CN115546444A

CN115546444A - Tooth reconstruction method, system and terminal based on three-dimensional tooth and crown registration

Info

Publication number: CN115546444A
Application number: CN202211496122.5A
Authority: CN
Inventors: 曾令辉; 杨彤
Original assignee: Acedental Ltd
Current assignee: Acedental Ltd
Priority date: 2022-11-28
Filing date: 2022-11-28
Publication date: 2022-12-30
Anticipated expiration: 2042-11-28
Also published as: CN115546444B

Abstract

According to the tooth reconstruction method, the tooth reconstruction system and the tooth reconstruction terminal based on the three-dimensional tooth and the dental crown registration, the dental crown of the mouth-scanned dental crown is replaced by the dental crown of the CBCT reconstructed tooth through the three-stage registration of the CBCT dental data and the mouth-scanned dental crown data, and a high-precision 3D tooth model is reconstructed by using a 3D curved surface reconstruction algorithm; the invention can not only generate high-precision teeth, but also meet the requirements of producing invisible orthodontic appliances and clinical diagnosis.

Description

Tooth reconstruction method, system and terminal based on three-dimensional tooth and crown registration

Technical Field

The invention relates to the field of image registration, in particular to a tooth reconstruction method, a tooth reconstruction system and a tooth reconstruction terminal based on three-dimensional tooth and crown registration.

Background

In recent years, as people gradually pay more attention to the tidiness and beauty of teeth, accurate extraction of a three-dimensional digital tooth model has important application in orthodontic clinical, particularly in the aspects of tooth arrangement and tooth biomechanical simulation. Intact teeth (including crowns and roots) can be reconstructed from 3D data, such as Magnetic Resonance Imaging (MRI), computed Tomography (CT), and Cone Beam Computed Tomography (CBCT). CBCT is currently the most popular imaging technique in dental imaging because it better separates soft and hard tissues. Although CBCT imaging can reconstruct a 3D tooth model, since the accuracy of CBCT is typically between 200-450 microns, the loss of detail information on the crown surface is severe, and the manufacture of invisible appliances requires more accurate crown detail information.

Disclosure of Invention

In view of the above-mentioned shortcomings of the prior art, it is an object of the present invention to provide a tooth reconstruction method, system and terminal based on three-dimensional tooth and crown registration, which are used to solve the above technical problems in the prior art.

To achieve the above and other related objects, the present invention provides a tooth reconstruction method based on three-dimensional tooth and crown registration, the method comprising: respectively obtaining corresponding tooth triangular mesh data and mouth-scanning dental crown triangular mesh data from the acquired CBCT tooth data and mouth-scanning dental crown data; wherein the dental triangular mesh data comprises: tooth triangular mesh data for each tooth; the oral scan dental crown triangle mesh data includes: crown triangle mesh data for each oral scan crown; performing coarse registration on each tooth and each oral-scan dental crown based on the tooth triangular mesh data and the oral-scan dental crown triangular mesh data to obtain coarse registration data; wherein the coarse registration data comprises: one or more dental crown gross registration data pairs; wherein each tooth crown coarse registration data pair comprises: tooth triangular mesh data of a tooth; or comprises the following steps: a registration dental triangular mesh data of a registered tooth and a crown triangular mesh data of a registration oral-scan crown corresponding to the registered tooth; performing tooth crown rough registration data pair screening on the rough registration data, and performing crown coordinate transformation on one or more screened tooth crown rough registration data to obtain a tooth crown registration data pair corresponding to one or more screened tooth crown rough registration data pairs; globally sampling the dental crown registration data pairs of the teeth, and obtaining a preliminary transformation matrix for transforming the oral-scanning dental crowns to dental crown positions on the corresponding teeth based on the calculated feature descriptors of the sampling points; calculating according to the tooth crown registration data pair and the preliminary transformation matrix to obtain a fine registration transformation matrix based on a closest point iterative algorithm; transforming each of the oral-scan dental crowns in the oral-scan dental crown triangular mesh data to a dental crown corresponding position of a corresponding tooth in the dental triangular mesh data based on the fine registration transformation matrix to obtain dental crown triangular mesh data of each of the transformed dental crowns transformed by each of the oral-scan dental crowns; removing crowns of teeth having crowns in the tooth triangular mesh data based on the crown triangular mesh data of each transformed crown for obtaining triangular mesh teeth formed with voids; wherein the triangular mesh tooth formed with the gap includes: a plurality of triangular mesh tooth roots after the tooth crowns are removed and corresponding triangular mesh transformation tooth crowns; and based on a Possion surface reconstruction algorithm, obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the formed gaps.

In an embodiment of the present invention, the obtaining the corresponding tooth triangular mesh data and the mouth-scan crown triangular mesh data from the acquired CBCT tooth data and the mouth-scan crown data respectively includes: performing rigid body transformation on the CBCT tooth data and the oral scanning dental crown data to obtain CBCT tooth transformation data and oral scanning dental crown transformation data with three-axis axial data; marking the positions of teeth in the CBCT tooth transformation data and triangularizing image data to obtain tooth triangular grid data of each tooth; wherein the types of teeth include: maxillary and mandibular teeth; acquiring the position information of each oral-scanning dental crown in the oral-scanning dental crown transformation data, and cutting to acquire dental crown triangular grid data of each oral-scanning dental crown based on the position information; wherein the types of oral dental crowns include: maxillary and mandibular oral dental crowns.

In an embodiment of the invention, the coarsely registering each tooth and each oral crown based on the tooth triangle mesh data and the oral crown triangle mesh data to obtain coarsely registered data includes: respectively and randomly sampling a plurality of points on each oral scanning dental crown to obtain position information corresponding to a plurality of oral scanning dental crown sampling points; respectively and randomly sampling teeth corresponding to each oral scanning dental crown in the tooth triangular grid data within a sampling position range so as to obtain position information corresponding to tooth sampling points with the same number as the oral scanning dental crown sampling points of the corresponding oral scanning dental crown; wherein the range of sampleable positions of each tooth is related to a height value of the corresponding oral crown in the Z-axis of the three axes; calculating to obtain a translation vector for the movement of each oral scanning dental crown according to the position information of the sampling mass center corresponding to each oral scanning dental crown sampling point and the position information of the sampling mass center corresponding to each dental sampling point; obtaining a rotation matrix for rotating each oral-scanning dental crown according to a matrix formed by the position information of each oral-scanning dental crown sampling point and a matrix formed by the position information of each dental sampling point; and obtaining crown triangular grid data of each registered crown according to the position information of each oral-scanning crown sampling point based on the translation vector and the rotation matrix so as to form one or more crown rough registration data pairs for each tooth by using the crown triangular grid data of each tooth.

In an embodiment of the present invention, the screening of the coarse registration data for the pair of dental crown coarse registration data, and performing crown coordinate transformation on the screened one or more pairs of dental crown coarse registration data to obtain the pair of dental crown registration data corresponding to each pair of the screened dental crown coarse registration data includes: acquiring a tooth crown rough registration data pair simultaneously having tooth triangular mesh data of a registered tooth and crown triangular mesh data of a registered oral-scan crown corresponding to the registered tooth; cutting the crown of each registered tooth based on the height value of the registration oral scanning crown corresponding to each registration tooth in the obtained rough registration data pair of the crown of the tooth on the Z axis in three axes, and calculating the surface area of the cut crown; screening the rough registration data pairs of the tooth crowns according to the rough registration data pairs of the tooth crowns of the teeth to the surface area of the cut tooth crowns of the corresponding registered teeth based on the set surface area difference threshold; transforming the crown triangle grid data of the registration oral-scan crowns in the screened crown rough registration data pairs of the teeth to corresponding positions of crowns of the corresponding registration teeth to form corresponding crown registration data pairs of the teeth; wherein the dental crown registration data pair comprises: tooth triangular mesh data of a registered tooth and crown transformed triangular mesh data of a registered oral scan crown corresponding to the registered tooth.

In an embodiment of the present invention, the globally sampling the dental crown registration data pairs and obtaining a preliminary transformation matrix for transforming each oral scan dental crown to a dental crown position on a corresponding tooth based on the calculated feature descriptors of the sampling points includes: respectively down-sampling the tooth triangular grid data of the registered tooth in each tooth crown registration data pair and the tooth crown conversion triangular grid data of the registered mouth-scanned tooth crown of the registered tooth according to the number of grid vertexes of the tooth triangular grid data and calculating the feature descriptors of each grid vertex of the tooth triangular grid data of the registered tooth and the tooth crown conversion triangular grid data of the registered mouth-scanned tooth crown so as to obtain feature descriptor data corresponding to each tooth crown registration data; and based on a fast global registration algorithm, obtaining a preliminary transformation matrix for transforming each oral-scanning dental crown to the dental crown position on the corresponding tooth according to the feature description sub-data corresponding to the dental crown registration data of each tooth.

In an embodiment of the invention, the obtaining a fine registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix calculation based on the closest point iterative algorithm includes: and calculating to obtain a fine registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm taking the preliminary transformation matrix as an initialization matrix.

In an embodiment of the present invention, the removing the crown of the tooth having the crown in the tooth triangular mesh data for obtaining the triangular mesh tooth formed with the gap based on the crown triangular mesh data of each transformed crown comprises: calculating a mesh vertex set within a preset distance threshold from a mesh vertex set in the transformed crown triangular mesh data of the transformed crown corresponding to the distance in the tooth triangular mesh data of one or more teeth with crowns in the tooth triangular mesh data; removing the mesh vertex sets of all teeth from the tooth triangular mesh data to obtain triangular mesh data of all tooth roots with crown removed; one or more interstitial triangular mesh teeth are formed based on the transformed crown triangular mesh data of the transformed crown and the triangular mesh data of each root of the removed crown.

In an embodiment of the present invention, the obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the formed gaps based on the mission surface reconstruction algorithm includes: calculating tooth root normal vector data consisting of normal vectors of mesh vertices on the surfaces of the tooth roots of the tooth triangular mesh data and tooth crown normal vector data consisting of normal vectors of mesh vertices on the surfaces of the oral-scan tooth crowns of the oral-scan tooth crown triangular mesh data; constructing an energy minimization function according to the tooth root normal vector data and the tooth crown normal vector data; and based on a Possion surface reconstruction algorithm, performing model reconstruction according to the energy minimization function and the triangular mesh teeth with the gaps to obtain a three-dimensional tooth model.

To achieve the above and other related objects, the present invention provides a tooth reconstruction system based on three-dimensional tooth and crown registration, the system comprising: the triangular grid data acquisition module is used for respectively acquiring corresponding tooth triangular grid data and oral-scanning dental crown triangular grid data from the acquired CBCT tooth data and the oral-scanning dental crown data; wherein the dental triangular mesh data comprises: tooth triangular mesh data for each tooth; the oral-scan crown triangle mesh data comprises: crown triangle mesh data for each oral scan crown; the rough registration module is connected with the triangular grid data acquisition module and is used for carrying out rough registration on each tooth and each oral-scanning dental crown based on the tooth triangular grid data and the oral-scanning dental crown triangular grid data so as to obtain rough registration data; wherein the coarse registration data comprises: one or more dental crown gross registration data pairs; wherein each tooth crown coarse registration data pair comprises: tooth triangular mesh data of a tooth; or comprises the following steps: a registration dental triangular mesh data of a registered tooth and a crown triangular mesh data of a registration oral-scan crown corresponding to the registered tooth; the data screening module is connected with the rough registration module and is used for screening the rough registration data of the tooth crowns and carrying out crown coordinate transformation on the screened rough registration data of one or more tooth crowns so as to obtain tooth crown registration data pairs corresponding to the screened rough registration data pairs of one or more tooth crowns; the fine-grained registration module is connected with the data screening module and used for carrying out overall sampling on the registration data pairs of the dental crowns of the teeth and obtaining a preliminary transformation matrix for transforming the dental crowns of the oral scans to the positions of the dental crowns on the corresponding teeth based on the calculated feature descriptors of the sampling points; the precise registration module is connected with the data screening module and the fine-grained registration module and used for calculating and obtaining a precise registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm; the mouth-scanning dental crown transformation module is connected with the precise registration module and is used for transforming each mouth-scanning dental crown in the mouth-scanning dental crown triangular grid data to the corresponding dental crown position of the corresponding tooth in the tooth triangular grid data based on the precise registration transformation matrix so as to obtain the dental crown triangular grid data of each transformed dental crown transformed by each mouth-scanning dental crown; a crown removing module connected with the oral scanning crown transformation module and the triangular mesh data acquisition module and used for removing the crown of the tooth with the crown in the tooth triangular mesh data based on the crown triangular mesh data of each transformed crown so as to obtain the triangular mesh tooth with the gap; wherein the triangular mesh tooth formed with the gap includes: a plurality of triangular mesh tooth roots after the tooth crowns are removed and corresponding triangular mesh transformed tooth crowns; and the model reconstruction module is connected with the dental crown removal module and is used for obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the gaps based on the Possion surface reconstruction algorithm.

To achieve the above and other related objects, the present invention provides a tooth reconstruction terminal based on three-dimensional tooth and crown registration, comprising: one or more memories and one or more processors; the one or more memories for storing a computer program; the one or more processors, coupled to the memory, are configured to execute the computer program to perform the tooth reconstruction method based on three-dimensional tooth and crown registration.

As described above, the present invention is a method, system and terminal for reconstructing teeth based on three-dimensional registration of teeth and crowns, and has the following advantages: according to the three-stage registration of the CBCT dental data and the oral-scanning dental crown data, the oral-scanning dental crown is replaced by the CBCT reconstructed dental crown, and a high-precision 3D dental model is reconstructed by using a 3D curved surface reconstruction algorithm; the invention can not only generate high-precision teeth, but also meet the requirements of producing invisible orthodontic appliances and clinical diagnosis (including root treatment, tooth alignment and the like).

Drawings

Fig. 1 is a flowchart illustrating a tooth reconstruction method based on three-dimensional tooth and crown registration according to an embodiment of the present invention.

Fig. 2 is a flowchart illustrating a tooth reconstruction method based on three-dimensional tooth and crown registration according to an embodiment of the present invention.

Fig. 3 is a schematic structural diagram of a tooth reconstruction system based on three-dimensional tooth and crown registration according to an embodiment of the present invention.

Fig. 4 is a schematic structural diagram of a tooth reconstruction terminal based on three-dimensional tooth and crown registration according to an embodiment of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the features in the following embodiments and examples may be combined with each other without conflict.

It is noted that in the following description, reference is made to the accompanying drawings which illustrate several embodiments of the present invention. It is to be understood that other embodiments may be utilized and that mechanical, structural, electrical, and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description is not to be taken in a limiting sense, and the scope of embodiments of the present invention is defined only by the claims of the issued patent. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Spatially relative terms, such as "upper," "lower," "left," "right," "lower," "below," "lower," "over," "upper," and the like, may be used herein to facilitate describing one element or feature's relationship to another element or feature as illustrated in the figures.

Throughout the specification, when a part is referred to as being "connected" to another part, this includes not only a case of being "directly connected" but also a case of being "indirectly connected" with another element interposed therebetween. In addition, when a certain part is referred to as "including" a certain component, unless otherwise stated, other components are not excluded, but it means that other components may be included.

The terms first, second, third, etc. are used herein to describe various elements, components, regions, layers and/or sections, but are not limited thereto. These terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the scope of the present invention.

Also, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises," "comprising," and/or "comprising," when used in this specification, specify the presence of stated features, operations, elements, components, items, species, and/or groups, but do not preclude the presence, or addition of one or more other features, operations, elements, components, items, species, and/or groups thereof. The terms "or" and/or "as used herein are to be construed as inclusive or meaning any one or any combination. Thus, "A, B or C" or "A, B and/or C" means "any of the following: a; b; c; a and B; a and C; b and C; A. b and C ". An exception to this definition will occur only when a combination of elements, functions or operations are inherently mutually exclusive in some way.

The invention provides a tooth reconstruction method, a system and a terminal based on three-dimensional tooth and dental crown registration.A dental crown of a CBCT reconstructed tooth is replaced by a dental crown of a mouth-scanned dental crown through three-stage registration of CBCT tooth data and mouth-scanned dental crown data, and a high-precision 3D tooth model is reconstructed by using a 3D curved surface reconstruction algorithm; the invention can not only generate high-precision teeth, but also meet the requirements of producing invisible orthodontic appliances and clinical diagnosis (including root treatment, tooth alignment and the like).

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings so that those skilled in the art can easily implement the embodiments of the present invention. The present invention may be embodied in many different forms and is not limited to the embodiments described herein.

Fig. 1 shows a flowchart of a tooth reconstruction method based on three-dimensional tooth and crown registration according to an embodiment of the present invention.

The method comprises the following steps:

step S11: and acquiring CBCT tooth data and oral scan dental crown data of the patient to respectively obtain corresponding tooth triangular mesh data and oral scan dental crown triangular mesh data.

In detail, the dental triangular mesh data includes: tooth triangular mesh data for each tooth; wherein the tooth triangular mesh data of each tooth comprises triangular mesh data of a tooth root and triangular mesh data of a tooth crown; the oral scan dental crown triangle mesh data includes: crown triangular mesh data for each oral-scan crown;

it should be noted that each oral crown corresponds to a tooth in the tooth triangular mesh data.

Optionally, step S11 includes:

performing rigid body transformation on the CBCT tooth data and the oral scanning dental crown data to obtain CBCT tooth transformation data and oral scanning dental crown transformation data with three-axis axial data; specifically, CBCT dental data and oral scan dental crown data obtained by scanning a user's teeth with a CBCT scanner and an intraoral scanner are acquired; performing rigid body transformation on the CBCT tooth data to obtain CBCT tooth transformation data with three-axis (X, Y and Z axis which are vertical to each other) axial data; transforming the oral-scanning dental crown data to obtain the oral-scanning dental crown transformation data in a triangular mesh form which is axially consistent with the CBCT dental transformation data;

marking the positions of teeth in the CBCT tooth transformation data and triangularizing image data to obtain tooth triangular grid data of each tooth; wherein the types of teeth include: maxillary and mandibular teeth; in particular, the method comprises the following steps of,

calculating the position information of each tooth in the CBCT tooth transformation data in the DICOM file by using a CBCT tooth division algorithm, and marking each tooth in the CBCT tooth transformation data based on the position information; then carrying out image triangulation on the CBCT tooth transformation data for marking the tooth position by using a Marching Cube algorithm so as to obtain tooth triangular grid data of each tooth;

acquiring the position information of each oral-scanning dental crown in the oral-scanning dental crown transformation data, and cutting to acquire dental crown triangular grid data of each oral-scanning dental crown based on the position information; wherein the types of oral dental crowns include: maxillary and mandibular oral dental crowns. Specifically, the position of each oral-scanned dental crown in the oral-scanned dental crown transformation data is calculated by using an oral-scanned dental division algorithm, and each dental crown is cut out based on the position information, so that dental crown triangular mesh data of each oral-scanned dental crown is obtained.

Optionally, the CBCT dental data is subjected to rigid body transformation (including translation transformation and rotation transformation) by using a CBCT software tool, such that the Z-axis vertical horizontal plane is from bottom to top (ensuring that the upper teeth are above the Z-axis of the lower teeth), the X-axis vertical coronal plane is from front to back (relative to the occlusal image of the observed person), the Y-axis vertical sagittal plane is from left to right (relative to the occlusal image of the observed person), and the three axial directions satisfy the right-hand law. Similarly, the mouth scan crown transformation data is subjected to rigid body transformation using a mesh processing tool such that X, Y, Z three axes are axially coincident with the corresponding CBCT image.

Step S12: coarse registration is performed on each tooth and each oral-scan crown based on the tooth triangular mesh data and the oral-scan crown triangular mesh data to obtain coarse registration data.

In detail, the coarse registration data comprises: one or more dental crown gross registration data pairs; wherein each pair of tooth crown gross registration data comprises: tooth triangular mesh data for a tooth (indicating the absence of a crown); or comprises the following steps: tooth triangular mesh data for a registered tooth and crown triangular mesh data for a registered oral crown corresponding to the registered tooth (indicating the presence of both a tooth root and a tooth crown).

Optionally, step S12 includes:

respectively and randomly sampling a plurality of points (grid vertexes) on each oral scanning dental crown to obtain position information corresponding to a plurality of oral scanning dental crown sampling points; specifically, a height value h of each oral-scanning dental crown on a Z axis in three axes is obtained, and N (N is more than or equal to 1) points are randomly sampled on each oral-scanning dental crown to obtain position information corresponding to N sampling points of the oral-scanning dental crown; (including XYZ-axis coordinate information);

respectively and randomly sampling teeth corresponding to each oral scanning dental crown in the tooth triangular grid data within a sampling position range so as to obtain position information corresponding to tooth sampling points with the same number as the oral scanning dental crown sampling points of the corresponding oral scanning dental crown; wherein the range of sampleable positions of each tooth is related to a height value of the corresponding oral crown in the Z-axis of the three axes; specifically, finding out a corresponding tooth in the tooth triangular grid data according to a tooth position corresponding to each oral-scanning dental crown, if the tooth is an upper jaw tooth, recording the minimum value of the grid vertex of the corresponding tooth triangular grid data in the Z-axis direction as minz, randomly sampling N points from points of Z coordinates in the tooth triangular grid data between minz and minz + h, and obtaining position information (including XYZ-axis coordinate information) corresponding to N tooth sampling points; carrying out the same sampling operation on all the upper teeth; if the tooth is a mandibular tooth, the maximum value of the mesh vertex of the corresponding tooth triangular mesh data in the Z-axis direction is maxz, N points are randomly sampled from the points of the Z coordinate between maxz and maxz-h in the tooth triangular mesh data, and the position information (including XYZ-axis coordinate information) corresponding to the N tooth sampling points is obtained; the exact same sampling operation was performed for all mandibular teeth.

Calculating to obtain a translation vector for the movement of each oral scanning dental crown according to the position information of the sampling mass center corresponding to each oral scanning dental crown sampling point and the position information of the sampling mass center corresponding to each dental sampling point; specifically, a maxillary dental crown sampling point center of mass mA1 and a maxillary dental crown sampling point center of mass mB1 are respectively calculated according to each maxillary dental crown sampling point and each maxillary dental sampling point, a translation vector is recorded as t 1= mB 1-mA 1, and each maxillary dental crown sampling point is moved along t 1; respectively calculating a lower jaw dental crown sampling point mass center mA2 and a lower jaw dental crown sampling point mass center mB2 according to each mouth-scanning dental crown sampling point corresponding to the lower jaw dental crown and each tooth sampling point corresponding to the lower jaw dental crown, recording the translation vector as t2= mB 2-mA 2, and moving each mouth-scanning dental crown sampling point of the lower jaw dental crown along t 2;

obtaining a rotation matrix for rotating each oral scanning dental crown according to a matrix formed by the position information of each oral scanning dental crown sampling point and a matrix formed by the position information of each dental sampling point; specifically, the matrix formed by sampling points of each scanned dental crown of the dental crown is recorded as

The matrix formed by the corresponding maxillary tooth sampling points is recorded

Establishing a least square problem:

（1）

wherein

Is a matrix of rotations of the optical system,

represents the Frobenius norm; wherein, the first and the second end of the pipe are connected with each other,

；（2）

solving the problem by Singular Value Decomposition (SVD) to obtain a rotation matrix corresponding to the maxillary dental crown. Likewise, a corresponding rotation matrix of the mandibular crown can be obtained.

And obtaining crown triangular grid data of each registered crown according to the position information of each oral-scanning crown sampling point based on the translation vector and the rotation matrix so as to form one or more crown rough registration data pairs for each tooth by using the crown triangular grid data of each tooth. Specifically, the maxillary dental crowns are moved along the corresponding translation vector direction of the maxillary dental crowns, then the maxillary dental crowns are rotated by using the rotation matrixes corresponding to the maxillary dental crowns to obtain dental crown triangular grid data of each maxillary registration dental crown, and one or more maxillary dental crown rough registration data pairs are formed by combining the dental triangular grid data of the maxillary dental crowns respectively corresponding to the maxillary dental crowns; moving the lower jaw dental crowns along the corresponding translation vector direction of the lower jaw dental crowns, then rotating the lower jaw dental crowns by using the rotation matrix of the corresponding upper jaw dental crowns to obtain dental crown triangular grid data of each upper jaw registration dental crown, and combining the dental triangular grid data of the lower jaw dental crowns respectively corresponding to the upper jaw dental crowns to form one or more lower jaw dental crown rough registration data pairs.

Step S13: and performing tooth crown rough registration data pair screening on the rough registration data, and performing crown coordinate transformation on the screened one or more tooth crown rough registration data to obtain a tooth crown registration data pair corresponding to the screened one or more tooth crown rough registration data pairs.

Alternatively, since the CBCT scanner can identify the tooth roots and the mouth scan can identify only the tooth crowns, the number of teeth may not be consistent with the number of tooth crowns; meanwhile, since the accuracy of the CBCT scanner is lower than that of the oral scan, the geometric details of the crown surface of the triangular mesh tooth may be greatly different from the surface details of the triangular mesh crown, which may cause noise in the subsequent registration, and therefore step S13 includes:

acquiring a tooth crown rough registration data pair simultaneously having tooth triangular mesh data of a registered tooth and crown triangular mesh data of a registered oral-scan crown corresponding to the registered tooth;

based on the height values of the registration oral-scan crowns corresponding to the registration teeth in the obtained rough registration data pairs of the tooth crowns on the Z axis in three axes, cutting the tooth crowns of the registration teeth, and calculating the surface areas of the cut tooth crowns and the registration oral-scan crowns corresponding to the registration teeth;

screening the rough registration data pairs of the teeth crowns of the teeth according to the rough registration data of the teeth crowns of the teeth, the surface area of the cut teeth crowns of the corresponding registration teeth and the surface area of the registration oral-scanning teeth crowns corresponding to the registration teeth; specifically, comparing the surface area of the cut crown of the registered tooth corresponding to each pair of rough registration data of the crown of the tooth with the surface area of the registration oral-scan crown corresponding to the registered tooth, and if the difference of the surface areas is greater than a surface area difference threshold value, discarding the pair of rough registration data of the crown of the tooth; otherwise, keeping the dental crown rough registration data pair;

transforming the crown triangle grid data of the registration oral-scan crowns in the screened crown rough registration data pairs of the teeth to corresponding positions of crowns of the corresponding registration teeth to form corresponding crown registration data pairs of the teeth; wherein the dental crown registration data pair comprises: tooth triangular mesh data of a registered tooth and crown transformed triangular mesh data of a registered oral scan crown corresponding to the registered tooth. Specifically, the retained crown triangular mesh data of the registration mouth-scanned crown of each tooth crown rough registration data pair is transformed to the corresponding position of the crown of the corresponding registration tooth, and the crown transformation triangular mesh data transformed by the registration mouth-scanned crown and the tooth triangular mesh data of the registration tooth corresponding to the registration mouth-scanned crown are obtained.

Step S14: globally sampling the dental crown registration data pairs of each tooth, and obtaining a preliminary transformation matrix for transforming each oral dental crown to a dental crown position on the corresponding tooth based on the calculated feature descriptors of each sampling point.

The method comprises the steps of calculating the similarity of the surface forms of tooth triangular grid data of a registration tooth in each tooth crown registration data pair and tooth crown conversion triangular grid data of a registration oral-scan tooth crown of the registration tooth, selecting tooth pairs with high similarity, sampling on the tooth pairs, calculating feature descriptors of surface points, and performing overall fine-grained registration based on the feature descriptors.

Optionally, step S14 includes:

respectively down-sampling the tooth triangular grid data of the registered tooth in each tooth crown registration data pair and the tooth crown conversion triangular grid data of the registered mouth-scanned tooth crown of the registered tooth according to the number of grid vertexes of the tooth triangular grid data and calculating the feature descriptors of each grid vertex of the tooth triangular grid data of the registered tooth and the tooth crown conversion triangular grid data of the registered mouth-scanned tooth crown so as to obtain feature descriptor data corresponding to each tooth crown registration data; specifically, the tooth triangular mesh data of the registered tooth in each tooth crown registration data pair and the tooth crown conversion triangular mesh data of the registered oral-scan tooth crown of the registered tooth are down-sampled according to the number of mesh vertexes thereof, and the normal vector of the mesh vertexes after down-sampling and the feature descriptor of each mesh vertex are calculated, including but not limited to: fast Point Feature Histogram (FPFH);

based on a fast global registration algorithm, obtaining a preliminary transformation matrix for transforming each oral-scanning dental crown to a dental crown position on a corresponding tooth according to the feature description subdata corresponding to the dental crown registration data of each tooth; specifically, a fast global registration algorithm is utilized, and then the feature descriptors of each mesh vertex are used as input to register the downsampled points to obtain a preliminary transformation matrix for transforming each oral crown to a crown position on a corresponding tooth.

Step S15: and calculating to obtain a fine registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm.

Optionally, step S15 includes: and calculating to obtain a fine registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm taking the preliminary transformation matrix as an initialization matrix.

Step S16: transforming each of the oral-scan crown triangular mesh data to a crown corresponding position of a corresponding tooth in the tooth triangular mesh data based on the fine registration transformation matrix to obtain crown triangular mesh data for each of the transformed crowns transformed by each of the oral-scan crowns.

It should be noted that. For each upper and lower jaw tooth, if the upper and lower jaw crowns do not exist, the original upper and lower jaw triangular mesh tooth is directly used as a final tooth, and a subsequent algorithm processing flow is not entered.

Step S17: removing crowns of teeth having crowns in the tooth triangular mesh data based on the crown triangular mesh data of each transformed crown for obtaining triangular mesh teeth formed with voids; wherein the triangular mesh tooth formed with the gap includes: a plurality of triangular mesh tooth roots after removing the tooth crowns and corresponding triangular mesh transformed tooth crowns.

Optionally, step S17 includes: calculating a mesh vertex set within a preset distance threshold from a mesh vertex set in the transformed crown triangular mesh data of the transformed crown corresponding to the distance in the tooth triangular mesh data of one or more teeth with crowns in the tooth triangular mesh data; specifically, if a crown of the maxillary tooth exists, given a distance threshold of d microns, a set of vertices of all maxillary teeth within a preset distance threshold (set of d microns) from mesh vertices in transformed crown triangular mesh data of the transformed crown is calculated; the mandibular teeth are made in the same way as the maxillary teeth;

removing the mesh vertex sets of all teeth from the tooth triangular mesh data to obtain triangular mesh data of all tooth roots with crown removed; specifically, removing the vertex sets from the maxillary triangular mesh teeth to obtain triangular mesh tooth roots without crowns; the mandibular teeth are made in the same way as the maxillary teeth;

one or more interstitial triangular mesh teeth are formed based on the transformed crown triangular mesh data of the transformed crown and the triangular mesh data of each root of the removed crown.

Step S18: and based on a Possion surface reconstruction algorithm, obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the formed gaps.

Optionally, step S18: calculating the normal vector data of the tooth root formed by the normal vectors of the grid vertexes on the surface of each tooth root of the tooth triangular grid data and the normal vector data of the tooth crown formed by the normal vectors of the grid vertexes on the surface of the oral-scanning tooth crown of each oral-scanning tooth crown triangular grid data by using the normal vector of each point of the triangular grid;

constructing an energy minimization function according to the tooth root normal vector data and the tooth crown normal vector data;

（3）

wherein the content of the first and second substances,

a function representing a scalar field of a function,

a gradient field representing a scalar field function,

normal vectors representing the vertices of the triangular mesh of the roots and crowns,

the coordinates of the vertices of the triangular mesh representing the tooth roots and the tooth crowns,

a weighting factor representing a penalty factor, which is used to control the weight ratio between the gradient field fit and the scalar field function fit of the scalar field function.

Based on a Possion surface reconstruction algorithm, performing model reconstruction according to the energy minimization function and the triangular mesh teeth with the gaps to obtain a three-dimensional tooth model; specifically, the normal vector of each point is calculated, so that the normal vector of each point does not need to be estimated by using the point and the adjacent points, and the problem is solved by adopting a numerical algorithm for solving an energy minimization function in a screening Possion surface reconstruction method, so that the high-precision 3D tooth model is obtained.

To better illustrate the above described tooth reconstruction method based on three-dimensional tooth and crown registration, the present invention provides the following specific embodiments.

Example 1: a tooth reconstruction method based on three-dimensional tooth and crown registration. Fig. 2 is a schematic flow chart of tooth reconstruction based on three-dimensional tooth and crown registration; the method comprises the following steps:

step 1: acquiring triangular mesh tooth and crown data:

and scanning the teeth of the user by using the CBCT scanner and the intraoral scanner to obtain CBCT dental occlusion DICOM image data and triangular mesh dental occlusion data of the user. The CBCT occlusion is subjected to rigid body transformation (including translation transformation and rotation transformation) by using a CBCT software tool, so that a Z-axis vertical horizontal plane is from bottom to top (the upper teeth are ensured to be above the Z-axis of the lower teeth), an X-axis vertical coronal plane is from front to back (relative to an occlusion image of an observed person), a Y-axis vertical sagittal plane is from left to right (relative to an occlusion image of the observed person), and the three axial directions meet the right-hand law. Similarly, the mesh processing tool is used to perform rigid body transformation on the oral occlusal data so that X, Y, Z three axial directions are axially coincident with the corresponding CBCT image. Calculating the position information of each tooth in the DICOM file by using a CBCT tooth division algorithm, marking each tooth based on the position information, and triangulating the image data of the teeth by using a Marching Cube algorithm to obtain tooth data of a triangular mesh; meanwhile, the position of each dental crown in the dental crown and gum data of the triangular mesh is calculated by using a mouth-scanning tooth-separating algorithm, and each dental crown is cut out based on the position information, so that the dental crown data of the triangular mesh is obtained.

Step 2, coarse registration of triangular mesh teeth and crowns:

the upper and lower teeth are roughly registered with the corresponding upper and lower crowns respectively, and because the registration algorithm is the same, the registration algorithm and flow will be described only by taking the upper jaw pair (CBCT tooth and corresponding oral scan crown) as an example. Firstly, for a single oral-scan dental crown, recording the height of the single oral-scan dental crown in the Z-axis direction as h, randomly sampling N points on the oral-scan dental crown, finding out a corresponding CBCT tooth according to the tooth position, recording the minimum value of the vertex of a triangular patch in the Z-axis direction as minz, randomly sampling N points from the points of the CBCT tooth surface point Z coordinate between minz and minz + h, (if the CBCT tooth is a mandibular tooth, recording the maximum value of the vertex of the triangular patch in the Z-axis direction as maxz, randomly sampling N points from the points of the CBCT tooth surface point Z coordinate between maxz and maxz-h), and carrying out completely same sampling operation on all maxillary tooth pairs. And respectively calculating the centroid mA of the upper half dental crown sampling point and the centroid mB of the dentition sampling point, recording the translation vector as t = mB-mA, and moving all the sampling points of the upper half dental crown along t. Recording a matrix formed by the upper half dental crown sampling points as A, recording a matrix formed by the upper half dental row sampling points as B, and establishing a least square problem: solving the problem using Singular Value Decomposition (SVD) yields a rotation matrix. And moving the upper half dental crown along the t, and then rotating the upper half dental crown by using a rotation matrix to obtain a final coarse registration result.

And step 3: screening the triangular mesh teeth and the corresponding crowns and cutting the crowns;

since the data processing algorithms of the upper and lower teeth are consistent with the data processing algorithms of the corresponding upper and lower crowns, the specific algorithm of this step will be described only by taking the upper teeth and the upper crowns as examples. Since the CBCT scanner can identify the tooth roots and the oral scan can only identify the tooth crowns, the number of maxillary teeth may not be consistent with the number of maxillary tooth crowns; meanwhile, the accuracy of the CBCT scanner is lower than that of oral scanning, so that the difference between the geometrical details of the surface of the dental crown of the triangular mesh tooth and the surface details of the dental crown of the triangular mesh is large, and noise is caused to subsequent registration, therefore, the following post-processing algorithm is adopted in the patent: firstly, finding a data pair with maxillary teeth and maxillary crowns, cutting out the crown parts corresponding to the maxillary teeth based on the height of the maxillary crowns, respectively calculating the surface areas of the maxillary teeth and the maxillary crowns, and discarding the data pair if the area difference is greater than a given threshold value; otherwise, the crown portion of the maxillary tooth is retained while the corresponding maxillary crown is transformed to the coordinate position of the maxillary tooth to form a data pair.

And 4, step 4: globally sampling and calculating a feature descriptor of a sampling point to perform fine-grained matching;

and performing fine-grained registration based on the calculated crowns of the upper and lower teeth and the corresponding data pairs of the upper and lower crowns, wherein the registration algorithm of the step is described by taking the registration of the crowns of the upper and lower teeth and the corresponding crowns of the upper and lower teeth as an example because the registration algorithm of the upper and lower teeth is the same as the registration algorithm of the lower teeth. First, the triangular mesh data of the crowns of the maxillary teeth and the corresponding maxillary crowns are down-sampled according to the number of vertexes, and normal vectors of the down-sampled vertexes and a feature descriptor of each vertex are calculated, including but not limited to: fast Point Feature Histogram (FPFH); and then taking the feature descriptor of each vertex as input, and further registering the down-sampled points by using a fast global registration algorithm to obtain a transformation matrix for transforming the maxillary dental crown to the dental crown position of the corresponding maxillary tooth.

And 5: carrying out fine registration by using a closest point iterative algorithm;

and further carrying out fine registration to obtain a final result based on the registration result obtained in the last step. Since the fine registration algorithms for the upper and lower jaws are the same, the registration algorithm of this step is described here only by way of example of the registration of the crowns of the upper teeth and the corresponding upper dental crowns. Firstly, extracting the dental crowns of a plurality of maxillary teeth calculated in the step 3 and point cloud data of corresponding maxillary dental crown data pairs; and (5) taking the transformation matrix obtained in the step (4) as an initialization transformation matrix of an Iterative Closest Point (ICP) algorithm, and calculating to obtain a fine registration transformation matrix.

Step 6: removing the crowns of the triangular mesh teeth;

since the treatment algorithm for the maxillary teeth is the same as that for the mandibular teeth, only the maxillary teeth treatment algorithm will be described herein. And transforming the crown of the upper half-mouth triangular mesh to the coordinate position of the dentition of the corresponding upper half-mouth triangular mesh based on the coarse registration transformation matrix, the fine-grained transformation matrix and the fine registration transformation matrix. Because the subsequent surface reconstruction algorithm needs to utilize the normal vector of each vertex of the triangular mesh, the normal vectors of each vertex of all triangular mesh teeth and the normal vector of each vertex of the oral scan dental crown triangular mesh are calculated according to a normal vector formula. For each maxillary tooth, if the maxillary crown does not exist, directly using the original maxillary triangular mesh tooth as a final tooth without entering a subsequent algorithm processing flow; if the maxillary dental crown exists, giving a distance threshold value of d micrometers, calculating vertex sets of all maxillary teeth within d micrometers of the transformed maxillary dental crown vertex set, removing the vertex sets from the maxillary triangular mesh teeth to obtain triangular mesh tooth roots without dental crowns, and finally forming triangular mesh teeth with gaps;

and 7: reconstructing teeth by using a screened Possion curved surface reconstruction method;

further treatment is performed on each of the gapped teeth resulting from the previous treatment. Taking the three-dimensional coordinates and normal vectors of each vertex to construct an energy minimization function: the normal vector of each point is calculated, so that the normal vector of each point does not need to be estimated by using the points and adjacent points, and the problem is solved by adopting a numerical algorithm for solving an energy minimization function in a screening Possion surface reconstruction method, so that the high-precision 3D tooth model is obtained.

Similar in principle to the embodiments described above, the present invention provides a dental reconstruction system based on three-dimensional tooth and crown registration.

Specific embodiments are provided below in conjunction with the attached figures:

fig. 3 shows a schematic structural diagram of a tooth reconstruction system based on three-dimensional tooth and crown registration in an embodiment of the invention.

The system comprises:

the triangular grid data acquisition module 31 is used for respectively acquiring corresponding tooth triangular grid data and oral scan dental crown triangular grid data from the acquired CBCT tooth data and oral scan dental crown data; wherein the dental triangular mesh data comprises: tooth triangular mesh data for each tooth; the oral scan dental crown triangle mesh data includes: crown triangle mesh data for each oral scan crown;

a rough registration module 32, connected to the triangular mesh data acquisition module 31, for performing rough registration on each tooth and each oral-scan dental crown based on the tooth triangular mesh data and the oral-scan dental crown triangular mesh data to obtain rough registration data; wherein the coarse registration data comprises: one or more dental crown gross registration data pairs; wherein each tooth crown coarse registration data pair comprises: tooth triangular mesh data of a tooth; or comprises the following steps: a registration dental triangular mesh data of a registered tooth and a crown triangular mesh data of a registration oral-scan crown corresponding to the registered tooth;

a data screening module 33, connected to the coarse registration module 32, configured to perform tooth crown coarse registration data pair screening on the coarse registration data, and perform crown coordinate transformation on the screened one or more tooth crown coarse registration data pairs to obtain tooth crown registration data pairs corresponding to the screened one or more tooth crown coarse registration data pairs;

the fine-grained registration module 34 is connected with the data screening module 33 and used for carrying out overall sampling on the registration data pairs of the dental crowns of the teeth and obtaining a preliminary transformation matrix for transforming the dental crowns of the oral scans to the positions of the dental crowns of the corresponding teeth based on the calculated feature descriptors of the sampling points;

the precise registration module 35 is connected with the data screening module 33 and the fine-grained registration module 34, and is used for calculating and obtaining a precise registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm;

an oral-scanning crown transformation module 36 connected to the precise registration module 35 for transforming each oral-scanning crown in the oral-scanning crown triangular mesh data to a corresponding crown position of a corresponding tooth in the tooth triangular mesh data based on the precise registration transformation matrix to obtain crown triangular mesh data of each transformed crown transformed by each oral-scanning crown;

a crown removal module 37 connected to the oral crown transformation module 36 and the triangular mesh data acquisition module 31 for removing a crown of a tooth having a crown in the tooth triangular mesh data based on the crown triangular mesh data of each transformed crown for obtaining a triangular mesh tooth formed with a gap; wherein the triangular mesh tooth formed with the space includes: a plurality of triangular mesh tooth roots after the tooth crowns are removed and corresponding triangular mesh transformed tooth crowns;

and the model reconstruction module 38 is connected with the dental crown removal module 37 and is used for obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the gaps formed based on the Possion surface reconstruction algorithm.

It should be noted that the division of each module in the system embodiment of fig. 3 is only a division of a logical function, and all or part of the actual implementation may be integrated into one physical entity or may be physically separated. And these units can be implemented entirely in software, invoked by a processing element; or may be implemented entirely in hardware; and part of the units can be realized in the form of calling software by the processing element, and part of the units can be realized in the form of hardware.

Since the implementation principle of the tooth reconstruction system based on the three-dimensional tooth and crown registration has been described in the foregoing embodiments, repeated description is omitted here.

Fig. 4 shows a schematic structural diagram of a tooth reconstruction terminal 40 based on three-dimensional tooth and crown registration in an embodiment of the invention.

The tooth reconstruction terminal 40 based on three-dimensional tooth and crown registration includes: a memory 41 and a processor 42. The memory 41 is used for storing a computer program; the processor 42 runs a computer program to implement a tooth reconstruction method based on three-dimensional tooth and crown registration as described in fig. 1.

Alternatively, the number of the memories 41 may be one or more, the number of the processors 42 may be one or more, and fig. 4 illustrates one example.

Optionally, the processor 42 in the tooth reconstruction terminal 40 based on the three-dimensional tooth and crown registration loads one or more instructions corresponding to the progress of the application program into the memory 41 according to the steps shown in fig. 1, and the processor 42 executes the application program stored in the first memory 41, so as to implement various functions in the tooth reconstruction method based on the three-dimensional tooth and crown registration shown in fig. 1.

Optionally, the memory 41 may include, but is not limited to, a high speed random access memory, a non-volatile memory. Such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state storage devices; the Processor 42 may include, but is not limited to, a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, or discrete hardware components.

Optionally, the Processor 42 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.

The present invention also provides a computer readable storage medium storing a computer program which when executed implements a tooth reconstruction method based on three-dimensional tooth and crown registration as shown in fig. 1. The computer-readable storage medium may include, but is not limited to, floppy diskettes, optical disks, CD-ROMs (compact disc-read only memories), magneto-optical disks, ROMs (read-only memories), RAMs (random access memories), EPROMs (erasable programmable read only memories), EEPROMs (electrically erasable programmable read only memories), magnetic or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing machine-executable instructions. The computer readable storage medium may be a product that is not accessed by the computer device or may be a component that is used by an accessed computer device.

In conclusion, according to the tooth reconstruction system based on the registration of the three-dimensional tooth and the dental crown, the dental crown of the CBCT reconstructed tooth is replaced by the dental crown of the oral-scanning dental crown through the three-stage registration of the CBCT tooth data and the oral-scanning dental crown data, and a high-precision 3D tooth model is reconstructed by using a 3D curved surface reconstruction algorithm; the invention can not only generate high-precision teeth, but also meet the requirements of producing invisible orthodontic appliances and clinical diagnosis (including root treatment, tooth alignment and the like). Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.

The foregoing embodiments are merely illustrative of the principles of the present invention and its efficacy, and are not to be construed as limiting the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims

1. A method for dental reconstruction based on three-dimensional registration of teeth and crowns, the method comprising:

respectively obtaining corresponding tooth triangular mesh data and mouth-scanning dental crown triangular mesh data from the acquired CBCT tooth data and mouth-scanning dental crown data; wherein the dental triangular mesh data comprises: tooth triangular mesh data for each tooth; the oral scan dental crown triangle mesh data includes: crown triangle mesh data for each oral scan crown;

performing rough registration on each tooth and each oral-scanning dental crown based on the tooth triangular mesh data and the oral-scanning dental crown triangular mesh data to obtain rough registration data; wherein the coarse registration data comprises: one or more dental crown gross registration data pairs; wherein each tooth crown coarse registration data pair comprises: tooth triangular mesh data of a tooth; or comprises the following steps: a registration dental triangular mesh data of a registered tooth and a crown triangular mesh data of a registration oral-scan crown corresponding to the registered tooth;

performing tooth crown rough registration data pair screening on the rough registration data, and performing crown coordinate transformation on one or more screened tooth crown rough registration data to obtain a tooth crown registration data pair corresponding to one or more screened tooth crown rough registration data pairs;

globally sampling the dental crown registration data pairs of the teeth, and obtaining a preliminary transformation matrix for transforming the oral-scanning dental crowns to dental crown positions on the corresponding teeth based on the calculated feature descriptors of the sampling points;

based on a closest point iterative algorithm, calculating according to the tooth crown registration data pair and the preliminary transformation matrix to obtain a fine registration transformation matrix;

transforming each of the oral-scan dental crowns in the oral-scan dental crown triangular mesh data to a dental crown corresponding position of a corresponding tooth in the dental triangular mesh data based on the fine registration transformation matrix to obtain dental crown triangular mesh data of each of the transformed dental crowns transformed by each of the oral-scan dental crowns;

removing crowns of teeth having crowns in the tooth triangular mesh data based on the crown triangular mesh data of each transformed crown for obtaining triangular mesh teeth formed with voids; wherein the triangular mesh tooth formed with the gap includes: a plurality of triangular mesh tooth roots after the tooth crowns are removed and corresponding triangular mesh transformed tooth crowns;

and based on a Possion surface reconstruction algorithm, obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the formed gaps.

2. A tooth reconstruction method based on three-dimensional tooth and crown registration according to claim 1, wherein the obtaining of corresponding tooth triangle mesh data and mouth-scanned crown triangle mesh data from the acquired CBCT tooth data and mouth-scanned crown data, respectively, comprises:

performing rigid transformation on the CBCT tooth data and the oral-scanning dental crown data to obtain CBCT tooth transformation data and oral-scanning dental crown transformation data with three-axis axial data;

marking the positions of teeth in the CBCT tooth transformation data and triangularizing image data to obtain tooth triangular grid data of each tooth; wherein the types of teeth include: maxillary and mandibular teeth;

acquiring the position information of each oral-scanning dental crown in the oral-scanning dental crown transformation data, and cutting to acquire dental crown triangular grid data of each oral-scanning dental crown based on the position information; wherein the types of oral dental crowns include: maxillary and mandibular oral dental crowns.

3. A method for tooth reconstruction based on three-dimensional tooth and crown registration according to claim 1, wherein the coarse registration of each tooth and each oral-scan crown based on the tooth triangular mesh data and the oral-scan crown triangular mesh data to obtain coarse registration data comprises:

respectively and randomly sampling a plurality of points on each oral scanning dental crown to obtain position information corresponding to a plurality of oral scanning dental crown sampling points;

respectively and randomly sampling teeth corresponding to each oral scanning dental crown in the tooth triangular grid data within a sampling position range so as to obtain position information corresponding to tooth sampling points with the same number as the oral scanning dental crown sampling points of the corresponding oral scanning dental crown; wherein the range of sampleable positions of each tooth is related to a height value of the corresponding oral crown in the Z-axis of the three axes;

calculating to obtain a translation vector for the movement of each oral scanning dental crown according to the position information of the sampling mass center corresponding to each oral scanning dental crown sampling point and the position information of the sampling mass center corresponding to each dental sampling point;

obtaining a rotation matrix for rotating each oral scanning dental crown according to a matrix formed by the position information of each oral scanning dental crown sampling point and a matrix formed by the position information of each dental sampling point;

and obtaining crown triangular grid data of each registered crown according to the position information of each oral-scanning crown sampling point based on the translation vector and the rotation matrix so as to form one or more crown rough registration data pairs for each tooth by using the crown triangular grid data of each tooth.

4. A method for reconstructing a tooth based on three-dimensional tooth and crown registration according to claim 1, wherein the step of performing a crown rough registration data pair screening on the rough registration data, and performing a crown coordinate transformation on the screened one or more crown rough registration data to obtain a crown registration data pair corresponding to each screened crown rough registration data pair comprises:

cutting the crown of each registered tooth based on the height value of the registration oral scanning crown corresponding to each registration tooth in the obtained rough registration data pair of the crown of the tooth on the Z axis in three axes, and calculating the surface area of the cut crown;

screening the rough registration data pairs of the tooth crowns according to the rough registration data pairs of the tooth crowns of the teeth to the surface area of the cut tooth crowns of the corresponding registered teeth based on the set surface area difference threshold;

transforming the crown triangular grid data of the registration mouth-scanned crowns in the screened coarse registration data pairs of the crowns of the teeth to corresponding positions of crowns of the corresponding registration teeth to form corresponding crown registration data pairs of the teeth; wherein the dental crown registration data pair comprises: tooth triangular mesh data of a registered tooth and crown transformed triangular mesh data of a registered oral scan crown corresponding to the registered tooth.

5. A tooth reconstruction method based on three-dimensional tooth and crown registration according to claim 1, wherein the globally sampling each pair of tooth crown registration data and obtaining a preliminary transformation matrix for transforming each oral crown to a crown position on the corresponding tooth based on the computed feature descriptors of each sampling point comprises:

respectively down-sampling the tooth triangular grid data of the registered tooth in each tooth crown registration data pair and the tooth crown conversion triangular grid data of the registered mouth-scanned tooth crown of the registered tooth according to the number of grid vertexes of the tooth triangular grid data and calculating the feature descriptors of each grid vertex of the tooth triangular grid data of the registered tooth and the tooth crown conversion triangular grid data of the registered mouth-scanned tooth crown so as to obtain feature descriptor data corresponding to each tooth crown registration data;

and based on a fast global registration algorithm, obtaining a preliminary transformation matrix for transforming each oral-scanning dental crown to the dental crown position on the corresponding tooth according to the feature description sub-data corresponding to the dental crown registration data of each tooth.

6. A tooth reconstruction method based on three-dimensional tooth and crown registration according to claim 1, wherein said obtaining a fine registration transformation matrix from said pair of tooth crown registration data and said preliminary transformation matrix calculation based on a closest point iterative algorithm comprises:

and calculating to obtain a fine registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm taking the preliminary transformation matrix as an initialization matrix.

7. A tooth reconstruction method based on three-dimensional tooth and crown registration according to claim 1, wherein the removing the crown of the tooth with crown in the tooth triangular mesh data for obtaining the triangular mesh tooth formed with the gap based on the triangular mesh data of each transformed crown comprises:

calculating a mesh vertex set within a preset distance threshold from a mesh vertex set in the transformed crown triangular mesh data of the transformed crown corresponding to the distance in the tooth triangular mesh data of one or more teeth with crowns in the tooth triangular mesh data;

removing the mesh vertex sets of all teeth from the tooth triangular mesh data to obtain triangular mesh data of all tooth roots with crown removed;

8. The tooth reconstruction method based on three-dimensional tooth and crown registration according to claim 1, wherein the Poission surface reconstruction algorithm, obtaining the reconstructed three-dimensional tooth model from the triangular mesh teeth formed with the gaps comprises:

calculating tooth root normal vector data consisting of normal vectors of mesh vertices on the surfaces of the tooth roots of the tooth triangular mesh data and tooth crown normal vector data consisting of normal vectors of mesh vertices on the surfaces of the oral-scan tooth crowns of the oral-scan tooth crown triangular mesh data;

and based on a Possion surface reconstruction algorithm, performing model reconstruction according to the energy minimization function and the triangular mesh teeth with the gaps to obtain a three-dimensional tooth model.

9. A dental reconstruction system based on three-dimensional tooth and crown registration, the system comprising:

the triangular grid data acquisition module is used for respectively acquiring corresponding tooth triangular grid data and oral-scanning dental crown triangular grid data from the acquired CBCT tooth data and the oral-scanning dental crown data; wherein the dental triangular mesh data comprises: tooth triangular mesh data for each tooth; the oral scan dental crown triangle mesh data includes: crown triangle mesh data for each oral scan crown;

the rough registration module is connected with the triangular grid data acquisition module and is used for carrying out rough registration on each tooth and each oral-scanning dental crown based on the tooth triangular grid data and the oral-scanning dental crown triangular grid data so as to obtain rough registration data; wherein the coarse registration data comprises: one or more tooth crown gross registration data pairs; wherein each tooth crown coarse registration data pair comprises: tooth triangular mesh data of a tooth; or comprises the following steps: a registration dental triangular mesh data of a registered tooth and a crown triangular mesh data of a registration oral-scan crown corresponding to the registered tooth;

the data screening module is connected with the rough registration module and is used for screening the rough registration data of the tooth crowns and carrying out crown coordinate transformation on the screened rough registration data of one or more tooth crowns so as to obtain tooth crown registration data pairs corresponding to the screened rough registration data pairs of one or more tooth crowns;

the fine-grained registration module is connected with the data screening module and used for carrying out overall sampling on the registration data pairs of the dental crowns of the teeth and obtaining a preliminary transformation matrix for transforming the dental crowns of the oral scans to the positions of the dental crowns on the corresponding teeth based on the calculated feature descriptors of the sampling points;

the precise registration module is connected with the data screening module and the fine-grained registration module and used for calculating and obtaining a precise registration transformation matrix according to the tooth crown registration data pair and the preliminary transformation matrix based on a closest point iterative algorithm;

the mouth-scanning dental crown transformation module is connected with the precise registration module and is used for transforming each mouth-scanning dental crown in the mouth-scanning dental crown triangular grid data to the corresponding dental crown position of the corresponding tooth in the tooth triangular grid data based on the precise registration transformation matrix so as to obtain the dental crown triangular grid data of each transformed dental crown transformed by each mouth-scanning dental crown;

a crown removing module connected with the oral scanning crown transformation module and the triangular mesh data acquisition module and used for removing the crown of the tooth with the crown in the tooth triangular mesh data based on the crown triangular mesh data of each transformed crown so as to obtain the triangular mesh tooth with the gap; wherein the triangular mesh tooth formed with the gap includes: a plurality of triangular mesh tooth roots after the tooth crowns are removed and corresponding triangular mesh transformation tooth crowns;

and the model reconstruction module is connected with the dental crown removal module and is used for obtaining a reconstructed three-dimensional tooth model according to the triangular mesh teeth with the gaps on the basis of a Possion surface reconstruction algorithm.

10. A dental reconstruction terminal based on three-dimensional tooth and crown registration, comprising: one or more memories and one or more processors;

the one or more memories for storing a computer program;

the one or more processors, coupled to the memory, to execute the computer program to perform the method of any of claims 1-8.