130 further includes the following steps. Step S122: and calculating a similarity matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches. Step S124: and calculating a normalized Laplace matrix according to the similarity matrix. Step S136: calculating m+1 eigenvectors corresponding to the minimum m+1 eigenvalues of the normalized Laplace matrix; wherein m represents the number of teeth;
step S138: dividing the elements in the m+1 eigenvectors into m+1 classes by using a k-average clustering method or a maximum inter-class variance method, namely dividing triangular patches in a dental triangle network model into m+1 classes, wherein m classes respectively represent m teeth, and the other class represents gums.
This is a specific implementation of step S130.
Further, in the step S136, each feature vector corresponds to a feature attribute; the m+1 feature vectors respectively represent different feature attributes; the number of elements contained in each feature vector is the same as the number of vertexes in the dental triangle mesh model, and each element represents one feature attribute of one vertex. This is a specific description of the feature vectors and feature values.
The invention also provides a first dental model segmentation device which comprises the following units. A model building unit for building a digitized dental model having surface color information and geometric feature information. The segmentation unit is used for segmenting teeth and gingiva of the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the dental model. The above-described device is a first embodiment of the present invention for dividing the entire teeth and the entire gums of the dental model.
The invention also provides a second dental model segmentation device, which comprises the following units. A model building unit for building a digitized dental model having surface color information and geometric feature information. The segmentation unit is used for carrying out the segmentation of each tooth and the tooth and gum on the dental model by utilizing a spectral clustering method and combining the surface color information and the geometric characteristic information of the dental model. The above-described device is a second embodiment of the present invention, and is used to divide the entire gingiva and each individual tooth of the dental model.
According to the invention, the intraoral information of the patient is acquired by using the intraoral scanner, in particular to the acquisition of the digital dental model, wherein the digital dental model comprises the digital dental model and the digital gingival model, and the scanning result of the intraoral scanner comprises the color information of the intraoral of the patient, such as a white dental image and a red gingival image, the acquired digital dental model can be more accurately segmented by utilizing the color difference in combination with a specific segmentation method, and further, a single digital dental model can be segmented, so that more accurate data information is provided for the design of a subsequent correction scheme, and the subsequently manufactured invisible dental corrector for correction is more attached to the scheme of the patient and is more approximate to an expected correction target.
Drawings
Fig. 1 is a flowchart of a first embodiment of a dental model segmentation method according to the present invention.
Fig. 2 is a flow chart of one specific implementation of step S120 in fig. 1.
Fig. 3 is a flowchart of a second embodiment of a dental model segmentation method provided by the present invention.
Fig. 4 is a flow chart of one specific implementation of step S130 in fig. 3.
Fig. 5 is a schematic structural view of a first embodiment of the dental model dividing device provided by the invention.
Fig. 6 is a schematic structural view of a second embodiment of the dental model dividing device provided by the invention.
The reference numerals in the drawings illustrate: 10 is a model building unit; 20. 30 is a dividing unit.
Detailed Description
Referring to fig. 1, an embodiment of a dental model segmentation method according to the present invention includes the following steps.
Step S110: a digitized dental model is created having surface color information and geometric feature information. This step is, for example, to acquire intraoral data from an intraoral scanner in a human mouth and to create a triangle mesh model of the dental jaw based on the intraoral data. Since the mouth scan data contains the surface color information of each part, the dental model created based on the mouth scan data also has the surface color information.
Preferably, the digital dental model is a digital model using a triangular mesh. Each triangular mesh comprises three vertexes and three sides, and the vertexes and the sides enclose a triangular surface patch.
Step S120: and (3) carrying out tooth and gum segmentation on the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the digital dental model.
Referring to fig. 2, the step S120 further includes the following steps.
Step S122: and calculating a similarity matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches.
The calculation formula of the distance Dist (i, j) between adjacent triangular patches i, j in the dental triangular mesh model is as follows;
where Col_Dist (i, j) is the color function between adjacent triangular patches i, j, ang_Dist (α) ij ) Is the angular function between adjacent triangular patches i, j, and avg represents the average.
Col_dist (i, j) =col (i) -Col (j); where Col (i) represents the color of triangular patch i.
Ang_Dist(α ij )＝η(1-cosα ij ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein alpha is ij Is the dihedral angle between adjacent triangular patches i, j; η represents the concave-convex state between the adjacent triangular patches i, j. Alpha ij And the angle is larger than 180 degrees, the adjacent triangular patches i and j are convex, and eta is a first value. Alpha ij And the angle is smaller than 180 degrees, the adjacent triangular patches i and j are concave, and eta is a second value.
The similarity matrix W is an n x n matrix, where n represents the number of triangular plates of the dental model. The values of the corresponding positions in the similarity matrix W are obtained from a gaussian function of the distance Dist (i, j) between adjacent triangular patches i, j in the dental triangular mesh model. The calculation formula of each element in the similarity matrix W is as follows;
wherein,
step S124: and calculating a normalized Laplace matrix according to the similarity matrix.
For example, a normalized Laplace matrix employs a symmetric normalized Laplace (Symmetric normalized Laplacian) matrix L sym 。
L sym ＝D -1/2 LD -1/2 ＝I–D -1/2 WD -1/2 . Wherein D is a diagonal matrix with elements on the diagonal of D 1 ，d 2 ，…，d n ；I is an identity matrix, and elements on diagonal lines of the identity matrix are all 1.
As another example, the normalized laplace matrix employs a random walk normalized laplace (Random walk normalized Laplacian) matrix L rw 。
L rw ＝D -1 L＝I–D -1 W。
Wherein D is a diagonal matrix with elements on the diagonal of D 1 ，d 2 ，…，d n ；I is an identity matrix, and elements on diagonal lines of the identity matrix are all 1.
Step S126: and calculating two eigenvectors corresponding to the minimum eigenvalue and the next smallest eigenvalue of the normalized Laplace matrix.
Each feature vector corresponds to a feature attribute; the two feature vectors respectively represent different feature attributes; the number of elements contained in each feature vector is the same as the number of vertexes in the dental triangle mesh model, and each element represents one feature attribute of one vertex.
Step S128: the elements in the two eigenvectors are divided into two classes by using a k-means (k-means) clustering method or a maximum inter-class variance method, namely, triangular patches in a dental triangle network model are divided into two classes, wherein one class represents teeth and the other class represents gums.
The maximum inter-class variance method is to traverse all elements in the feature vector corresponding to the small feature value, divide the elements of the feature vector into two classes when finding the best element as a segmentation threshold, and calculate the mean value and the inter-class variance of the elements of the two classes so as to maximize the inter-class variance.
Referring to fig. 3, a second embodiment of the dental model segmentation method according to the present invention includes the following steps.
Step S110: a digitized dental model is created having surface color information and geometric feature information. Preferably, an intraoral scanner is used to acquire intraoral data in a human mouth and a digitized dental model is created based on the intraoral data. Preferably, the digitized dental model is a digital model using triangular meshes, each of which is referred to as a triangular patch.
Step S130: and (3) carrying out segmentation between each tooth and between the teeth and gingiva on the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the dental model.
Referring to fig. 4, the step S130 further includes the following steps.
Step S122: and calculating a similarity matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches.
Step S124: and calculating a normalized Laplace matrix according to the similarity matrix.
Step S136: calculating m+1 eigenvectors corresponding to the minimum m+1 eigenvalues of the normalized Laplace matrix; wherein m represents the number of teeth;
each feature vector corresponds to a feature attribute; the m+1 feature vectors respectively represent different feature attributes; the number of elements contained in each feature vector is the same as the number of vertexes in the dental triangle mesh model, and each element represents one feature attribute of one vertex.
Step S138: dividing the elements in the m+1 eigenvectors into m+1 classes by using a k-average clustering method or a maximum inter-class variance method, namely dividing triangular patches in a dental triangle network model into m+1 classes, wherein m classes respectively represent m teeth, and the other class represents gums.
Referring to fig. 5, an embodiment of a dental model dividing apparatus according to the present invention corresponds to an embodiment of a dental model dividing method shown in fig. 1. An embodiment of the dental model segmentation apparatus comprises a model building unit 10 and a segmentation unit 20.
The model building unit 10 is used for building a digital dental model with surface color information and geometric feature information;
the segmentation unit 20 is used for segmenting teeth and gums of the dental model by using a spectral clustering method in combination with surface color information and geometric feature information of the dental model.
Referring to fig. 6, a second embodiment of the dental model dividing apparatus provided by the present invention corresponds to the second embodiment of the dental model dividing method shown in fig. 3. The embodiment of the dental model dividing device comprises a model building unit 10 and a dividing unit 30.
The model building unit 10 is used for building a digital dental model with surface color information and geometric feature information;
the segmentation unit 30 is used for performing the segmentation of each tooth and gum of the dental model by using a spectral clustering method in combination with surface color information and geometric feature information of the dental model.
The invention can accurately divide teeth and gingiva of the dental model by referring to the surface color information of the dental model.
The above are only preferred embodiments of the present invention, and are not intended to limit the present invention. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (13)
1. A dental model segmentation method is characterized by comprising the following steps: :
step S110: establishing a digital dental model with surface color information and geometric characteristic information;
step S120: carrying out tooth and gum segmentation on the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the digital dental model;
the step S120 further includes the steps of:
step S122: calculating a similarity matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches;
step S124: calculating a normalized Laplace matrix according to the similarity matrix;
step S126: calculating two eigenvectors corresponding to the minimum eigenvalue and the next smallest eigenvalue of the normalized Laplace matrix;
step S128: the elements in the two eigenvectors are divided into two classes by using a k-means clustering method or a maximum inter-class variance method, namely, triangular patches in a dental triangle network model are divided into two classes, wherein one class represents teeth and the other class represents gums.
2. The method according to claim 1, wherein in step S110, an intraoral scanner is used to acquire intraoral data in the human mouth, and a digital dental model is created based on the intraoral data.
3. The dental model segmentation method according to claim 1, wherein in the step S110, the digitized dental model is a digital model using triangular meshes, each of which is called a triangular patch.
4. The dental model segmentation method according to claim 1, wherein in the step S122, a calculation formula of a distance (i, j) between adjacent triangular patches i, j in the dental triangular mesh model is as follows;
where Col_Dist (i, j) is the color function between adjacent triangular patches i, j, ang_Dist (α) ij ) Is an angle function between adjacent triangular patches i and j, and avg represents average calculation;
col_dist (i, j) =col (i) -Col (j); where Col (i) represents the color of triangular patch i;
Ang_Dist(α ij )＝η(1-cosα ij ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein alpha is ij Is the dihedral angle between adjacent triangular patches i, j; η represents the concave-convex state between the adjacent triangular patches i, j.
5. The dental model segmentation method according to claim 4, wherein in the step S122, the similarity matrix W is an n×n matrix, where n represents the number of triangular plates of the dental model; the calculation formula of each element in the similarity matrix W is as follows;
wherein,
6. the dental model segmentation method according to claim 5, wherein in the step S124, the normalized laplace matrix is a symmetric normalized laplace matrix L sym ；
L sym ＝D -1/2 LD -1/2 ＝I–D -1/2 WD -1/2 ；
Wherein D is a diagonal matrix with elements on the diagonal of D 1 ，d 2 ，…，d n ；
I is an identity matrix, and elements on diagonal lines of the identity matrix are all 1.
7. The dental model segmentation method according to claim 5, wherein in the step S124, the normalized laplace matrix is a random walk normalized laplace matrix L rw ；
L rw ＝D -1 L＝I–D -1 W；
Wherein D is a diagonal matrix with elements on the diagonal of D 1 ，d 2 ，…，d n ；
I is an identity matrix, and elements on diagonal lines of the identity matrix are all 1.
8. The dental model segmentation method according to claim 6 or 7, wherein in the step S126, each feature vector corresponds to a feature attribute; the two feature vectors respectively represent different feature attributes; the number of elements contained in each feature vector is the same as the number of vertexes in the dental triangle mesh model, and each element represents one feature attribute of one vertex.
9. A dental model segmentation method is characterized by comprising the following steps: :
step S110: establishing a digital dental model with surface color information and geometric characteristic information;
step S130: dividing each tooth and gum of the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the dental model;
the step S130 further includes the steps of:
step S122: calculating a similarity matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches;
step S124: calculating a normalized Laplace matrix according to the similarity matrix;
step S136: calculating m+1 eigenvectors corresponding to the minimum m+1 eigenvalues of the normalized Laplace matrix; wherein m represents the number of teeth;
step S138: dividing the elements in the m+1 eigenvectors into m+1 classes by using a k-average clustering method or a maximum inter-class variance method, namely dividing triangular patches in a dental triangle network model into m+1 classes, wherein m classes respectively represent m teeth, and the other class represents gums.
10. The dental model segmentation method according to claim 9, wherein in the step S110, the digitized dental model is a digital model using triangular meshes, each of which is called a triangular patch.
11. The dental model segmentation method according to claim 10, wherein in the step S136, each feature vector corresponds to a feature attribute; the m+1 feature vectors respectively represent different feature attributes; the number of elements contained in each feature vector is the same as the number of vertexes in the dental triangle mesh model, and each element represents one feature attribute of one vertex.
12. A dental model dividing device, comprising the following units:
a model building unit for building a digital dental model with surface color information and geometric feature information;
the segmentation unit is used for segmenting teeth and gingiva of the dental model by utilizing a spectral clustering method and combining surface color information and geometric characteristic information of the dental model;
the segmentation unit is also used for calculating a similar matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches; calculating a normalized Laplace matrix according to the similarity matrix; calculating two eigenvectors corresponding to the minimum eigenvalue and the next smallest eigenvalue of the normalized Laplace matrix; the elements in the two eigenvectors are divided into two classes by using a k-means clustering method or a maximum inter-class variance method, namely, triangular patches in a dental triangle network model are divided into two classes, wherein one class represents teeth and the other class represents gums.
13. A dental model dividing device, comprising the following units:
a model building unit for building a digital dental model with surface color information and geometric feature information;
the segmentation unit is used for carrying out segmentation between each tooth and between the teeth and the gingiva on the dental model by utilizing a spectral clustering method and combining the surface color information and the geometric characteristic information of the dental model;
the segmentation unit is also used for calculating a similar matrix according to the distance between adjacent triangular patches in the dental triangular mesh model, wherein the distance between the adjacent triangular patches is comprehensively obtained by a color function and an angle function between the adjacent triangular patches; calculating a normalized Laplace matrix according to the similarity matrix; calculating m+1 eigenvectors corresponding to the minimum m+1 eigenvalues of the normalized Laplace matrix; wherein m represents the number of teeth; dividing the elements in the m+1 eigenvectors into m+1 classes by using a k-average clustering method or a maximum inter-class variance method, namely dividing triangular patches in a dental triangle network model into m+1 classes, wherein m classes respectively represent m teeth, and the other class represents gums.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311758275.7A CN117830323A (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
| CN201911244466.5A CN111275710B (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911244466.5A CN111275710B (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311758275.7A Division CN117830323A (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111275710A CN111275710A (en) | 2020-06-12 |
| CN111275710B true CN111275710B (en) | 2024-02-27 |
Family
ID=71111840
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311758275.7A Pending CN117830323A (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
| CN201911244466.5A Active CN111275710B (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311758275.7A Pending CN117830323A (en) | 2019-12-06 | 2019-12-06 | Dental model segmentation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN117830323A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706971A (en) * | 2009-11-27 | 2010-05-12 | 西安科技大学 | Automatic division method of dental crowns in dental models |
| CN102147935A (en) * | 2010-02-09 | 2011-08-10 | 常州高新技术产业开发区三维工业技术研究所有限公司 | Method suitable for segmenting dental triangle mesh curved surface from dental jaw triangle mesh curved surface |
| CN105046750A (en) * | 2015-08-24 | 2015-11-11 | 杭州美齐科技有限公司 | Method for automatically segmenting whole dental triangular mesh model |
| WO2016197370A1 (en) * | 2015-06-11 | 2016-12-15 | 深圳先进技术研究院 | Segmentation and reconstruction method and device for teeth and alveolar bone |
| CN106504331A (en) * | 2016-09-27 | 2017-03-15 | 西安科技大学 | Tooth modeling method based on three-dimensional model search |
| WO2018220248A1 (en) * | 2017-06-01 | 2018-12-06 | Pedro Perales Padilla | Method for designing a dental prosthesis using digital techniques |
| CN108986111A (en) * | 2018-07-02 | 2018-12-11 | 西安增材制造国家研究院有限公司 | A kind of three-dimensional dentognathic model dividing method for area of computer aided stealth correction |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102011010975A1 (en) * | 2011-02-10 | 2012-08-16 | Martin Tank | Method and analysis system for geometrical analysis of scan data of oral structures |
| US20170325909A1 (en) * | 2016-05-12 | 2017-11-16 | Fei Gao | Method for facilitating a tooth repositioning dental treatment |
| CN108986123A (en) * | 2017-06-01 | 2018-12-11 | 无锡时代天使医疗器械科技有限公司 | The dividing method of tooth jaw three-dimensional digital model |
-
2019
- 2019-12-06 CN CN202311758275.7A patent/CN117830323A/en active Pending
- 2019-12-06 CN CN201911244466.5A patent/CN111275710B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706971A (en) * | 2009-11-27 | 2010-05-12 | 西安科技大学 | Automatic division method of dental crowns in dental models |
| CN102147935A (en) * | 2010-02-09 | 2011-08-10 | 常州高新技术产业开发区三维工业技术研究所有限公司 | Method suitable for segmenting dental triangle mesh curved surface from dental jaw triangle mesh curved surface |
| WO2016197370A1 (en) * | 2015-06-11 | 2016-12-15 | 深圳先进技术研究院 | Segmentation and reconstruction method and device for teeth and alveolar bone |
| CN105046750A (en) * | 2015-08-24 | 2015-11-11 | 杭州美齐科技有限公司 | Method for automatically segmenting whole dental triangular mesh model |
| CN106504331A (en) * | 2016-09-27 | 2017-03-15 | 西安科技大学 | Tooth modeling method based on three-dimensional model search |
| WO2018220248A1 (en) * | 2017-06-01 | 2018-12-06 | Pedro Perales Padilla | Method for designing a dental prosthesis using digital techniques |
| CN108986111A (en) * | 2018-07-02 | 2018-12-11 | 西安增材制造国家研究院有限公司 | A kind of three-dimensional dentognathic model dividing method for area of computer aided stealth correction |
Non-Patent Citations (2)
| Title |
|---|
| 李剑 ; 严斌 ; 王林 ; 张文健 ; 鲍旭东 ; 赵孟 ; .数字化三维牙颌模型中牙齿边界的自动提取.口腔医学.2008,(07),全文. * |
| 范然 ; 钮叶新 ; 金小刚 ; 顾书华 ; 施洁 ; 王慧明 ; .计算机辅助牙齿隐形正畸系统.计算机辅助设计与图形学学报.2013,(01),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111275710A (en) | 2020-06-12 |
| CN117830323A (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105447908B (en) | Dental arch model generation method based on oral cavity scan data and CBCT data | |
| US11735306B2 (en) | Method, system and computer readable storage media for creating three-dimensional dental restorations from two dimensional sketches | |
| US10713396B2 (en) | Method of aligning intra-oral digital 3D models | |
| US8135569B2 (en) | System and method for three-dimensional complete tooth modeling | |
| JP6872556B2 (en) | Population-based surface mesh reconstruction | |
| US11622843B2 (en) | Processing digital dental impression | |
| CA3024372A1 (en) | Method for estimating at least one of shape, position and orientation of a dental restoration | |
| JP2019536561A (en) | Method and system for brace removal from dentition mesh | |
| CN111784754B (en) | Tooth orthodontic method, device, equipment and storage medium based on computer vision | |
| CN111968146B (en) | Three-dimensional dental mesh model segmentation method | |
| WO2023274413A1 (en) | Three-dimensional scanning system, auxiliary member, processing method and apparatus, device and medium | |
| US10456225B2 (en) | Method for clearing of virtual representations of objects | |
| US20230127167A1 (en) | Processing digital dental impression | |
| CN112308895A (en) | Method for constructing realistic dentition model | |
| CN115457198A (en) | Tooth model generation method and device, electronic equipment and storage medium | |
| CN114549540A (en) | Method for automatically fusing oral scanning tooth data and CBCT (Cone Beam computed tomography) data and application thereof | |
| CN111275710B (en) | Dental model segmentation method and device | |
| CN107638224A (en) | A kind of shape-position parameter measurement method for digitizing artificial tooth | |
| WO2022174747A1 (en) | Method for segmenting computed tomography image of teeth | |
| CN114758073A (en) | Oral cavity digital system based on RGBD input and flexible registration | |
| US20180130184A1 (en) | Method for clearing of virtual representatations of objects | |
| CN117788747A (en) | Dental crown three-dimensional reconstruction method by fusing oral scan data and CBCT | |
| WO2024088360A1 (en) | Method for registering three-dimensional digital models of dentition | |
| US20230087800A1 (en) | Automated tooth administration in a dental restoration workflow | |
| CN108814742B (en) | Design method and device of dental tray, user terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 122, 2305 Zuchongzhi Road, Pudong New Area, Shanghai, 201203 Applicant after: Zhengya Dental Technology (Shanghai) Co.,Ltd. Address before: Room 122, 2305 Zuchongzhi Road, Pudong New Area, Shanghai, 201203 Applicant before: SHANGHAI SMARTEE DENTI-TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |