CN112396609A - Dentition segmentation method, dentition segmentation device and electronic equipment - Google Patents
Dentition segmentation method, dentition segmentation device and electronic equipment Download PDFInfo
- Publication number
- CN112396609A CN112396609A CN201910702794.9A CN201910702794A CN112396609A CN 112396609 A CN112396609 A CN 112396609A CN 201910702794 A CN201910702794 A CN 201910702794A CN 112396609 A CN112396609 A CN 112396609A
- Authority
- CN
- China
- Prior art keywords
- dentition
- tooth
- model
- class
- feature points
- 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.)
- Granted
Links
- 210000004513 dentition Anatomy 0.000 title claims abstract description 199
- 230000036346 tooth eruption Effects 0.000 title claims abstract description 199
- 230000011218 segmentation Effects 0.000 title claims abstract description 129
- 238000000034 method Methods 0.000 title claims abstract description 85
- 230000006870 function Effects 0.000 claims description 33
- 238000004422 calculation algorithm Methods 0.000 claims description 23
- 238000009499 grossing Methods 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 11
- 238000009792 diffusion process Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 239000013598 vector Substances 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 238000005295 random walk Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 2
- 238000009826 distribution Methods 0.000 claims description 2
- 230000005484 gravity Effects 0.000 claims description 2
- 239000011505 plaster Substances 0.000 claims description 2
- 238000000513 principal component analysis Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 238000010845 search algorithm Methods 0.000 claims description 2
- 230000001131 transforming effect Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 15
- 238000012545 processing Methods 0.000 description 15
- 230000002093 peripheral effect Effects 0.000 description 12
- 238000012937 correction Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000003860 storage Methods 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000005055 memory storage Effects 0.000 description 2
- 238000009877 rendering Methods 0.000 description 2
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000013013 elastic material Substances 0.000 description 1
- 210000004195 gingiva Anatomy 0.000 description 1
- 210000004283 incisor Anatomy 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000036244 malformation Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 210000000214 mouth Anatomy 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20156—Automatic seed setting
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- General Health & Medical Sciences (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Dental Tools And Instruments Or Auxiliary Dental Instruments (AREA)
Abstract
A dentition segmentation method, a dentition segmentation device and electronic equipment are provided, wherein the dentition segmentation method comprises the following steps: selecting a first class of feature points on the digital dentition model, wherein the first class of feature points are grid vertexes for guiding the segmentation of each single tooth in the dentition; classifying second feature points in the digital dentition model according to the first feature points, and determining teeth to which the second feature points belong, wherein the second feature points are mesh vertexes used for representing the whole shape of the digital dentition model; and respectively merging the second type of characteristic points belonging to each tooth to obtain the digital tooth area of each single tooth after the digital dentition model is segmented. Because the two types of feature points are selected based on the whole digital dentition model, the classification information of the feature points covers the classification features of the whole digital dentition model, even if noise data exist in the model, the noise data can be evenly distributed into the global data, and the single tooth can be more accurately segmented.
Description
Technical Field
The invention relates to the technical field of tooth orthodontics, in particular to a dentition segmentation method, a tooth segmentation device and electronic equipment.
Background
The malformation of the jaw is one of three diseases of the oral cavity, and has high prevalence rate. The traditional tooth correction is mainly to arrange a bracket and an arch wire on the surface of a dentition, plays a role in correction and corrects in a twisting and pushing mode. For aesthetic reasons, the appliance is transferred from the labial side to the lingual side and cannot be seen in appearance, but the method has high requirements on doctors, causes great oral injury to patients, enhances the foreign body sensation and has high manufacturing cost. With the progress of technology, the invisible correction is more and more accepted and used by patients, the bracket-free invisible correction is a transparent elastic material movable correction device designed and manufactured by computer assistance, the bracket-free invisible correction device is a sequential continuous correction device, and the aim of correcting teeth is achieved by continuously moving the teeth in a small range. The appliance can control the size of the correction force and the time of the correction force, only some teeth can move at different stages, and other teeth are used as anchorage, so that the correction of the teeth is completed. In this process, it is necessary to segment a single tooth and then perform subsequent tooth arrangement, so the quality of the segmentation of the single tooth affects the accuracy of the subsequent tooth arrangement.
Most of the existing three-dimensional tooth segmentation methods are to separate single independent teeth by extracting tooth morphology features of a tooth jaw model by using a graphic image processing technology, however, due to the limitation of three-dimensional scanning precision, boundary lines of dentition and gingiva on a triangular mesh tooth jaw model and boundary lines between adjacent teeth are usually fused together, and tooth morphologies are different, so that a good segmentation result is difficult to obtain by using the traditional segmentation method, and the problem is particularly prominent when a patient has malformed teeth. The prior incisor process usually adopts a method of manually drawing or adjusting contour lines of all teeth on a digital tooth-jaw model according to the experience of a designer, and the accuracy of the method depends on the experience degree of the doctor and is not easy to guarantee.
It is difficult to realize fully automatic tooth segmentation, because the tooth shapes are different and the tooth arrangement of different people is also very different, so a large number of triangular patches for user interactive setting of marks are also needed.
Disclosure of Invention
The application provides a dentition segmentation method, a dentition segmentation device and electronic equipment, which can realize automatic segmentation of teeth of a triangular patch without user interaction and ensure the integrity of the segmented teeth.
According to a first aspect, there is provided a method of tooth segmentation for a digitized dentition model, comprising the steps of:
selecting a first class of characteristic points on a digital dentition model to be segmented, wherein the digital dentition model is a triangular patch model, and the first class of characteristic points are triangular patch vertexes which are selected based on the digital dentition model and are used for guiding the segmentation of each single tooth in the dentition;
classifying second feature points in the digital dentition model according to the first feature points to determine teeth to which the second feature points belong, wherein the second feature points are selected based on the digital dentition model and are used for representing the vertexes of a triangular patch of the overall shape of the digital dentition model;
and respectively merging the second characteristic points belonging to each tooth to obtain the digital tooth area of each single tooth after the digital dentition model is segmented.
According to a second aspect, there is provided a tooth segmentation method for a digital dental model, comprising the steps of:
determining an occlusion plane of a digital dental model, wherein the digital dental model is a triangular patch model;
extracting characteristic areas of dentition and gum boundary in the digital dental model;
projecting the characteristic region to the occlusion plane to construct a characteristic binary image, calculating the outer contour of the characteristic binary image and mapping the outer contour to the characteristic region to obtain a gum line;
segmenting the digitized dental model into dentition regions and gum regions based on a gum line;
and (3) segmenting the tooth area by adopting the dentition segmentation method to obtain the digital area of each single tooth.
According to a third aspect, there is provided a dentition segmentation apparatus for digitizing a dentition model, comprising:
the digital dentition segmentation system comprises a selection module, a segmentation module and a segmentation module, wherein the selection module is used for selecting a first class of characteristic points on a digital dentition model to be segmented, the digital dentition model is a triangular patch model, and the first class of characteristic points are triangular patch vertexes which are selected based on the digital dentition model and are used for guiding the segmentation of each single tooth in the dentition;
the classification module is used for classifying second feature points in the digital dentition model according to the first feature points and determining teeth to which the second feature points belong, wherein the second feature points are selected based on the digital dentition model and are used for representing the vertexes of a triangular patch of the overall shape of the digital dentition model;
and the merging module is used for respectively merging the second type of characteristic points belonging to each tooth to obtain the digital tooth area of each single tooth after the digital dentition model is segmented.
According to a fourth aspect, there is provided an electronic device comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by the processor to implement the dentition segmentation method described above.
According to a fifth aspect, there is provided a tooth segmentation apparatus for digitizing a dental model, comprising:
the determination module is used for determining an occlusion plane of the digital dental model, and the digital dental model is a triangular patch model;
the extraction module is used for extracting characteristic areas of dentition and gum boundary in the digital dental model;
the obtaining module is used for projecting the characteristic region to the occlusion plane to construct a characteristic binary image, calculating the outer contour of the characteristic binary image and mapping the outer contour to the characteristic region to obtain a gum line;
a first segmentation module for segmenting the digitized dental model into dentition regions and gum regions based on a gum line;
and the second segmentation module is used for segmenting the tooth area by adopting the tooth segmentation method to obtain the digital area of each single tooth.
According to a sixth aspect, there is provided an electronic device comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by the processor to implement the tooth segmentation method described above.
According to the technical scheme, the technical effects are as follows:
the first type of feature points are selected on the whole digital dentition model, then the second type of feature points on the digital dentition model are classified and collected according to the first type of feature points, and the segmentation of the single tooth is realized.
In addition, compared with the existing method that the feature points are marked by user interaction in the automatic segmentation of the triangular patch, the technical scheme of the method is that the first class feature points and the second class feature points are selected by automatic identification, and the automatic tooth segmentation can be completed without user interaction identification.
Drawings
FIG. 1 is a flowchart illustrating a dentition segmentation method according to an embodiment;
FIG. 2 is a flowchart illustrating a first class of feature points being automatically identified according to a first embodiment;
FIG. 3 is a schematic diagram of a digitized dentition model;
FIG. 4 is a diagram illustrating a height function value found in a digitized dentition model as a local minimum;
FIG. 5 is a schematic diagram illustrating arrangement of feature points of a first type in a digital dentition model;
FIG. 6 is a schematic diagram of a digitized region of a single tooth after the digitized dentition model is segmented;
FIG. 7 is a flowchart of a tooth segmentation method according to the second embodiment;
FIG. 8 is a schematic diagram of the segmentation of tooth areas and tooth silver areas in the digital dental model;
FIG. 9 is a block diagram showing a dentition segmentation apparatus according to a third embodiment;
FIG. 10 is a block diagram showing the configuration of an electronic apparatus according to a fourth embodiment;
FIG. 11 is a block diagram of a tooth segmentation apparatus according to a fifth embodiment;
fig. 12 is a block diagram of an electronic device according to a sixth embodiment.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and accompanying drawings.
The first embodiment is as follows:
the present embodiment provides a tooth segmentation method for a digitized dentition model, and a specific flowchart is shown in fig. 1, which specifically includes the following steps.
S100: selecting a first class of characteristic points on a digital dentition model to be segmented, wherein the digital dentition model is a triangular patch model.
S101: and classifying the second class of feature points in the digital dentition model according to the first class of feature points, and determining the tooth to which each second class of feature point belongs.
S102: and respectively merging the second characteristic points belonging to each tooth to obtain the digital tooth area of each single tooth after the digital dentition model is segmented.
The first type of characteristic points are triangular patch vertexes which are selected based on a digital dentition model and used for guiding the segmentation of each single tooth in the dentition, and the second type of characteristic points are triangular patch vertexes which are selected based on the digital dentition model and used for representing the overall shape of the digital dentition model; that is, the first type of feature points is used to guide the segmentation of the dentition, and the second type of feature points is the feature points when the dentition is specifically segmented; through the segmentation guidance of the first class of feature points, the second class of feature points can be accurately classified to each tooth, and the segmentation precision of dentitions is further improved.
The first type of feature points are selected on the whole digital dentition model, then the second type of feature points on the digital dentition model are classified and collected according to the first type of feature points, and the segmentation of the single tooth is realized.
The above steps will be described in detail below.
In step S100, the first-class feature points are obtained by an automatic identification method, and this example provides a specific implementation method for automatically identifying the first-class feature points, and a specific flowchart thereof is shown in fig. 2, and specifically includes the following steps.
S200: and defining a vertex height function of the triangular patch mesh in the digital dentition model.
The digital dentition model is a triangular patch model, as shown in fig. 3, the digital dentition model is a mesh model composed of all triangular patches, and each triangular patch includes a corresponding vertex and an edge.
In step S200, a vertex height function is defined according to the vertex curvature of the triangular patch in the digitized dentition model and the distance from the vertex to the base of the dentition, and specifically, the vertex height function of the digitized dentition model is defined as: h (-H)1)+(-H2) Where H is the vertex height function, H1For digitizing teethVertex curvature of column model, H2To digitize the distance from the apex of the dentition model to the base of the dentition, and, for ease of subsequent operations, H1Concretely, the result obtained by normalizing the calculated vertex curvature, H1∈[-1,1]，H2Concretely, the method is a result obtained by normalizing the distance from the top point of the calculated digital dentition model to the bottom of the dentition, H2∈[0,1]。
In other embodiments, the vertex height function of the digitized dentition model may also be defined as H ═ α (-H)1)+(-H2) And alpha represents H1And H2Relative weight therebetween.
Other types of vertex height functions may be used in a particular application, and therefore the present example is not particularly limited with respect to the form of definition of the vertex height function.
S201: and searching a local minimum point of the vertex height function from the digital dentition model.
In step S201, a local minimum point Q of the vertex height function is foundiI.e. the height function value is less than the height function value of the vertex of a ring of neighborhoods, each local minimum point QiRepresenting a ring neighborhood SiThe relative center of (a). For other vertices on the mesh, the vertex falls into the corresponding one-ring neighborhood along the direction of maximum change of the height function value, and the search result is shown in fig. 4.
S202: and determining the first type characteristic point according to the local minimum point.
In step S202, the method specifically includes the following steps:
step 1) calculating the area height of the area to which each local minimum point belongs according to the height of each local minimum point and the height of the area edge of the area to which the local minimum point belongs.
Specifically, for each ring neighborhood SiDefining the height of the region, specifically equal to the function value of the height of the edge of the region and the center point Q of the interior of the regioniThe difference in the height function values;
and 2) screening a plurality of regions with the minimum region height, and taking local minimum points in the screened regions as first-class feature points.
Specifically, a ring neighborhood SiSorting according to the height of the regions, selecting n regions (n is 50-150) with the smallest height of the regions, and finally, selecting the center points Q of the n regionsiNamely the position of the first class characteristic point.
The first-class feature points are screened out through the height of the region, so that the number of the feature points participating in subsequent calculation is reduced, and the first-class feature points can be ensured to cover the global features of the digital dentition model. In one example, the first type feature points are screened as shown in FIG. 5.
As can be seen, step S202 is specifically based on QiThe first-class characteristic points are determined, local minimum points, namely points which are farthest away from the bottom surface of the dentition and are more convex and points with larger tooth height, are found through the height function, the first-class characteristic points are obtained according to the local minimum points, and the local minimum points can be found in a large range on the whole model, so that the obtaining range of the first-class characteristic points can be selected from the whole digital dentition model, and the whole characteristics of the digital dentition model can be reflected better.
The above provides a specific implementation manner of automatically identifying the first class feature point in step S100, and a technical solution obtained by a person skilled in the art through simple modification of the above implementation manner should also be within the scope of the present invention.
In step S101, the second-class feature points in the digitized dentition model are classified according to the first-class feature points, and the tooth to which each second-class feature point belongs is determined. The first type of feature points refer to the feature points automatically identified and acquired in step S100, and the second type of feature points refer to vertices of each triangular patch on the digital dentition model, that is, the second type of feature points are all vertices on the digital dentition model and include the first type of feature points, or the second type of feature points refer to vertices of each triangular patch mesh on the digital dentition model except the first type of feature points. Therefore, the number of feature points of the second type may be larger than the number of feature points of the first type.
In step S101, specifically, the edge weights on the digitized dentition model are set, and the first-class feature points are used to classify the second-class feature points, which specifically includes the following steps:
(1) edge weights are set.
When the edge weight is set, the triangular mesh of the digital dentition model is regarded as a graph structure, and the edge weight w of the graph is setij：wij＝||vi-vj||·exp(-dij)；
wherein ,vi and vjRespectively representing the coordinates between vertex i and vertex j, dij＝η·||N(vi)-N(vj)||2，N(vi) and N(vj) Representing the normal vectors of the vertex i and the vertex j, further setting the value of eta, if the edge eijIf the edge is a concave edge, η is 1, if the edge e isijWhen the flange is convex, η is 0.2.
(2) The second class of feature points are classified according to the first class of feature points by using a clustering algorithm or a graph cut algorithm or a random walk algorithm, and the embodiment provides the following specific implementation mode.
Classifying the second class of feature points on the digital tooth model according to the first class of feature points by using a clustering algorithm, wherein the process of clustering classification is as follows:
1) and calculating the shortest path from all the second-class characteristic points to each first-class characteristic point.
Wherein, Dijastra algorithm (Dijastra algorithm) can be adopted to calculate the shortest path, and the edge e on the graph in the calculation processijIs the edge weight w set as aboveijThe Dijastra algorithm is a shortest path algorithm from one vertex to other vertexes, and solves the shortest path problem in the weighted graph. The Dijkstra algorithm is mainly characterized in that the Dijkstra algorithm expands outwards layer by taking a starting point as a center until the expansion reaches a terminal point. In other embodiments, the a-x algorithm may also be used to calculate the shortest path from all the feature points of the second class to each feature point of the first class.
Dijastra algorithm and A algorithm are basic methods for calculating the shortest path, and are well known to those skilled in the art, and are not described in detail in this embodiment.
2) And judging whether the shortest path from the second type feature point to the first type feature point is smaller than the shortest paths to other first type feature points, if so, determining that the second type feature point belongs to the tooth where the first type feature point is located.
Classifying the second class of feature points on the digital dentition model according to the first class of feature points by using a graph cut algorithm, wherein the classification process comprises the following steps:
1) setting a corresponding segmentation energy function for each first-class feature point: e (l) ═ r (l) + b (l);
wherein E (L) is the energy of segmentation;
r (L) is the probability that the second type feature point belongs to the first type feature point region, and R (L) ═ Σ Ri，Ri＝-ln(Pr(i | first-class feature point region)), Pr(i | first class feature point region) ═ e-L1，L1The shortest path from the second type feature point to the first type feature point is formed;
b (L) is the length of the segmentation boundary,
2) and if the segmentation energy function is minimized, the second class of feature points with the minimum segmentation energy are assigned to the teeth where the first class of feature points corresponding to the segmentation energy function are located.
Classifying the second class of feature points on the digital tooth model according to the first class of feature points by using a random walk algorithm, wherein the classification process comprises the following steps:
1) setting the phi value of the first class characteristic point to be 1;
2) setting the phi value of the second class of characteristic points as follows:
wherein ,wijIs the edge weight between vertex i and vertex j, N1(i) Representing a ring neighborhood vertex set of the second type feature point i;
3) and solving other vertex phi values, which is equivalent to solving a linear equation system, and attributing all the second type feature points with phi being more than 0 to the teeth where the corresponding first type feature points are located.
The three classification methods can be selected according to actual situations in practical application, and are not limited herein.
In step S102, the second type feature points belonging to each tooth are respectively merged to obtain a digitized tooth region of each individual tooth after the digitized tooth model is segmented, specifically including the steps of:
1) defining the region merging energy: wij＝∑wk·lk/∑lk；
wherein ,WijIs the combined energy of region i and region j, wk and lkThe edge weight and the edge length at which region i and region j intersect.
2) And calculating the merging energy of all adjacent regions, selecting the adjacent region with the minimum merging energy to merge, and taking the region with the minimum merging energy value as the segmented digital tooth model of the single tooth.
Specifically, the merging energy of all adjacent regions is updated, and then the above steps S101 and S102 are iterated until the merged energy value reaches the local minimum, at this time, the region with the local minimum merged energy value is the region of the divided single tooth, the final number of the regions is the number of teeth, and a schematic diagram of the divided single tooth is shown in fig. 6.
It can be seen that, in the embodiment, when the dentition is segmented, the single tooth is segmented by adopting the mode of selecting the feature points from the whole digital dentition model and then respectively determining the tooth to which the feature points belong.
The steps of the above methods are divided for clarity, and the implementation may be combined into one step or split some steps into multiple steps, which are within the scope of the present invention as long as the same logical relationship is included; it is within the scope of the patent to add insignificant modifications to the algorithms or processes or to introduce insignificant design changes to the core design without changing the algorithms or processes.
Example two:
the present embodiment provides a tooth segmentation method for a digital dental model, and a flowchart thereof is shown in fig. 7, which specifically includes the following steps.
S700: and determining the occlusion plane of the digital dental model.
Wherein, digital dental model is triangle patch model, and the digital dental model can be a digital dental model obtained by intraoral scanning or by scanning dental plaster model.
The specific implementation manner of determining the occlusion plane in step S700 is:
calculating the position of the center of gravity of the digital dental model
Performing principal component analysis by using all vertexes of the digital dental model, and taking the feature vector of the minimum principal component as the normal vector of the occlusion plane
Acquiring an occlusion plane according to an occlusion plane equation:
wherein ,
point coordinates on the occlusal plane.
S701: and extracting characteristic areas of dentition and gum boundary in the digital dental model.
In step S701, the specific implementation steps of acquiring the feature region are:
s7011: vertex curvature values for all vertices of the digitized dental model are calculated.
Specifically, the maximum principal curvature value of each grid vertex is calculated by using a local cubic surface fitting method, and the maximum principal curvature values of all the grid vertices are normalized to [0,1 ];
averagely dividing the normalized curvature value set into L intervals
Calculating the cumulative distribution probability density of the curvature of each section:
wherein ,
representing the curvature value kiSatisfy the requirement of
The number of the grid vertexes L is 0,1,2, 1, L-1;
and transforming the vertex curvature value of each grid vertex to obtain the curvature value of each transformed grid vertex, wherein the transformation formula is as follows:
n is the total number of mesh vertices.
S7012: and extracting an initial characteristic region of the digital dental model according to the vertex curvature value.
In particular, according to formula F0＝{pi|k(pi) H, (i ═ 1, 2.., N) } extracting an initial feature region, wherein F is0As the initial feature region, k (p)i) For mesh vertex piThe curvature value after transformation, H is the curvature threshold.
S7013: and mapping the initial characteristic region into an undirected connected graph according to the connection relation between the grid vertexes.
Specifically, the initial feature region F0Mapping to undirected connectivity graph G according to connection relation between grid vertexes0，G0Middle edge weight edge (v)i,vj) Defined as follows:
wherein E is the set of all triangular patches of the three-dimensional dental model.
S7014: and calculating the connected components of the undirected connected graph.
Specifically, the connected components of the undirected connected graph G0 are calculated, the number of mesh vertexes in each connected component is counted, the connected components with the smaller number of mesh vertexes, i.e., the noise region, is deleted, and only the connected components with the largest number of mesh vertexes are reserved, i.e., the characteristic region F of the boundary between the dentition, the gum and the adjacent teeth.
S7015: and counting the number of the network vertexes in each communication component, wherein the communication component with the most network vertexes is a characteristic region of dentition and gum boundary.
S702: and projecting the characteristic region to an occlusion plane to construct a characteristic binary image, calculating the outer contour of the characteristic binary image and mapping the outer contour to the characteristic region to obtain a gum line.
The specific implementation of step S702 is as follows:
projecting the characteristic region onto an occlusion plane, and constructing a characteristic binary image;
filling holes in the characteristic binary image;
calculating the outer contour of the feature binary image after the hole is filled;
and mapping the outer contour into the characteristic area to obtain a gum line.
S703: the digitized dental model is segmented into dentition regions and gum regions based on a gum line.
Step S703 specifically includes the following steps:
1) randomly selecting a network vertex on the digital dental model as a diffusion seed vertex;
2) diffusing outwards by taking the peak of the seed as a center by utilizing a breadth-first search algorithm until the peak of the grid in the gum line is positioned;
3) the diffused area and the non-diffused area constitute the dentition area and the gum area, respectively, and a schematic diagram of the division of the dentition area and the gum area is shown in fig. 8.
A specific example of implementing the step S703 is as follows:
step 1: and selecting a vertex which is used for representing the tooth point on the dental jaw model, marking the vertex and adding the vertex to the sequence to be diffused.
Step 2: selecting a vertex in the sequence to be diffused, judging each vertex in a ring of neighborhood vertex set of the vertex, if the vertex in a ring of neighborhood is judged to be an undispersed vertex and a non-gum line vertex, marking the undispersed vertex as a tooth point, and adding the tooth point into the sequence to be diffused.
Step 2 is a diffusion process, and the non-diffused vertices labeled as tooth points are continuously added to the sequence to be diffused to wait for diffusion.
And step 3: and traversing and searching the vertexes in the sequence to be diffused, and repeating the step 2 to diffuse the searched vertexes until the sequence to be diffused is empty.
S704: and (4) segmenting the tooth area by adopting a dentition segmentation method to obtain the digital area of each single tooth.
In step S704, the adopted dentition segmentation method is specifically the dentition segmentation method in the first embodiment, and for the specific segmentation process, reference is made to the first embodiment, which is not described in detail in this embodiment.
After the tooth area is obtained, the tooth area is segmented by adopting the dentition segmentation method of the first embodiment, the first class of feature points are selected on the whole digital dentition model, then the second class of feature points on the digital dentition model are classified and collected according to the first class of feature points, and the segmentation of the single tooth is realized.
Further, after obtaining the divided dentition area and gum area, the method further comprises the step of smoothing the boundary between the dentition area and the gum area, wherein the smooth dentition area and gum area boundary is realized by the following steps:
1) obtaining a smooth energy function: e ═ E1+E2；
wherein ,E1Is Σ | φi-φi0|2Representing the smooth posterior gingival partDeviation of the cut region from the smooth anterior tooth gingival cut region, whereini0Represents the tooth and gum segmentation result before smoothing, and the phi is the tooth vertexi01, said phi for the gum apexi01, phi asiRepresenting the tooth and gum segmentation result after smoothing;
E2is composed of
Represents a smoothing energy term, wherein N is1(i) A ring neighborhood vertex set representing vertex i, said n representing a ring neighborhood vertex number;
2) minimizing a smooth energy function to obtain a phi value after each vertex is smooth;
wherein, all areas formed by the vertexes phi & gt 0 are tooth model areas after being smoothed, and all areas formed by the vertexes phi & lt 0 are gum model areas after being smoothed.
It should be noted that, the steps S700-S703 are a specific implementation scheme for achieving the separation of the dentition region and the gingival region, and those skilled in the art can also use other schemes to achieve the separation of the dentition region and the gingival region in the digital dental model, for example, those skilled in the art can also apply a deep artificial neural network to the present example to achieve the separation of the dentition region and the gingival region.
Example three:
based on the first embodiment, this example provides a dentition segmentation apparatus for a digitized dentition model, a block diagram of which is shown in fig. 9, and the dentition segmentation apparatus includes:
a selecting module 901, configured to select a first class of feature points on a digital dentition model to be segmented, where the digital dentition model is a triangular patch model, and the first class of feature points are triangular patch vertices that are selected based on the digital dentition model and are used for guiding segmentation of each single tooth in a dentition;
the classification module 902 is configured to classify second-class feature points in the digitized dentition model according to the first-class feature points, and determine teeth to which the second-class feature points belong, where the second-class feature points are vertices of a triangular patch that is selected based on the digitized dentition model and is used for representing the overall shape of the digitized dentition model;
and a merging module 903, configured to merge the second type of feature points belonging to each tooth, respectively, to obtain a digitized tooth region of each individual tooth after the digitized dentition model is segmented.
For the specific process of automatically identifying and selecting the first type of feature points, please refer to step S100 in the first embodiment, which is not repeated in this embodiment.
The classifying module 902 classifies the second class of feature points according to the first class of feature points, and determines the teeth to which each second class of feature point belongs, and for a specific classifying process, reference is made to step S101 in the first embodiment, which is not repeated herein, and similarly, reference is also made to the first embodiment for a description of definitions of the first class of feature points and the second class of feature points.
The merging module 903 merges the second type of feature points on each tooth to obtain a digitized region of each individual tooth, please refer to step S102 in the first embodiment.
With reference to the dentition segmentation method of the first embodiment, the dentition segmentation apparatus of the present embodiment implements segmentation of a single tooth by selecting a first type of feature points on the whole dentition model, and classifying and re-aggregating a second type of feature points on the model according to the first type of feature points.
Example four:
according to a first embodiment, this embodiment provides an electronic device, a block diagram of which is shown in fig. 10, and the electronic device 100 may be a tablet computer, a notebook computer, or a desktop computer. The electronic device 100 may also be referred to by other names, such as portable terminal, laptop terminal, desktop terminal, and the like.
Generally, the electronic device 100 includes a processor 1001 and a memory 1002, and the processor 1001 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 1001 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 1001 may also include a main processor and a coprocessor, where the main processor is a processor for Processing data in an awake state, and is also referred to as a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state.
In some embodiments, the processor 1001 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 1001 may further include an AI (Artificial Intelligence) processor for processing a computing operation related to machine learning.
Therefore, the electronic device 100 of the present application, which executes the segmentation method of the digitized dentition model provided in the first embodiment through at least one instruction, at least one program, a code set, or an instruction set, has the following advantages:
the first type of feature points are selected on the whole digital dentition model, then the second type of feature points on the model are classified and collected according to the first type of feature points, and the segmentation of the single tooth is realized.
In some embodiments, the electronic device 100 may further include: a peripheral interface 1003 and at least one peripheral. The processor 1001, memory 1002 and peripheral interface 1003 may be connected by a bus or signal line. Various peripheral devices may be connected to peripheral interface 1003 via a bus, signal line, or circuit board. Specifically, the peripheral device includes: at least one of radio frequency circuitry 1004, touch screen display 1005, camera 1006, audio circuitry 1007, positioning components 1008, and power supply 1009.
Example five:
based on the second embodiment, this example provides a tooth segmentation apparatus for a digital dental model, a structural block diagram of which is shown in fig. 11, the tooth segmentation apparatus includes:
the determining module 1101 is used for determining an occlusion plane of the digital dental model, wherein the digital dental model is a triangular patch model;
the extraction module 1102 is used for extracting characteristic areas of dentition and gum boundaries in the digital dental model;
an obtaining module 1103, configured to project the feature region onto the occlusion plane to construct a feature binary image, calculate an outer contour of the feature binary image, and map the outer contour into the feature region to obtain a gum line;
a first segmentation module 1104 for segmenting the digitized dental model into dentition regions and gum regions based on a gum line;
and a second segmentation module 1105, configured to segment the tooth region by using a dentition segmentation method to obtain each individual tooth.
The specific way of determining the occlusal plane by the determining module 1101 is referred to step S700 of the second embodiment, the specific way of extracting the characteristic region of the boundary between the dentition and the gum by the extracting module 1102 is referred to step S701 of the second embodiment, the specific way of acquiring the gum line by the acquiring module 1103 is referred to step S702 of the second embodiment, the specific way of dividing the dentition region and the gum region by the first dividing module 1104 is referred to step S703 of the second embodiment, and the tooth region is divided by the second dividing module 1105 by the dentition dividing method of the first embodiment to acquire each single tooth, so the specific way of dividing the tooth region by the second dividing module 1105 is referred to the dentition dividing method of the first embodiment.
The tooth segmentation device of the present example segments the tooth region by combining the dentition segmentation method of the first example, and therefore the tooth region segmentation realized by the tooth segmentation device of the present example has the following advantages:
the first type of feature points are selected on the digital dentition model, then the second type of feature points on the digital dentition model are classified and collected according to the first type of feature points, and the single tooth is segmented.
Example six:
according to a second embodiment, this embodiment provides an electronic device, a block diagram of which is shown in fig. 12, and the electronic device 200 may be a tablet computer, a notebook computer, or a desktop computer. The electronic device 200 may also be referred to by other names, such as portable terminal, laptop terminal, desktop terminal, and the like.
Generally, the electronic device 200 includes a processor 2001 and a memory 2002, and the processor 2001 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 2001 may be implemented in at least one hardware form of a DSP (Digital Signal Processing), an FPGA (Field-Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 2001 may also include a main processor and a coprocessor, the main processor being a processor for Processing data in an awake state, also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state.
In some embodiments, the processor 2001 may be integrated with a GPU (Graphics Processing Unit) that is responsible for rendering and drawing the content that the display screen needs to display. In some embodiments, the processor 2001 may further include an AI (Artificial Intelligence) processor for processing computing operations related to machine learning.
The memory 2002 may include one or more computer-readable storage media, which may be non-transitory. The memory 2002 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in the memory 2002 is used to store at least one instruction, at least one program, set of codes, or set of instructions for execution by the processor 2001 to implement the tooth segmentation method provided by embodiment two herein.
Therefore, the electronic device 200 of the present application performs the tooth segmentation method provided in the second embodiment by at least one instruction, at least one program, a code set, or an instruction set, wherein the segmentation for the tooth region has the following advantages:
the first type of feature points are selected on the whole digital dentition model, then the second type of feature points on the digital dentition model are classified and collected according to the first type of feature points, and the segmentation of the single tooth is realized.
In some embodiments, the electronic device 200 may further include: a peripheral interface 2003 and at least one peripheral. The processor 2001, memory 2002 and peripheral interface 2003 may be connected by buses or signal lines. Various peripheral devices may be connected to peripheral interface 2003 through a bus, signal line, or circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 2004, a touch display 2005, a camera 2006, an audio circuit 2007, a positioning assembly 2008, and a power supply 2009.
The present invention has been described in terms of specific examples, which are provided to aid understanding of the invention and are not intended to be limiting. For a person skilled in the art to which the invention pertains, several simple deductions, modifications or substitutions may be made according to the idea of the invention.
Claims (26)
1. A dentition segmentation method of a digital dentition model is characterized by comprising the following steps:
selecting a first class of characteristic points on a digital dentition model to be segmented, wherein the digital dentition model is a triangular patch model, and the first class of characteristic points are triangular patch vertexes which are selected based on the digital dentition model and are used for guiding the segmentation of each single tooth in the dentition;
classifying second feature points in the digital dentition model according to the first feature points to determine teeth to which the second feature points belong, wherein the second feature points are selected based on the digital dentition model and are used for representing the vertexes of a triangular patch of the overall shape of the digital dentition model;
and respectively merging the second characteristic points belonging to each tooth to obtain the digital tooth area of each single tooth after the digital dentition model is segmented.
2. The dentition segmentation method as claimed in claim 1 wherein the step of selecting a first class of feature points on the digitized dentition model to be segmented comprises:
defining a vertex height function of a triangular patch in the digital dentition model;
searching a local minimum point of the vertex height function from the digital dentition model;
and determining the first type characteristic point according to the local minimum point.
3. The dentition segmentation method according to claim 2 wherein the step of defining the vertex height function of the triangular patch in the digitized dentition model comprises the steps of: and defining the vertex height function according to the vertex curvature of a triangular patch in the digital dentition model and the distance from the vertex to the bottom of the dental jaw.
4. The dentition segmentation method of claim 3 wherein the specific steps of defining the vertex height function are:
calculating the apex curvature H of the digitized dentition model1And normalized to [ -1,1]；
Respectively calculating the distance H from each vertex of the digital dentition model to the bottom of the dental jaw2And normalized to [0, 1]]；
The vertex height function is defined as H ═ H (-H)1)+(-H2) Where H is a vertex height function.
5. The dentition segmentation method according to claim 2 wherein the determining the first class of feature points from the local minimum points is:
calculating the height of the region to which each local minimum point belongs according to the height of each local minimum point and the height of the region edge of the region to which the local minimum point belongs;
and screening a plurality of areas with the minimum area height, and taking the local minimum point in the screened areas as the first-class characteristic point.
6. The dentition segmentation method according to claim 1 wherein the second class of feature points on the digitized dentition model are classified according to the first class of feature points using a clustering algorithm or a graph cut algorithm or a random walk algorithm.
7. The dentition segmentation method according to claim 6 wherein the classifying the second class of feature points in the digitized dentition model according to the first class of feature points determines the tooth to which each of the second class of feature points belongs, and specifically comprises:
classifying the second class of feature points on the digital dentition model according to the first class of feature points by using a clustering algorithm, and the method specifically comprises the following steps:
calculating the shortest path from all the second-class characteristic points to each first-class characteristic point;
and judging whether the shortest path from the second class of feature points to the first class of feature points is smaller than the shortest paths to other first class of feature points, if so, determining that the second class of feature points belongs to the teeth where the first class of feature points are located.
8. The dentition segmentation method according to claim 6 wherein the classifying the second class of feature points in the digitized dentition model according to the first class of feature points determines the tooth to which each of the second class of feature points belongs, and specifically comprises:
classifying second-class feature points on the digital dentition model according to the first-class feature points by using a graph cut algorithm, wherein the method specifically comprises the following steps:
setting a corresponding segmentation energy function for each first-class feature point: e (l) ═ r (l) + b (l);
wherein E (L) is the energy of segmentation;
r (L) is the probability that the second type feature point belongs to the first type feature point region, and R (L) ═ Σ Ri，Ri＝-ln(Pr(i | first-class feature point region)), Pr(i | first class feature point region) ═ e-L1L1 is the shortest path from the second class of feature point to the first class of feature point;
b (L) is the length of the segmentation boundary,
and if the segmentation energy function is minimized, the second class of feature points with the minimum segmentation energy are attributed to the teeth where the first class of feature points corresponding to the segmentation energy function are located.
9. The dentition segmentation method according to claim 6 wherein the classifying the second class of feature points in the digitized dentition model according to the first class of feature points determines the tooth to which each of the second class of feature points belongs, and specifically comprises:
classifying the second class of feature points on the digital dentition model according to the first class of feature points by using a random walk algorithm, and the specific steps are as follows:
setting the phi value of the first class characteristic point to be 1;
setting the phi value of the second class of characteristic points as follows:
wherein ,wijIs the edge weight between vertex i and vertex j, N1(i) Representing a ring neighborhood vertex set of the second type feature point i;
all the second class feature points with phi > 0 are assigned to the teeth where the corresponding first class feature points are located.
10. The dentition segmentation method according to claim 1, wherein the step of respectively merging the second feature points belonging to each tooth to obtain the digitized tooth area of each individual tooth after the segmentation of the digitized tooth model comprises the steps of:
defining the region merging energy: wij＝∑wk·lk/∑lk；
wherein ,WijIs the combined energy of region i and region j, wk and lkThe edge weight and the edge length of the intersection of the area i and the area j are obtained;
and calculating the merging energy of all adjacent regions, selecting the adjacent region with the minimum merging energy to merge, and taking the region with the minimum merging energy value as the digital tooth region of the segmented single tooth.
11. The dentition segmentation method of claim 1 wherein the second type of feature points are vertices of triangular patches on the digitized dentition model or vertices of triangular patches on the digitized dentition model other than the first type of feature points.
12. A tooth segmentation method of a digital dental model is characterized by comprising the following steps:
determining an occlusion plane of a digital dental model, wherein the digital dental model is a triangular patch model;
extracting characteristic areas of dentition and gum boundary in the digital dental model;
projecting the characteristic region to the occlusion plane to construct a characteristic binary image, calculating the outer contour of the characteristic binary image and mapping the outer contour to the characteristic region to obtain a gum line;
segmenting the digitized dental model into dentition regions and gum regions based on a gum line;
the tooth area is segmented by the dentition segmentation method according to any one of claims 1-11 to obtain digitized areas of individual teeth.
13. The tooth segmentation method according to claim 12, wherein the determining of the occlusal plane of the digitized dental model comprises the steps of:
calculating the position of the center of gravity of the digital dental model
Performing principal component analysis by using all vertexes of the digital dental modelThe feature vector of the minimum principal component is used as the normal vector of the occlusion plane
Acquiring an occlusion plane according to an occlusion plane equation, wherein the occlusion plane equation is as follows:
wherein ,
point coordinates on the occlusal plane.
14. The tooth segmentation method according to claim 12, wherein the step of extracting the characteristic region of the dentition and the gum boundary in the digital dental model comprises the steps of:
calculating the vertex curvature values of all the vertexes of the digital dental model;
extracting an initial characteristic region of the digital dental model according to the vertex curvature value;
mapping the initial characteristic region into an undirected connected graph according to the connection relation between the vertexes of the triangular patch;
calculating connected components of the undirected connected graph;
and counting the number of the network vertexes in each communication component, wherein the communication component with the most number of the vertexes of the triangular patch is a characteristic region of dentition and gum boundary.
15. The tooth segmentation method according to claim 14, wherein the step of calculating the vertex curvature values of all the vertices of the digitized dental model comprises the steps of:
calculating the maximum principal curvature value of each grid vertex by using a local cubic surface fitting method, and normalizing the maximum principal curvature values of all the triangular patch vertices to [0,1 ];
averagely dividing the normalized curvature value set into L intervals
Calculating the cumulative distribution probability density of the curvature of each section:
wherein ,
representing the curvature value kiSatisfy the requirement of
The number of the grid vertexes L is 0,1,2, 1, L-1;
and transforming the vertex curvature value of each triangular patch vertex to obtain the curvature value of each triangular patch vertex after transformation, wherein the transformation formula is as follows:
and N is the total number of the vertexes of the triangular patch.
16. The tooth segmentation method according to claim 14, wherein the extracting of the initial feature region of the digitized dental model according to the vertex curvature value comprises the steps of:
according to formula F0＝{pi|k(pi) H, (i ═ 1, 2.., N) } extracting an initial feature region, wherein F is0As the initial feature region, k (p)i) For mesh vertex piThe curvature value after transformation, H is the curvature threshold.
17. The tooth segmentation method according to claim 12, wherein the step of projecting the characteristic region to the occlusion plane to construct a characteristic binary image, calculating an outer contour of the characteristic binary image and mapping the outer contour to the characteristic region to obtain a gum line comprises the steps of:
projecting the characteristic region onto the occlusion plane and constructing the characteristic binary image;
filling holes in the characteristic binary image;
calculating the outer contour of the feature binary image after the hole is filled;
and mapping the outer contour into the characteristic area to obtain a gum line.
18. The tooth segmentation method according to claim 12, wherein the step of segmenting the digitized jaw model into dentition regions and gum regions based on the gum line comprises the steps of:
randomly selecting a vertex of a triangular patch on the digital dental model as a diffusion seed vertex;
diffusing outwards by taking the peak of the seed as a center by using a breadth-first search algorithm until the peak position of the triangular surface sheet in the gum line is reached;
the diffused area and the non-diffused area constitute the dentition area and the gum area, respectively.
19. The tooth segmentation method according to claim 18, wherein the diffusion region is formed by the steps of:
selecting a vertex representing a tooth point on the digital dental model, marking the vertex and adding the vertex into a sequence to be diffused;
selecting a vertex in the sequence to be diffused, judging each vertex in a ring of neighborhood vertex set of the vertex, if the vertex in a ring of neighborhood is judged to be an undispersed vertex and a non-gum line vertex, marking the undispersed vertex as a tooth point, and adding the tooth point into the sequence to be diffused;
traversing and searching the vertexes in the sequence to be diffused, and diffusing the searched vertexes until the sequence to be diffused is empty.
20. The tooth segmentation method according to claim 12, wherein the digitized dental model is obtained by intraoral scanning or by scanning a dental plaster model.
21. The tooth segmentation method according to claim 12, further comprising a step of smoothing boundaries of the dentition region and the gum region after obtaining the segmented dentition region and gum region.
22. The tooth segmentation method according to claim 21, wherein the step of smoothing the boundary between the dentition region and the gum region comprises the steps of:
obtaining a smooth energy function: e ═ E1+E2；
wherein ,E1Is Σ | φi-φi0|2Representing the deviation of the gum segmentation area of the teeth after smoothing from the gum segmentation area of the teeth before smoothing, wherein phi isi0Represents the tooth and gum segmentation result before smoothing, and the phi is the tooth vertexi01, said phi for the gum apexi01, phi asiRepresenting the tooth and gum segmentation result after smoothing;
E2is composed of
Represents a smoothing energy term, wherein N is1(i) A ring neighborhood vertex set representing vertex i, said n representing a ring neighborhood vertex number;
minimizing the smooth energy function to obtain a phi value after each vertex is smooth;
wherein, all areas formed by the vertexes phi & gt 0 are tooth model areas after being smoothed, and all areas formed by the vertexes phi & lt 0 are gum model areas after being smoothed.
23. An dentition segmentation apparatus for digitizing a dentition model, comprising:
the system comprises a selection module, a segmentation module and a segmentation module, wherein the selection module is used for selecting a first class of characteristic points on a digital dentition model to be segmented, the digital dentition model is a triangular patch model, and the first class of characteristic points are triangular patch vertexes which are selected based on the digital dentition model and are used for guiding the segmentation of each single tooth in the dentition;
the classification module is used for classifying second feature points in the digital dentition model according to the first feature points and determining teeth to which the second feature points belong, wherein the second feature points are selected based on the digital dentition model and are used for representing the vertexes of a triangular patch of the overall shape of the digital dentition model;
and the merging module is used for respectively merging the second type of characteristic points belonging to each tooth to obtain the digital tooth area of each single tooth after the digital dentition model is segmented.
24. An electronic device comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by the processor to implement the dentition segmentation method of any of claims 1-11.
25. A tooth segmentation device for a digital dental model, comprising:
the determination module is used for determining an occlusion plane of the digital dental model, and the digital dental model is a triangular patch model;
the extraction module is used for extracting characteristic areas of dentition and gum boundary in the digital dental model;
the obtaining module is used for projecting the characteristic region to the occlusion plane to construct a characteristic binary image, calculating the outer contour of the characteristic binary image and mapping the outer contour to the characteristic region to obtain a gum line;
a first segmentation module for segmenting the digitized dental model into dentition regions and gum regions based on a gum line;
a second segmentation module for segmenting the tooth region by using the dentition segmentation method according to any one of claims 1 to 11 to obtain a digitized region of each individual tooth.
26. An electronic device comprising a processor and a memory, the memory having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by the processor to implement a tooth segmentation method according to any one of claims 12 to 22.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910702794.9A CN112396609B (en) | 2019-07-31 | 2019-07-31 | Dentition segmentation method, dentition segmentation device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910702794.9A CN112396609B (en) | 2019-07-31 | 2019-07-31 | Dentition segmentation method, dentition segmentation device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112396609A true CN112396609A (en) | 2021-02-23 |
| CN112396609B CN112396609B (en) | 2023-08-08 |
Family
ID=74601272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910702794.9A Active CN112396609B (en) | 2019-07-31 | 2019-07-31 | Dentition segmentation method, dentition segmentation device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112396609B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114638844A (en) * | 2022-03-18 | 2022-06-17 | 福州海狸家口腔科技有限公司 | Dental model generation method, storage medium and electronic equipment |
| CN116385474A (en) * | 2023-02-27 | 2023-07-04 | 雅客智慧（北京）科技有限公司 | Tooth scanning model segmentation method and device based on deep learning and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN105678747A (en) * | 2015-12-30 | 2016-06-15 | 青岛兰信医学科技有限公司 | Tooth mesh model automatic segmentation method based on principal curvature |
| CN106228549A (en) * | 2016-07-14 | 2016-12-14 | 嘉兴学院 | A kind of triangle gridding tooth dividing method based on path planning |
| WO2017020580A1 (en) * | 2015-08-03 | 2017-02-09 | 青岛海信医疗设备股份有限公司 | Medical tissue slice image 3d reconstruction method and apparatus |
| CN108364356A (en) * | 2018-02-28 | 2018-08-03 | 辽宁科技大学 | A kind of automatic division method of tooth three-dimensional grid model |
| CN108648283A (en) * | 2018-04-02 | 2018-10-12 | 北京正齐口腔医疗技术有限公司 | Tooth dividing method and device |
-
2019
- 2019-07-31 CN CN201910702794.9A patent/CN112396609B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| WO2017020580A1 (en) * | 2015-08-03 | 2017-02-09 | 青岛海信医疗设备股份有限公司 | Medical tissue slice image 3d reconstruction method and apparatus |
| 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 |
| CN105678747A (en) * | 2015-12-30 | 2016-06-15 | 青岛兰信医学科技有限公司 | Tooth mesh model automatic segmentation method based on principal curvature |
| CN106228549A (en) * | 2016-07-14 | 2016-12-14 | 嘉兴学院 | A kind of triangle gridding tooth dividing method based on path planning |
| CN108364356A (en) * | 2018-02-28 | 2018-08-03 | 辽宁科技大学 | A kind of automatic division method of tooth three-dimensional grid model |
| CN108648283A (en) * | 2018-04-02 | 2018-10-12 | 北京正齐口腔医疗技术有限公司 | Tooth dividing method and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114638844A (en) * | 2022-03-18 | 2022-06-17 | 福州海狸家口腔科技有限公司 | Dental model generation method, storage medium and electronic equipment |
| CN116385474A (en) * | 2023-02-27 | 2023-07-04 | 雅客智慧（北京）科技有限公司 | Tooth scanning model segmentation method and device based on deep learning and electronic equipment |
| CN116385474B (en) * | 2023-02-27 | 2024-06-04 | 雅客智慧（北京）科技有限公司 | Tooth scanning model segmentation method and device based on deep learning and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112396609B (en) | 2023-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11735306B2 (en) | Method, system and computer readable storage media for creating three-dimensional dental restorations from two dimensional sketches | |
| CN106228549B (en) | A kind of triangle gridding tooth dividing method based on path planning | |
| US20200402647A1 (en) | Dental image processing protocol for dental aligners | |
| RU2725280C1 (en) | Devices and methods for orthodontic treatment planning | |
| JP2023526134A (en) | Generation and Placement of Neural Network-Based Dental Appliances for Tooth Restoration | |
| Liao et al. | Automatic tooth segmentation of dental mesh based on harmonic fields | |
| CN112790879B (en) | Tooth axis coordinate system construction method and system of tooth model | |
| CN112396609B (en) | Dentition segmentation method, dentition segmentation device and electronic equipment | |
| CN112308861B (en) | Tooth and gum segmentation method, tooth segmentation method and electronic equipment | |
| Jung et al. | Combining volumetric dental CT and optical scan data for teeth modeling | |
| KR102250520B1 (en) | Method for recommending crown model and prosthetic CAD apparatus therefor | |
| US20220361992A1 (en) | System and Method for Predicting a Crown and Implant Feature for Dental Implant Planning | |
| WO2022193909A1 (en) | Method for removing accessory on dental three-dimensional digital model | |
| Woodsend et al. | Automatic recognition of landmarks on digital dental models | |
| CN108242056B (en) | Three-dimensional tooth grid data segmentation method based on harmonic field algorithm | |
| Ben-Hamadou et al. | 3DTeethSeg'22: 3D Teeth Scan Segmentation and Labeling Challenge | |
| TW202409874A (en) | Dental restoration automation | |
| CN113139908B (en) | Three-dimensional dentition segmentation and labeling method | |
| Mouritsen | Automatic segmentation of teeth in digital dental models | |
| US20230419631A1 (en) | Guided Implant Surgery Planning System and Method | |
| CN112932697B (en) | Hard palate area identification method and system in dental model | |
| CN115620053B (en) | Airway type determining system and electronic equipment | |
| US20230298272A1 (en) | System and Method for an Automated Surgical Guide Design (SGD) | |
| CN116229013A (en) | Dental crown generation method, device, equipment and medium | |
| US20240029380A1 (en) | Integrated Dental Restoration Design Process and System |
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 |
Address after: Room 122, 2305 Zuchongzhi Road, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai 201210 Applicant after: Zhengya Dental Technology (Shanghai) Co.,Ltd. Address before: Room 122, 2305 Zuchongzhi Road, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai 201210 Applicant before: SHANGHAI SMARTEE DENTI-TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |