CN110135397B

CN110135397B - Method for identifying dental crown part in three-dimensional tooth model

Info

Publication number: CN110135397B
Application number: CN201910447561.9A
Authority: CN
Inventors: 魏伟
Original assignee: Shanghai Emendi Material Technology Co ltd
Current assignee: Shanghai Emendi Material Technology Co ltd
Priority date: 2019-05-27
Filing date: 2019-05-27
Publication date: 2022-07-08
Anticipated expiration: 2039-05-27
Also published as: CN110135397A

Abstract

The invention discloses a method for identifying a dental crown part in a three-dimensional tooth model, which comprises the following steps: traversing each vertex of the dental model, and calculating and determining a first concave point set S; determining the number m of dental crowns to be identified; sequentially identifying models of the dental crowns according to the arrangement sequence of the actual dental crowns, and obtaining a dental crown set T { (T1, T2, …, Tm }, T1, T2, …, wherein Tm represents each dental crown; traversing the dental crown set T, judging whether the boundary vertex sets of two adjacent dental crowns have overlapping conditions, and if not, ending; if so, the overlap is removed. The invention can effectively improve the identification efficiency and accuracy.

Description

Method for identifying dental crown part in three-dimensional tooth model

Technical Field

The invention relates to the technical field of dental crown identification.

Background

Crown identification is the manipulation of a single target crown portion delineated in a single maxillofacial surface of a patient by specialized computer software. Three-dimensional tooth model files which can be displayed in a computer are obtained by scanning the tooth model of a patient in a 3D mode, and after the operation of identifying the tooth crowns, the three-dimensional characteristic information of the tooth crowns of each tooth is obtained.

The existing widely applied crown identification technology mainly has two defects:

1) the tooth model has high quality dependence. The scanning accuracy of the tooth model is not high, so that the dents at the dividing boundaries of the dental crown and the gum and the dental crown are not obvious, and the dividing boundaries of the dental crown and the gum and the dental crown cannot be accurately identified.

2) The adaptation range is limited. The segmentation boundaries between the crowns are inaccurate due to the areas without obvious depressions at the segmentation boundaries between the crowns and the gingiva and between the crowns and the dental crowns. The dividing boundaries between the crowns do not accurately reach the recessed areas.

The above problems reduce the working efficiency of the crown recognition operation and the accuracy of the crown recognition.

Disclosure of Invention

The invention aims to provide a method for identifying a dental crown part in a three-dimensional tooth model, which effectively improves the identification efficiency and accuracy.

The technical scheme for realizing the purpose is as follows:

a method of identifying a crown portion in a three-dimensional tooth model, comprising:

traversing each vertex of the dental model, and calculating and determining a first concave point set S;

determining the number m of dental crowns to be identified;

sequentially identifying models of the dental crowns according to the arrangement sequence of the actual dental crowns, and obtaining a dental crown set T { (T1, T2, …, Tm }, T1, T2, …, wherein Tm represents each dental crown;

traversing the dental crown set T, judging whether the boundary vertex sets of two adjacent dental crowns are overlapped, and if not, ending; if so, the overlap is removed.

Preferably, the mean value a of the included angle between the vertex normal vector and the triangle plane normal vector of the neighborhood is calculated for each vertex of the dental model, and if the mean value a of the included angle is less than 90 degrees, the vertex is a concave point and is added into the first concave point set S.

Preferably, the model of each crown is identified, comprising:

determining a right division plane1 and a left division plane2 of a single crown;

traversing the first pit set S, and finding a second pit set S1 between the right split plane1 and the left split plane 2;

removing the concave points which are not used for dividing the crown and the gum and the crown in the second concave point set S1 to obtain a third concave point set S2;

dividing the third concave point set S2 into a plurality of concave point sets according to the connectivity of vertexes of the dental model to obtain a concave point block set A, and sequencing the concave point block set A according to the shape of the third concave point set S2, so that concave point blocks adjacent to two subscripts in the concave point block set A are also adjacent in the dental model, and recording that the concave point block set A is { A1, A2, …, An }, A1, A2, … and An represents each concave point block;

screening vertexes except any two adjacent concave point blocks in the dental model, adding the vertexes into the concave point block set A, enabling the two adjacent concave point blocks to be communicated, and finally enabling all the concave point blocks to form a closed area which is marked as a closed area AC;

and screening out the point set B in the closed area AC, so that the point set B keeps the shape of the closed area AC, and the point set B forms a closed ring, all vertexes in the ring and on the ring form a model, all vertexes outside the ring and on the ring form another model, and selecting the model with a small space area as the model of the identified crown in the two models.

Preferably, the screening of the point set B in the closed region AC refers to repeating the operations of: deleting the outermost and innermost points of the annular closed area AC, and if only one point exists at a certain position, deleting the point to enable the annular area not to be closed, not deleting the point;

and then, sequentially connecting all the rest points to form a closed line, namely the point set B.

Preferably, assuming that there is an overlap between the boundary vertex sets B1 and B2 of the crown T1 and T2, the removing the overlap includes:

let B11, B12 be the set of overlapping partial boundary vertices and the set of non-overlapping partial boundary vertices of boundary vertex sets B1 and B2, respectively, B21, B22 be the set of overlapping partial boundary vertices and the set of non-overlapping partial boundary vertices of boundary vertex sets B2 and B1, respectively,

recalculating B11 and B21 as B11 'and B21' so that B11 'and B21' retain the shape of B11 and B21 and B11 'and B21' do not overlap;

b11 'and B12 form a new boundary vertex set B1' of the crown T1, the operation of identifying the model of the crown is performed, the crown T1 is updated to T1 ', the B21' and B22 form a new boundary vertex set B2 'of T2, the operation of identifying the model of the crown is performed, the crown T2 is updated to T2', and the operated crown set is still T, T ═ T1 ', T2', T3, …, Tm }.

The invention has the beneficial effects that: according to the invention, through sequentially identifying each single crown and removing the overlapping, the success rate of crown identification is improved, and the probability of wrong crown identification area caused by the quality of a patient tooth model is reduced. Meanwhile, the adaptation range of the dental crown identification algorithm is expanded, the situation that the dental crown segmentation boundary cannot be accurately identified due to the fact that the sunken areas at the dental crown and gum and the dental crown and dental crown segmentation boundary in the dental model of the patient are not obvious is avoided, and the identified dental crown segmentation boundary can accurately reach the sunken areas at the dental crown and gum and the dental crown and dental crown segmentation boundary.

Drawings

Fig. 1 is a flowchart of a method of identifying a crown portion in a three-dimensional tooth model of the present invention.

Detailed Description

The invention will be further explained with reference to the drawings.

Referring to fig. 1, the method for identifying a crown part in a three-dimensional tooth model according to the present invention comprises the steps of:

step S11, traversing vertexes of the dental model, calculating an included angle mean value a of a vertex normal vector and a neighborhood triangular surface normal vector for each vertex, and if the included angle mean value a is less than 90 degrees, the point is a concave point and is added into the first concave point set S.

In step S12, the number m of crowns to be identified is determined.

In step S13, models of the respective crowns are sequentially recognized in the order of arrangement of the actual crowns, and a crown set T { T1, T2, …, Tm }, is obtained. T1, T2, …, Tm represent the respective crowns. Specifically, the method comprises the following steps:

in step S131, the right and left split planes 1, 2 of a single crown are determined.

Step S132, traverse the first pit set S, and find out a second pit set S1 between the right split plane1 and the left split plane 2.

Step S133, removing the concave points in the second concave point set S1 that do not obviously divide the crown and the gum, and the crown to obtain a third concave point set S2. These pits are typically located at the boundaries between the teeth and the gums, at the valleys on the tooth surface, and at the valleys on the gum surface.

Step S134, dividing the third concave point set S2 into a plurality of concave point sets according to the connectivity of the vertexes of the dental model to obtain a concave point block set A, and sorting the concave point block set A according to the shape of the third concave point set S2, so that two concave point blocks with adjacent subscripts in the concave point block set A are also adjacent in the dental model, and the concave point block set A is { A1, A2, …, An }, A1, A2, …, and An represents each concave point block. As the dental model is a whole, A1 is adjacent to An. Definition of connectivity of dental model vertices: for any vertex p1 in the dental model and the other vertex p2 in the dental model, if p2 and p1 are on a triangular plane, the vertex p1 and the vertex p2 are communicated.

Step S135, closing any two adjacent concave point blocks in the concave point block set A, namely adding vertexes except any two adjacent concave point blocks in the screened dental model into the concave point block set A to enable the two adjacent concave point blocks to be communicated, and finally forming a closed area by all the concave point blocks, wherein the closed area is marked as a closed area AC. If the initial concave point block set A only has one block, the top points of the tooth jaw model except the concave point block are screened and added into the concave point block to form a closed area.

Step S136, extracting a frame of the closed region AC, that is, screening out a point set B in the closed region AC, so that the point set B maintains the shape of the closed region AC, and the point set B forms a closed ring. The extraction is generally carried out as follows: the closed region AC is a ring-shaped region, and the outermost and innermost points of the ring-shaped region are deleted, and if there is only one point at a certain position, the deletion of the point will make the ring-shaped region not closed, the deletion of the point will not be performed. Repeating the above operations to finally connect only all the points in sequence into a closed line, and ending.

In step S137, the point set B is a closed loop, all the vertexes inside and on the loop constitute one model, and all the vertexes outside and on the loop constitute the other model, so that the point set B divides the dental model into two models, and the model with a small spatial area is selected as the model of the identified crown. The point set B is the set of boundary vertices of the crown.

And step S14, traversing the dental crown set T, judging whether the boundary vertex sets of two adjacent dental crowns are overlapped, and if not, ending the process. If so, the overlap is removed. For example: boundary vertex sets B1 and B2 of crowns T1 and T2 have an overlap, B11 and B12 are respectively a set of overlapping partial boundary vertices and a set of non-overlapping partial boundary vertices of boundary vertex sets B1 and B2, B21 and B22 are respectively a set of overlapping partial boundary vertices and a set of non-overlapping partial boundary vertices of boundary vertex sets B2 and B1, and then B11 and B21 are recalculated to be B11 'and B21', so that B11 'and B21' keep shapes of B11 and B21 as much as possible, and B11 'and B21' have no overlap. Then B11 'and B12 perform step S137 operation on the new boundary vertex set B1' constituting T1, T1 is updated to T1 ', B21' and B22 perform step S137 operation on the new boundary vertex set B2 'constituting T2, and T2 is updated to T2', and the operated crown set is still T, T ═ T1 ', T2', T3, …, Tm }.

The above embodiments are provided only for illustrating the present invention and not for limiting the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and therefore all equivalent technical solutions should also fall within the scope of the present invention, and should be defined by the claims.

Claims

1. A method of identifying a crown portion in a three-dimensional tooth model, comprising:

determining the number m of dental crowns to be identified;

sequentially identifying models of the crowns according to the arrangement sequence of the actual crowns to obtain a crown set T { T1, T2, …, Tm }, T1, T2 and …, wherein Tm represents each crown;

traversing the dental crown set T, judging whether the boundary vertex sets of two adjacent dental crowns have overlapping conditions, and if not, ending; if so, removing the overlap;

identifying a model of each crown comprising:

screening vertexes, except any two adjacent concave point blocks, in the dental model and adding the vertexes into the concave point block set A, so that the two adjacent concave point blocks can be communicated, and finally, forming a closed area for all the concave point blocks, and recording the closed area as a closed area AC;

screening out a point set B in the closed area AC, so that the point set B keeps the shape of the closed area AC, the point set B forms a closed ring, all vertexes in the ring and on the ring form a model, all vertexes outside the ring and on the ring form another model, and selecting a model with a small space area from the two models as a model of the identified dental crown;

definition of the connectivity of the vertices: for any vertex p1 in the dental model and the other vertex p2 in the dental model, if p2 and p1 are on a triangular plane, the vertex p1 and the vertex p2 are communicated.

2. The method for identifying a crown part in a three-dimensional tooth model according to claim 1, wherein a mean value a of an included angle between a normal vector of the vertex and a normal vector of a triangular surface of a neighborhood is calculated for each vertex of the jaw model, and if the mean value a of the included angle is less than 90 degrees, the vertex is a concave point and is added to the first set of concave points S.

3. The method for identifying a crown part in a three-dimensional tooth model according to claim 1, wherein said screening out a point set B in the closed region AC means a repetition of: deleting the outermost and innermost points of the annular closed area AC, and if only one point exists at a certain position, deleting the point to enable the annular area not to be closed, and not deleting the point;

4. The method for identifying a crown portion in a three-dimensional tooth model according to claim 1, wherein assuming that there is an overlap between the boundary vertex sets B1 and B2 of the crown T1 and T2, said removing the overlap comprises:

recalculating B11 and B21 as B11 'and B21', so that B11 'and B21' maintain the shapes of B11 and B21, and B11 'and B21' do not have an overlapping condition;