CN116246046A

CN116246046A - Three-dimensional tooth model segmentation method and device

Info

Publication number: CN116246046A
Application number: CN202211542607.3A
Authority: CN
Inventors: 周金海; 王勇; 周达超
Original assignee: Guangzhou Heygears IMC Inc
Current assignee: Guangzhou Heygears IMC Inc
Priority date: 2022-12-02
Filing date: 2022-12-02
Publication date: 2023-06-09

Abstract

The application relates to a three-dimensional tooth model segmentation method and device. The method comprises the following steps: acquiring a gum line of the three-dimensional tooth model and a tooth area of the three-dimensional tooth model; extracting peak points of a gum line, and matching the peak points to obtain a peak point combination; determining a segmentation path between teeth in the tooth region based on the peak point combination; and carrying out segmentation processing on the tooth region according to the segmentation path to obtain the target segmented tooth. The method and the device solve the technical problem that the three-dimensional tooth model cannot be accurately segmented.

Description

Three-dimensional tooth model segmentation method and device

Technical Field

The application relates to the technical field of tooth orthodontics, in particular to a three-dimensional tooth model segmentation method and device.

Background

With the development of scientific and technological information, the technology of manufacturing technology, digital modeling technology, material science, numerical control technology and the like are rapidly developed and the subjects are mutually fused, and the computer technology is increasingly penetrated in various aspects of teaching, scientific research and clinical application in various fields of medicine and can be better mutually cooperated. With the development and popularization of measurement technology, people can conveniently obtain a digital tooth model, which plays an important role in oral clinical diagnosis and treatment process. The advent and development of 3D printing technology has become one of the current hot ports, and application of 3D printing technology to the medical field is becoming more and more common. The 3D printing has been applied in the medical field for over twenty years, and is widely applied to operations such as oral implantation, orthopaedics, neurosurgery and the like.

The inventors found that: in the current application scenario of orthodontic diagnosis and treatment, the result needs to be rapidly presented to a user, the diagnosis and treatment confidence is enhanced, namely, the tooth target position simulation is carried out, the tooth target position simulation refers to the scanning data of an input patient, the effect after correction can be obtained and displayed through simple operation and automatic operation of a system, and continuous animation changing from an original state to a target state can be demonstrated, so that doctors and patients can be better communicated, and the achievement is promoted. The method is used as a front link of orthodontic, bears the main responsibility of assisting the communication of doctors and patients, and the output result of target position simulation is one of the contents of follow-up link derived production, and tooth separation is an important link.

At present, in the related art, the three-dimensional tooth model segmentation technology belongs to a starting stage, and the high-efficiency processing requirement is difficult to meet.

Disclosure of Invention

The application provides a three-dimensional tooth model segmentation method and device, which are used for solving the technical problem that a three-dimensional tooth model cannot be accurately segmented.

According to an aspect of an embodiment of the present application, there is provided a three-dimensional tooth model segmentation method including: acquiring a gum line of the three-dimensional tooth model and a tooth area of the three-dimensional tooth model; extracting the peak points of the gum line, and matching the peak points to obtain a peak point combination; determining a dividing path between teeth in the tooth region based on the peak point combination; and carrying out segmentation processing on the tooth region according to the segmentation path to obtain the target segmented tooth.

According to another aspect of the embodiments of the present application, there is also provided a method for manufacturing a dental appliance, including:

obtaining a target segmented tooth according to any one of the methods described above;

printing to obtain a molded tooth model based on the target segmented teeth; wherein the shaped tooth model is used to obtain a dental appliance.

According to another aspect of the embodiments of the present application, there is also provided a three-dimensional tooth model segmentation apparatus including:

a memory and a processor, the memory storing a computer program which, when executed by the processor, performs any of the methods described above.

According to another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium storing a computer program which, when executed by a processor, performs any of the methods described above.

Compared with the related art, the technical scheme provided by the embodiment of the application has the following advantages:

after the gum line of the three-dimensional tooth model and the tooth area of the three-dimensional tooth model are obtained, the peak point of the gum line can be extracted, the peak point is paired, the peak point combination is obtained, then the segmentation path is determined according to the peak point combination, the tooth area is segmented according to the segmentation path, and the target segmented tooth is obtained, so that the effect of accurately dividing the three-dimensional tooth model is achieved, and the problem of low tooth dividing accuracy of the three-dimensional tooth model in the prior art is solved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described, and it is obvious to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort.

FIG. 1 is a schematic diagram of a hardware environment of an alternative three-dimensional tooth model segmentation method according to an embodiment of the present application;

FIG. 2 is a flow chart of an alternative three-dimensional tooth model segmentation method provided in accordance with an embodiment of the present application;

FIG. 3 is a schematic illustration of the peak and trough points of an alternative three-dimensional tooth model provided in accordance with an embodiment of the present application;

FIG. 4 is a schematic view of a tooth segmentation of an alternative three-dimensional tooth model provided in accordance with an embodiment of the present application;

FIG. 5 is a flow chart of an alternative three-dimensional tooth model provided in accordance with an embodiment of the present application;

FIG. 6 is a schematic illustration of an alternative combination of peak points of a three-dimensional tooth model provided in accordance with an embodiment of the present application;

FIG. 7 is a graph of segmentation of an alternative three-dimensional tooth model provided in accordance with an embodiment of the present application;

fig. 8 is a flowchart of an alternative gum line extraction method provided in accordance with an embodiment of the present application;

fig. 9 is a schematic diagram of a polygonal shaped patch for an alternative gum line extraction method provided in accordance with an embodiment of the present application;

fig. 10 is a diagram of an alternative gum line extraction method recognition model provided in accordance with an embodiment of the present application;

FIG. 11 is a schematic illustration of a convolution pooling operation based on edges in a target tooth model according to an alternative gum line extraction method provided in an embodiment of the present application;

FIG. 12 is a block diagram of an alternative three-dimensional tooth model segmentation apparatus according to an embodiment of the present application.

Detailed Description

For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present application based on the embodiments herein.

In the following description, suffixes such as "module", "component", or "unit" for representing elements are used only for facilitating the description of the present application, and are not of specific significance per se. Thus, "module" and "component" may be used in combination.

To solve the problems mentioned in the background art, according to an aspect of the embodiments of the present application, an embodiment of a three-dimensional tooth model segmentation method is provided.

Alternatively, in the embodiment of the present application, the above method may be applied to a hardware environment formed by the terminal 101 and the server 103 as shown in fig. 1. As shown in fig. 1, the server 103 is connected to the terminal 101 through a network, which may be used to provide services to the terminal or a client installed on the terminal, and a database 105 may be provided on the server or independent of the server, for providing data storage services to the server 103, where the network includes, but is not limited to: the terminal 101 may be a device for acquiring a three-dimensional dental model, such as a mouth-scan model scanner, or a terminal for receiving a three-dimensional dental model, such as a cell phone, a computer, etc., a wide area network, a metropolitan area network, or a local area network.

The three-dimensional tooth model segmentation method in the embodiment of the present application may be performed by the server 103, or may be performed jointly by the server 103 and the terminal 101.

As shown in fig. 2, the method may include the steps of:

step S202, acquiring a gum line of the three-dimensional tooth model and a tooth area of the three-dimensional tooth model;

step S204, extracting peak points of a gum line, and matching the peak points to obtain a peak point combination;

step S206, determining a segmentation path between teeth in the tooth area based on the peak point combination;

step S208, the tooth area is subjected to segmentation processing according to the segmentation path, and the target segmented tooth is obtained.

Optionally, the three-dimensional tooth model segmentation method is aimed at accurately dividing each tooth on the three-dimensional tooth model into individual teeth.

The three-dimensional tooth model may be an oral scan model, which refers to a three-dimensional model including a tooth portion of a user generated by scanning the inside of the user's oral cavity. The data of the three-dimensional tooth model is displayed on a computer or a server, and the shape or the style of the three-dimensional tooth model can be displayed through a display screen, so that a doctor can watch the three-dimensional tooth model conveniently.

Alternatively, the gum line may be a dividing line between the tooth area and the gum area on the three-dimensional tooth model, and the gum line may be determined in various manners, for example, by a manner of identifying a neural network model, or by a manner of identifying a two-dimensional projection image of the three-dimensional tooth model.

Alternatively, the peak points of the three-dimensional tooth model may be points higher than adjacent points on the gum line of the three-dimensional tooth model. For example, the line connecting points on the gum line of a three-dimensional tooth model continues to run high, beginning to descend at one of the points, which is then a peak point. The peak point is understood to be the highest point of the gum at the tooth gap between two teeth.

In this embodiment, the peak points may be paired, and each two peak points may be determined as a peak point combination, where two peak points in a peak point combination are used to determine the boundary between two teeth. Each peak-point combination comprises two peak points.

In this embodiment, after the combination of the peak points is determined, the dividing path between two teeth in the tooth area may be determined according to the two peak points.

The tooth region can be segmented according to the segmentation path to obtain individual teeth. As shown in fig. 3, there are peak points and trough points (also the other side of the tooth is not shown in fig. 3) on the gum line between the teeth and the gums of the three-dimensional tooth model, and the peak points are the higher points (possibly the highest points or not the highest points, on the gum line) on the gum between the two teeth. The trough point is the lower point on the gum line. The peak points are matched, the peak point of the tooth gap position between every two teeth can be matched into a peak point combination, the dividing path between the two teeth can be determined through the peak point combination, and finally the teeth are divided through the dividing path, so that a single tooth is obtained. The segmentation result may be as shown in fig. 4.

According to the method, the gum line of the three-dimensional tooth model and the tooth area of the three-dimensional tooth model are obtained, the peak points of the gum line are extracted, the peak points are paired to obtain the peak point combination, then the segmentation path is determined according to the peak point combination, the tooth area is segmented according to the segmentation path, and the method for segmenting the teeth is obtained, so that each tooth of the three-dimensional tooth model can be accurately segmented, and the effect of accurately segmenting the three-dimensional tooth model is achieved.

As an alternative example, acquiring a tooth region of a three-dimensional tooth model includes: and (3) carrying out segmentation processing on the three-dimensional tooth model based on the gum line to obtain a tooth region of the three-dimensional tooth model.

Optionally, in this embodiment, before the three-dimensional tooth model is divided, the three-dimensional tooth model may be divided by using a gum line, and after the division, a tooth area of the three-dimensional tooth model is obtained. The three-dimensional tooth model is segmented, so that the gingival area of the three-dimensional tooth model can be hidden or not processed, and the influence of the gingival area on tooth segmentation accuracy can be avoided when the three-dimensional tooth model is segmented later.

As an alternative example, before acquiring the gum line of the three-dimensional tooth model, the method further comprises: determining an initial orientation of the three-dimensional tooth model; the three-dimensional tooth model is adjusted from an initial orientation to a target orientation.

Optionally, in this embodiment, after obtaining the scan data of the user to obtain the three-dimensional tooth model of the user, the direction of the three-dimensional tooth model may be adjusted. The direction of the three-dimensional tooth model is adjusted in order to "straighten" the three-dimensional tooth model, so that for all three-dimensional tooth models, a fixed orientation can be adjusted, making curvature recognition more accurate. For example, the initial orientation may be different for all three-dimensional tooth models. If the desktop is taken as a horizontal plane, the direction of the upward normal to the desktop may be taken as the target orientation. Then, the three-dimensional tooth model is adjusted to the target orientation, which may be the tooth up state. After the three-dimensional tooth model is obtained, the orientation of the three-dimensional tooth model can be adjusted to the target orientation, so that the directions of all the obtained three-dimensional tooth models can be unified.

As an alternative example, determining the initial orientation of the three-dimensional tooth model includes: determining a minimum directional bounding box of the three-dimensional tooth model under the condition that the three-dimensional tooth model is a non-closed model; an initial orientation of the three-dimensional tooth model is determined from the minimum directional bounding box.

Alternatively, in this embodiment, the three-dimensional tooth model may be divided into two types, the first being a closed model and the second being a non-closed model. The closed model is formed by adding a plane as a bottom surface after stretching the edge.

For non-closed three-dimensional tooth models, a minimum directional bounding box may be used to enclose the three-dimensional tooth model. The two surfaces with the largest area of the minimum directional bounding box are the tooth facing side and the gingival part far away from the tooth of the three-dimensional tooth model. The direction of one normal vector can be selected as the target orientation from the two normal vectors of the two largest faces of the smallest directional bounding box.

As an alternative example, determining an initial orientation of the three-dimensional tooth model from the minimum directional bounding box includes: confirming the preset axial direction according to the two surfaces with the largest area of the minimum directional bounding box; acquiring a first average coordinate value of a boundary edge of the three-dimensional tooth model and a second average coordinate value of a vertex of a polygonal surface patch of the three-dimensional tooth model based on a preset axis direction; and determining the initial orientation according to the first average coordinate value and the second average coordinate value.

Alternatively, in this embodiment, the preset axis may be a preset direction, such as a coordinate axis of the three-dimensional space or other directions. And taking any one direction as a preset direction, wherein the direction of a preset axis is the same as the preset direction.

After determining the preset axis, a first average coordinate value of the boundary edge of the three-dimensional tooth model and a second average coordinate value of all vertices may be obtained. The boundary edge is a boundary between the tooth and the gum, the first average coordinate value of the boundary edge can be understood as an average value of the dividing points of the tooth and the gum, and the second average coordinate value can be understood as a center of gravity of the three-dimensional tooth model. The center of gravity of the three-dimensional tooth model is offset to one side of the tooth region. Therefore, of the two largest-area faces of the smallest directional bounding box bounding the three-dimensional tooth model, the face closer to the second average coordinate value is the side on which the tooth is located. Thus, the initial orientation of the three-dimensional tooth model can be determined.

As an alternative example, determining the initial orientation of the three-dimensional tooth model includes: under the condition that the three-dimensional tooth model is a closed model, determining a polygon surface patch group to which a polygon surface patch on the surface of the three-dimensional tooth model belongs, wherein the plane included angle of any two polygon surface patches in the same polygon surface patch group is smaller than a first threshold value; the initial orientation of the three-dimensional tooth model is determined from the sum of the areas of the polygonal patches in the set of polygonal patches.

In this embodiment, the closed model may be obtained by filling a non-closed model, or may be a closed three-dimensional tooth model obtained by direct interface scan when performing an oral cavity internal scan operation. The closed three-dimensional tooth model may determine the initial orientation by calculating the area of the set of polygonal patches that the polygonal patches make up.

As an alternative example, in the case where the three-dimensional tooth model is a closed model, determining the polygon group to which the polygon patch of the three-dimensional tooth model surface belongs includes: selecting any one first panel from all the panels which are not divided into the polygonal panel groups as a panel in one polygonal panel group, and determining the polygonal panel with the plane normal included angle smaller than a first threshold value with the first panel as a panel in the same polygonal panel group with the first panel, wherein in the initial condition, all the polygonal panels on the three-dimensional tooth model are not divided into the polygonal panel groups; and selecting any one target patch from all patches which are not divided into the polygonal patch groups as the patch in the other polygonal patch group, and determining the polygonal patch with the plane included angle smaller than the first threshold value with the target patch as the patch which is positioned in the same polygonal patch group with the target patch until all the polygonal patches are divided into the polygonal patch groups.

As an alternative example, in the case where the three-dimensional tooth model is a closed model, determining the polygon group to which the polygon patch of the three-dimensional tooth model surface belongs includes: selecting any one first panel from all the panels which are not divided into the polygonal panel groups as a panel in one polygonal panel group, and determining the polygonal panel with the normal line and the normal line of the first panel smaller than a first threshold value as a panel which is positioned in the same polygonal panel group with the first panel, wherein in the initial condition, all the polygonal panels on the three-dimensional tooth model are not divided into the polygonal panel groups; selecting any one target patch from all the patches which are not divided into the polygonal patch groups as the patch in the other polygonal patch group, and determining the polygonal patch with the normal line smaller than the first threshold value as the patch which is positioned in the same polygonal patch group with the target patch until all the polygonal patches are divided into the polygonal patch groups.

In this embodiment, the polygonal patches may be grouped to obtain different polygonal patch groups, where the surface angles between the polygonal patches in the polygonal patch groups are smaller than a first threshold.

When dividing the polygon panel group, any one of the polygon panels may be used as one of the polygon panels in one polygon panel group, and then, the polygon panels with the surface included angle smaller than the first threshold value or the surface included angle with the normal line and the normal line of the polygon panel in all the polygon panels are added to the polygon panel group based on the polygon panel. The included angle of each of the remaining polygonal patches with the base patch is greater than or equal to the first threshold. This operation is repeated until all the polygonal patches belong to one polygonal patch group. The polygon patches in each polygon patch group may be considered to lie on the same plane. And calculating the sum of areas of the polygonal patches in the polygonal patch group, wherein the polygonal patch in the polygonal patch group with the largest sum of areas is regarded as one side of the gum of the closed model, and the other side is one side of the tooth, so that the initial orientation of the three-dimensional tooth model is determined.

As an alternative example, acquiring a gum line of a three-dimensional tooth model includes: extracting characteristics of the three-dimensional tooth model; and identifying the characteristics to obtain the gum line.

In this embodiment, a neural network model may be used to extract the gum line of the three-dimensional tooth model. After the three-dimensional tooth model is input and input into the neural network model, the neural network model extracts the characteristics of the three-dimensional tooth model, convolves and pools the characteristics, and outputs the gum line obtained by recognition.

As an alternative example, extracting the peak point of the gum line includes: determining a coordinate value of each point on the gum line based on the target orientation; and determining the peak point from the gum line according to the coordinate value, wherein the coordinate value of the peak point is larger than the coordinate value of the adjacent point of the peak point on the gum line.

In this embodiment, since the orientation of the three-dimensional tooth model is adjusted to the target orientation, for a point on the gum line, if the coordinate value in the target direction is greater than that of the adjacent point, the point is referred to as a peak point.

If the distance between two adjacent peak points is too close, one point with larger coordinate value can be taken as the peak point, and the other point is not taken as the peak point, because the peak point is a tooth gap area between two teeth, the two peak points cannot be too close, and if the distance between the two peak points is too close, the fact that the peak points are wrongly determined is possibly indicated.

As an alternative example, pairing peak points, obtaining a peak point combination includes: dividing the peak points into a first peak point group and a second peak point group according to the position relation between the peak points and the three-dimensional tooth model; and combining each peak point in the first peak point group with a second peak point in the second peak point group as a group of peak points, wherein the second peak point is the closest peak point to the first peak point in the second peak point group, and the angle between the connecting line of the first peak point and the second peak point and the dental centerline of the three-dimensional tooth model is larger than a third threshold value.

Optionally, in this embodiment, when pairing peak points, two peak points closest to each other and whose connecting line passes through the dental centerline are paired into a pair of peak point combinations. The midline of a tooth is the line connecting the midpoints of the teeth on the three-dimensional tooth model. The connection of the peak points indicates that the peak points on two sides of the tooth are matched, but the peak points on one side of the tooth are not matched.

As an alternative example, after each peak point in the first peak point group is taken as a first peak point, and the first peak point is combined with a second peak point in the second peak point group as a set of peak points, the method further includes: deleting successfully matched peak points from the first peak point group and the second peak point group; and combining each peak point in the second peak point group with a fourth peak point in the first peak point group as a group of peak points, wherein the fourth peak point is the closest peak point to the third peak point in the first peak point group, and the angle between the connecting line of the third peak point and the fourth peak point and the dental centerline of the three-dimensional tooth model is larger than a third threshold value.

In this embodiment, pairing may be performed one by one from the peak points on one side of the tooth of the three-dimensional tooth model, for each peak point on one side of the tooth of the three-dimensional tooth model, the peak point with the smallest distance is selected from the other side of the tooth, and the included angle between the connecting line of the two peak points and the tooth center line meets the preset threshold condition, so that the two peak points are combined into one peak point combination, after pairing is completed, all the successfully paired peak points are deleted, and then pairing is performed again from the other side of the tooth, so as to achieve pairing of all the peak points.

Furthermore, if the peak point with the minimum distance is found only according to the distance, the pairing is easy to be wrong under the conditions of oblique teeth or incisors, and the pairing method can avoid the pairing mistake under the scene by adopting a mode that whether the included angle meets the preset threshold condition or not.

There may be various cases of pairing failure, such as that the distance between two peak points is too large, or that one peak point on one side and two peak points on the other side can be paired.

As an alternative example, determining the segmentation path between teeth in the tooth region based on the peak-point combination comprises: determining each peak point combination as a current combination; the shortest distance of two peak points in the current combination on the three-dimensional tooth model is determined as a segmentation path.

After the peak point combinations are determined, the split paths may be determined in accordance with the peak point combinations. The split path may be the shortest path between two peak points on the tooth.

As an alternative example, determining the segmentation path between teeth in the tooth region based on the peak-point combination comprises: determining each peak point combination as a current combination; determining the shortest distance of two peak points in the current combination on the three-dimensional tooth model as a first path; correcting the first path in a curvature superposition mode to obtain a target path; the target path is determined as a split path.

In this embodiment, the divided path may be used as a path for dividing the teeth, or may be adjusted, and after the shortest path between two peak points on the teeth is obtained, the path may be corrected by using a superimposed curvature, and the corrected path may be used as the divided path.

As an alternative example, correcting the first path by superimposing curvatures, obtaining the target path includes: each point on the first path is used as a current point, and a replacement point of the current point is determined for the current point in a preset range on the three-dimensional tooth model; wherein the replacement point is a point meeting the curvature threshold requirement; and taking the connecting line of the replacement points as a target path.

In this embodiment, the shortest path is corrected by using a mode of overlapping curvatures, that is, the points on the path are locally adjusted according to the curvatures, and each point on the shortest path is corrected. And taking each point as a current point, and if points meeting the curvature threshold requirement exist in a preset range near the current point on the three-dimensional tooth model, replacing the current point by using the points meeting the curvature threshold requirement. The replaced points are connected and then used as dividing paths.

The above three-dimensional tooth model segmentation method is described below with reference to an example.

The three-dimensional tooth model segmentation method can be mainly applied to a scene of correction display, and after tooth segmentation, the tooth positions are adjusted according to a correction scheme, and corrected teeth are generated and displayed. Further, the tooth segmentation technique can also be applied to the design of orthodontic guide plates, the design of dental models and the design of temporary teeth.

In this embodiment, if the teeth of the user are to be inspected or orthodontic, the oral cavity of the user may be scanned first to obtain data of the three-dimensional tooth model of the teeth of the user. The data may be displayed to the user and to the doctor via a display screen so that the doctor may indicate to the user the condition of the teeth.

The three-dimensional tooth model of the user resulting from the oral scan may be a closed or non-closed model. The three-dimensional tooth model includes a tooth region and a gum region, and the boundary between the tooth region and the gum region is not marked. Thus, if teeth are to be divided, a dividing line (gum line) between the teeth and gums and a dividing path between the teeth are determined.

That is, in this scheme, the scan data output by the 3D scanner is acquired; automatically aligning the model based on the identified model features; extracting characteristic values of the three-dimensional dental model through a curvature geometric algorithm based on the automatically-laid model, filtering and denoising the extracted characteristic values to obtain a gum line and a peak point thereof, and dividing a tooth part and a gum part to obtain a pure tooth part; and matching the peak points of the gum line, and determining the dividing paths between teeth by a curvature and shortest path method to obtain all the divided teeth.

In one embodiment, the three-dimensional tooth model segmentation method can be applied to an orthodontic full-automatic production process. Firstly, a digital three-dimensional model of a patient is obtained, the segmentation method is carried out on the digital three-dimensional model to obtain a single tooth model, and then, the malpositioned teeth are automatically arranged, and a plurality of sets of orthodontic tooth models are generated. And simultaneously, positioning parts and identification information are arranged on the sets of orthodontic tooth models. Printing the multiple sets of orthodontic tooth models, and pressing the preheated polymer material film on the dental model to form a shell-shaped film after printing. Shooting the bottom of a dental model (with a diaphragm attached to the surface of the dental model) through an image acquisition terminal (CCD), identifying identification information through an OCR (optical character recognition) technology, sending the identification information to a server, and calling a corresponding marking instruction or cutting instruction in a database according to the identification information by the server. When a marking instruction is executed, determining a 3D label to be printed; determining boundary constraint of the 3D label, and determining a label area from the bottom plate according to the boundary constraint, wherein the label area is used for printing the 3D label; and engraving the 3D label into the label area. When a cutting instruction is executed, the image acquisition terminal acquires an image of the bottom of a product to be cut; identifying an identification code in an image acquired by the image acquisition terminal, and transmitting the identification code or identification information contained in the identification code to the receiving terminal; the receiving terminal matches with the corresponding cutting instruction, and the cutting execution terminal performs cutting operation on the product to be cut according to the cutting instruction. The bottom surface of each product to be cut is provided with a corresponding identification code, different products correspond to different identification codes, the product to be cut is arranged above the image acquisition terminal, the identification codes in the images can be identified by acquiring the images of the bottoms of the products to be cut, the identification codes correspond to corresponding cutting instructions, and the cutting execution terminal executes corresponding cutting operations according to the cutting instructions to remove redundant parts so as to finally obtain the product.

It should be noted that, the above scheme for automatically arranging the dislocated teeth may be: establishing a tooth arrangement coordinate system; defining single tooth characteristic points and establishing a tooth local coordinate system; on the basis, the position and the posture of each tooth in the dentition are analyzed from a low-dimensional angle, the coordinate translation quantity and the local coordinate axis rotation quantity of the teeth are calculated respectively by adopting a weighted fitting optimization method, the association constraint of the tooth pose and a space dentition curve is formed, the collision detection method of a rectangular bounding box is combined, the tooth pose is adjusted in the space dentition curve constraint range by designing an iterative algorithm based on the steepest descent method, and the automatic arrangement of the teeth is completed.

FIG. 5 is an alternative flow chart of the present embodiment, which divides the three-dimensional tooth model into mainly the following steps:

1 import model

The imported model is the imported three-dimensional tooth model. The three-dimensional tooth model is a model obtained by oral scanning. The three-dimensional tooth model is imported into a system, which can display the construction of the three-dimensional tooth model through a display screen.

The three-dimensional tooth model may be a scan model (non-closed model) output by a 3D scanner, and the input three-dimensional tooth model may be in any direction. In this scenario, any other tooth model type with a plane does not affect the implementation of the present invention.

A three-dimensional tooth model is a digitized three-dimensional body made up of a series of polygonal patches.

2 model automatic centering

The three-dimensional tooth model may have a different orientation after being introduced into the model. Thus, the three-dimensional tooth model can be aligned by adjusting the direction of the three-dimensional tooth model to the target direction.

Aiming at the imported three-dimensional tooth model, the characteristics are identified, the purpose of identifying the characteristics is to find the alignment angle of the three-dimensional tooth model, the characteristics are inconsistent due to different application types of the different tooth models, and the characteristics can be classified according to the different tooth model types through extraction of the characteristics. When the model is to be adjusted, the initial orientation of the three-dimensional tooth model needs to be determined and then adjusted to the target orientation.

There are various ways to determine the initial orientation. For example, the number of the cells to be processed,

1) If the model is a mouth scan model and such model is a non-closed model, the directional containing box (Oriented Bounding Box, OBB) technique is first applied to determine the containing box of the containing model, and the direction of one normal of the two sides with the largest area of the 6 sides of the containing box is determined as the final tooth orientation. The method is to determine the size and direction of the box according to the geometry of the object, and the box is not required to be perpendicular to the coordinate axis. This allows the most compact containing box to be selected as best suited.

After the containing box is obtained, a first average coordinate value of the boundary edge of the three-dimensional tooth model and a second average coordinate value of the vertex of the polygonal surface piece of the three-dimensional tooth model are obtained based on the preset axial direction. Then, the side of the face with the second average coordinate value closer to the face is determined as the tooth facing side, and the initial facing of the three-dimensional tooth model is determined.

The containment box is rotated by a rotation matrix according to the angle of the target, so that the model is rotated to the target orientation.

2) If the model is a closed model with a flat bottom surface, the model is straightened according to the mode that the largest flat bottom surface of the model is attached to the bottom, and the method for detecting the largest plane of the teeth comprises the following steps: setting a certain polygonal surface patch, superposing the set polygonal surface patch and a 3D tooth model formed by the polygonal surface patch, setting an error threshold e, and considering that the set polygonal surface patch is uneven with the surface patch on the 3D tooth model when the error threshold e is larger than a certain value; otherwise, the two parts are considered to be in the same plane. When the set polygonal surface patch is on the same plane with a certain surface patch of the tooth model, the set polygonal surface patch and the certain surface patch are overlapped together, and the next polygonal surface patch is continuously searched and an error threshold value is judged. The above steps are cycled until a plane with the largest tooth model is obtained. And determining the current normal vector of the plane with the largest tooth model, and acquiring the target normal vector of the plane with the largest tooth model.

According to a cross multiplication method, a rotation angle and a rotation axis are solved according to vector values before and after rotation (the rotation angle and the rotation axis are obtained by solving a current normal vector and a target normal vector by using the cross multiplication method), the cross multiplication method is a binary operation of vectors in a vector space, and the operation result is a vector instead of a scalar; by the above-mentioned rotation angle and its rotation axis, an arbitrary model can be rotated to a desired spatial position.

3 extracting gum line based on feature recognition

The purpose of identifying the gum line from which the three-dimensional tooth model is extracted is to divide the three-dimensional tooth model according to the gum line. The gum line may be determined in a number of ways. For example, it may be determined by a neural network model into which the three-dimensional tooth model may be input, the gum line of the three-dimensional tooth model being annotated by the neural network model. Alternatively, a three-dimensional tooth model may be projected onto a two-dimensional plane, and then the tooth area on the two-dimensional plane is identified, and the boundary between the tooth and the gum, i.e., the gum line, is marked on the corresponding three-dimensional tooth model.

Taking a neural network model as an example to identify a gum line, placing the model to a designated position, and then carrying out filtering and noise removal on the characteristic values extracted from the tooth model according to a geometrical calculation method of curvature to obtain a three-dimensional tooth model outline; the curvature plan calculation method of (a) is to indicate the degree of surface roughness, which is also called a feature in this embodiment, with respect to the rotation rate of the tangential angle to the arc length of a certain surface on the dental cast. The characteristic value of the concave-convex area of the dental model (the true gum line on the dental model is also reflected by concave-convex) can be obtained through a curvature method;

To obtain the initial model feature value, the noise needs to be removed, namely, the noise is uneven in other parts of the teeth except the gum line area, and the part is also considered as a feature, which affects the subsequent fitting with the contour line, so that the model feature value needs to be filtered or removed, and finally, the optimal feature value is obtained.

The characteristic value is simplified into a contour line, namely an initial gum line, and the initial gum line is processed by adopting a principal component analysis method to obtain a point in a main direction because the initial gum line is not a smooth curve and has the conditions of local folding and deviation. By using the data to determine if the principal directions are consistent, such as where the curve bends at the peaks and troughs are redirected, the points of the principal directions can be determined. Aiming at the situation, the main contour shape is obtained by using a principal component analysis method, the points of the main contour direction are determined, the unsmooth area of the line segment is avoided, and the finally required smooth gum line is obtained.

4 dividing gums and teeth based on gum line

After the gum line is identified, the tooth portion and gum portion of the three-dimensional tooth model may be segmented according to the location of the gum line. The purpose of dividing the tooth portion and the gum portion is to make the tooth division more accurate.

5 obtaining and pairing peak points based on gum line

The above-mentioned points determining the principal direction of the contour, i.e. the peak-valley points, are determined. According to the scheme, the insertion control point is determined according to the curvature bending degree of the gum line (the wave crest and the wave trough point are all bending areas), and the wave crest point can be judged and the wave trough point can be filtered according to the up-down distance (the difference of the heights in the target direction) because the wave trough point is always lower than the wave crest point, so that the complete wave crest point is obtained.

After the peak points are obtained, the two peak points which are shortest in distance and pass through the dental midline through the connecting line are paired to serve as the starting point and the ending point of tooth segmentation. For example, as shown in fig. 6, the peak points on both sides of the tooth are paired. (one peak-point combination of one dotted line connection in fig. 6, not all dotted lines are shown).

6 obtaining a tooth segmentation curve according to the curvature threshold and the minimum path based on the point pairing

Because the teeth are contacted and connected before the teeth, the teeth are cut through curves to form independent single teeth, the scheme firstly adopts a shortest distance mode to determine an initial cutting curve, the shortest distance is from a matched initial peak point, walks along the surface of the triangular surface patch of the dental model, and reaches a final peak point by the shortest distance. Since the shortest distance is not the best cutting line, the shortest distance is corrected by superimposing the curvature, and since the contact between teeth is uneven, the shortest distance dividing line is superimposed and established by two methods within a range of large curvature. The determined segmentation path is shown in fig. 7.

7 dividing the teeth according to the dividing curve

And dividing all teeth according to the dividing curve to obtain divided teeth.

Optionally, as shown in fig. 8, the gum line extraction method may include the steps of:

step S222, a target tooth model is obtained, wherein the target tooth model comprises a plurality of polygonal patches;

step S224, determining edge categories of edges of the plurality of polygonal patches, wherein the edge categories comprise tooth edges and gum edges;

step S226, determining the gum line of the target tooth model according to the edge type of the edge.

The target tooth model may be an oral scan model, which refers to a digital three-dimensional model including teeth and gum parts of a user generated by scanning the inside of the oral cavity of the user; of course, the digital three-dimensional model can be obtained by taking an impression of the tooth and gum portions and then scanning the impression. The data of the three-dimensional tooth model is displayed on a computer or a server, and the shape or the style of the three-dimensional tooth model can be displayed through a display screen, so that a doctor can watch the three-dimensional tooth model conveniently.

The polygonal surface piece is a surface piece constituting the surface of the target tooth model (the tooth portion and the non-tooth portion, the surface of the entire model), and may specifically be a triangular surface piece, a quadrangular surface piece, a pentagonal surface piece … … surface piece, or the like. Each of the polygonal patches may lie in one plane and different polygonal patches may lie in the same or different planes. The surface of the target tooth model is composed of a plurality of polygonal surface pieces, each polygonal surface piece comprises a plurality of edges, and two adjacent polygonal surface pieces share one edge. The end points of the edges are the vertexes of the corresponding polygonal surface pieces, and one vertex can be shared by a plurality of polygonal surface pieces.

It should be noted that the target tooth model includes a tooth portion and a gum portion, and the gum line is a boundary between the tooth portion and the gum portion. The sides constituting the polygonal face sheet may be located at the tooth portion or the gum portion, and thus may be classified into tooth sides and gum sides. Based on this, in the present embodiment, the gum line of the target tooth model can be determined according to the edge class of the edge of the polygonal patch, and the determined gum line is used to segment the tooth portion and the gum portion of the target tooth model.

In this embodiment, after the target tooth model is obtained, the edge type of the edge of the polygonal surface patch on the target tooth model may be determined by determining the edge type of the edge, so as to determine the gum line of the target tooth model, and the edge serving as a component of the polygonal surface patch may further refine the target tooth model, so as to improve the accuracy of gum line extraction of the tooth model.

As an alternative example, determining the edge class of the edges of the plurality of polygonal patches includes: acquiring target characteristics of edges of a plurality of polygonal patches; based on the target features, edge categories are determined for edges of the plurality of polygonal patches.

In this embodiment, the target feature includes a geometric feature, however, in this embodiment, the target feature may also include a non-geometric feature, which is not specifically limited herein. In this embodiment, the edge class may be determined by a method of extracting the target feature of the edge of the polygonal patch and identifying the target feature. It should be noted that, when the target feature includes a geometric feature, the edge classification may consider the topological structure relationship of the space, so as to improve the accuracy of gum line identification.

In this embodiment, the target features of all sides on the target tooth model may be extracted, or some of the sides may be selected. The edge class of the edge is determined by identifying the target feature of the extracted edge.

As an alternative example, the target characteristics of the edge include at least one of: the method comprises sharing a dihedral angle between a first polygonal face sheet and a second polygonal face sheet of the edge, a first curvature value of a first vertex of the edge, a second curvature value of a second vertex of the edge, a first distance from a vertex of the first polygonal face sheet away from the edge to the edge, a second distance from a vertex of the second polygonal face sheet away from the edge to the edge, a first diagonal angle of the edge in the first polygonal face sheet, and a second diagonal angle of the edge in the second polygonal face sheet. In one application scenario, the target features of the edge include the seven items described above.

As another alternative example, the target features of the edge may include at least one of the following features: the first curvature value of a first vertex of the edge, the second curvature value of a second vertex of the edge, the first spatial coordinates of the first vertex, the second spatial coordinates of the second vertex, the length of the edge, the angle of the edge to an adjacent edge, the first distance of a vertex of the first polygon away from the edge to the edge, the second distance of a vertex of the second polygon away from the edge to the edge, the first diagonal of the edge in the first polygon, the second diagonal of the edge in the second polygon, the first normal of the first vertex, the second normal of the second vertex, the normal of the first polygon, the normal of the second polygon.

The dihedral angle between the first polygonal surface piece and the second polygonal surface piece sharing one side is the included angle between the two polygonal surface pieces sharing one side. The first vertex and the second vertex are two end points of the edge, and the curvature value is a value of curvature of the surface piece corresponding to the vertex at the vertex. The curvature is defined by differentiation for the rotation rate of the tangential angle to the arc length at a point on the curve, indicating the extent to which the curve deviates from a straight line. A value indicating the degree of bending of the curve at a certain point. The greater the curvature, the more curved the curve at the corresponding point is at that point. The above-mentioned space coordinates are coordinates in a three-dimensional rectangular coordinate system where the vertex is located. The three-dimensional rectangular coordinate system may be a coordinate system in which the target tooth model is located, for example, a plane in which a base plate of the target tooth model is located may be defined as an xOy plane, and a normal direction of the plane is defined as a z-axis. The direction of the coordinate axes may be predetermined. For example, the gingival plane of the target tooth model is defined as the plane on which the X-axis and the Y-axis lie, and the tooth direction is defined as the Z-axis direction. The length of the edge is the distance between two endpoints of the edge; because there may be multiple adjacent edges of an edge, the angle between an edge and an adjacent edge may be multiple, between 0-180 degrees; when the polygon is a quadrangle or more, the distance from the vertex to the side of the polygon can be more than one, and the distance from each vertex to the side can be different; similarly, there may be one or more diagonal corners of the sides in the polygonal panel, where there are multiple, the angle of each diagonal corner may be different. For example, as shown in fig. 9, the polygonal panels are triangular panel 302, quadrangular panel 304, and pentagonal panel 306, respectively, and each side of the three polygonal panels corresponds to at least one or at least two of the features described above.

Specifically, when extracting features, finding out a polygonal patch corresponding to the edge of the target tooth model; the included angle between the normals of the two polygonal patches can be obtained through cross multiplication according to the normals of the polygonal patches, the included angle between the dihedral angles of the two polygonal patches and the normals is complementary, and the dihedral angles of the two polygonal patches can be obtained through subtracting the included angle between the normals by 180 degrees; obtaining curvature values of the two vertexes according to a curvature geometric calculation method or other methods; the distance from the vertex far away from the edge to the edge of the polygonal surface piece can be obtained according to the point-to-straight line distance; and obtaining two vectors according to the directions of the two sides, and obtaining the included angle of the two sides at the vertex of the polygonal surface piece according to the two vectors through dot multiplication, so as to obtain the diagonal angle of the sides in each polygonal surface piece. It should be noted that, the spatial coordinates corresponding to each vertex and the proximity relations of the points, sides and faces are known, and each target feature can be obtained based on the known spatial coordinates.

As an alternative example, determining the edge class of an edge based on the target feature includes: carrying out dimension lifting on the target features of the initial dimension of the edge to obtain first features of the first dimension; performing dimension reduction on the first feature to obtain a second feature of the target dimension; the edge is determined to be a tooth edge or a gum edge according to the magnitude of the value of the second characteristic. That is, in this embodiment, the target dimension may be two-dimensional, corresponding to the tooth side and gum side, respectively. Of course, in other embodiments, the dimensions may be multi-dimensional, as long as the edge class of the corresponding edge can be determined, which is not limited herein.

Alternatively, in this embodiment, after extracting the target feature of the edge, the initial dimension of the target feature may be determined according to the kind, number, and the like of the extracted target feature. The method comprises the steps of firstly carrying out dimension lifting on target features, lifting the target features to a first dimension to obtain first features, then carrying out dimension reduction on the first features, and reducing the first features to the target dimension, wherein the first features after dimension reduction are not the target features but the second features of the target dimension because the dimensions are lifted and then reduced, so that the edge types of the corresponding edges can be obtained.

Taking the initial dimension as seven dimensions as an example, the seven-dimensional target features of the corresponding edge are input into the edge classification network to classify the corresponding edge. Specifically, firstly 7 dimensions are input, 7 geometric features are respectively corresponding to the 7 dimensions, the dimension of the target feature is increased to N dimensions through one convolution operation, secondly, the dimension of the target feature is kept unchanged through k identical convolution operations, and the number of edges is reduced through pooling operation. Then repeating the above processes, namely one dimension-increasing convolution+k dimension-same convolutions+pooling, so that the dimension of the target feature is sequentially changed into 2N, 4N and 8N, wherein in the last dimension-increasing cycle, the dimension-increasing operation is performed after k dimension-same convolutions so as to gradually recover the number of edges, then dimension-decreasing convolution is performed so that the feature dimension is reduced to 4N, then k identical convolutions are performed so as to keep the feature dimension unchanged, and then repeating the above processes, namely the dimension-increasing operation+one dimension-decreasing convolution+k dimension-increasing convolutions so that the feature dimension is sequentially changed into 4N and 2N, N, and the description is that after the last dimension-decreasing cycle, the feature dimension is further directly reduced to 2 dimensions. The dashed arrows in fig. 10 represent feature fusion operations, and the final two-dimensional values are probability values of the classes of the tooth edge and the gum edge, and the class with the large probability value can be selected as the class of the edge, so that each edge of the model is divided into the tooth edge or the gum edge.

The above is merely an example and is not intended to be limiting. For example, the initial dimension is not limited to 7 dimensions, the number of cycles in the process of increasing and decreasing the dimension is not limited, and N is an integer not less than 7, such as 7, 8, 9, 10, etc., and in a specific embodiment, the number of cycles can be adjusted according to practical situations.

It is further noted that the convolution and pooling in this step is different from the convolution pooling operation of the image neighborhood, which is based on edges in the target tooth model, as shown in fig. 11.

In order to eliminate the influence of different edge sequences (such as (a, b, c, d) and (c, d, a, b)) in the convolution operation, the original edge feature is processed according to the following formula (1), wherein f (a) and f' (a) respectively represent the original feature and the processed feature of the edge a, and the convolution operation definition of the processed edge e is shown in the formula (2), wherein { w } _k I k=0, 1,2,3,4} represents the weight parameters that need to be trained. The pooling operation converts 5 edges (e, a, b, c, d) into 2 edges (h, i), which are defined as (3), and the upper pooling operation is opposite to the pooling operation, converts 2 edges (h, i) into (e ', a ', b ', c ', d '), which are defined as (4).

C(e)＝w ₀ f'(e)+w ₁ f'(a)+w ₂ f'(b)+w ₃ f'(c)+w ₄ f'(d) (2)

/>

In addition, before edge classification, the classification network needs to be trained to obtain the network weight value and some super-parameter values: for example, k and N, specifically, multiple sets of geometric feature data and class values (such as gingival corresponding to 0 and tooth corresponding to 1) of the class of the corresponding edge are input into an edge classification network for training, for example, 500 sets, 1000 sets or 2000 sets of classified tooth model data are input, and each set of tooth model data contains 5000, 10000 or 20000 sets of geometric feature data and class data corresponding to the edge, so that the classification network can classify the edge according to the geometric feature data of the edge. It should be noted that, in this embodiment, in combination with the deep learning technology, the method based on edge classification can automatically identify gum line data, so as to greatly improve identification accuracy and efficiency, improve efficiency of the whole tooth diagnosis and treatment process such as stealth orthodontic, and enhance experience of users in diagnosis and treatment.

As an alternative example, determining the gum line of the target tooth model according to the edge class of the edge includes: determining the patch type of the polygonal patch according to the edge type, wherein the patch type comprises a tooth surface patch and a gum surface patch; the gum line is extracted based on the common edges of the polygonal patches that are different in patch category and adjacent.

Alternatively, in this embodiment, after determining the class of the edge, the class of the polygon patch may be determined according to the class of the edge. As will be appreciated, a dental patch is a patch that is positioned on the dental portion of a dental model, and a gingival patch is a patch that is positioned on the gingival portion. The patch category is determined according to the edge category of the edge, and the patch category of the polygonal patch may be determined according to all edges of the polygonal patch, or the patch category of the polygonal patch may be determined according to at least one edge of the polygonal patch.

As an alternative example, determining the patch class of the polygonal patch according to the edge class includes: in the polygonal surface sheet, when the number of the gingival sides is larger than the number of the tooth sides, determining the polygonal surface sheet as a gingival surface sheet; in the polygonal face piece, when the number of gum edges is smaller than the number of tooth edges, the polygonal face piece is determined to be a tooth face piece.

Alternatively, in this embodiment, the patch type of the polygonal patch is determined according to the edge types of all the edges of the polygonal patch. The number of tooth edges and gum edges in the polygonal face sheet is determined according to the number of tooth edges and gum edges in the polygonal face sheet.

Traversing all sides of the target tooth model, and marking the sides as common sides if two corresponding surface patch categories of the sides are different.

As an alternative example, extracting a gum line based on common edges of polygonal patches that differ in patch category and are adjacent includes: determining a first feature point based on the vertex of the common edge; determining a target feature point from the first feature points according to the curvature relation; and determining the gum line according to the target characteristic points.

Alternatively, in this embodiment, since the class of the polygonal patch may be determined according to the class of the edge, after the class of the patch is determined, the common edge of the polygonal patches with different classes is the boundary between the tooth patch and the gum patch. Thus, the gum line can be determined from all the common edges of the tooth and gum flaps.

In this embodiment, the first feature point may be determined from the vertex of the common edge, then the target feature point may be determined from the first feature point, and the gum line may be determined according to the target feature point. The target feature points may be calculated from the curvature.

In this embodiment, the vertex of one common edge is taken as the starting point, the connected common edge is searched, and the vertices of the connected common edges are recorded in the anticlockwise or clockwise order as the feature points. Wherein the first feature point may be all or part of the recorded feature points, in particular the first feature point is located on the initial tooth model, so that the accuracy of the extracted gum line is higher. The target feature points are feature points of which parts have specific features in the first feature points.

As an alternative example, determining the first feature point based on the vertex of the common edge includes: when the target tooth model is a model obtained by performing edge contraction processing on the initial tooth model, determining the nearest point of the vertex on the shared edge on the initial tooth model as a first characteristic point; in the case where the target tooth model is the initial tooth model, the vertex on the common edge is determined as the first feature point.

Optionally, the initial tooth model is an initially obtained tooth model. The initial tooth model may be subjected to an edge-shrinking treatment to obtain a target tooth model, or the initial tooth model may be directly used as the target tooth model. Whether or not to perform edge contraction processing on the initial tooth model affects whether or not the vertex of the common edge is directly used as the first feature point. In this embodiment, the first feature point is the vertex of the shared edge, or the closest point of the shared edge on the initial tooth model.

That is, in this embodiment, after searching for the connected common edge with the vertex of one common edge as the starting point and recording the vertices in the counterclockwise or clockwise order as the feature points, the recorded feature points may be projected back to the original model, and the point closest to the recorded feature points on the original model may be found as the new feature point.

The KDTree is a tree data structure that stores example points in k-dimensional space for quick retrieval. In this embodiment, all the recorded feature points may be traversed, and the closest point to the feature points may be found in the KDTree as a new feature point. The new feature point is the first feature point.

As an optional example, determining the target feature point from the first feature point according to the curvature relationship includes: first feature points, of which the curvature is larger than that of adjacent first feature points, are determined as target feature points.

After the first feature point is determined according to whether the initial tooth model is subjected to the edge contraction processing, the target feature point may be determined from the first feature point according to the curvature of the first feature point. Each first feature point corresponds to a curvature. In this embodiment, the first feature point with a curvature larger than that of the adjacent first feature points is used as the target feature point, that is, the point where the curve is located and is more curved is determined as the target feature point, so that the actual trend of the curve can be more represented.

After the recorded characteristic points are projected back to the original model, the curvature of the discrete points can be estimated by using a curvature geometric algorithm, and the characteristic points with large curvature are reserved as target characteristic points; the reason for retaining the characteristic points with large curvature is that the points with large curvature are often turning points, and the turning points are strong in characteristic and representative.

As an alternative example, determining the gum line from the target feature points includes: performing interpolation operation on the target feature points to obtain interpolated target feature points; and sequentially connecting the interpolated target feature points to form a gum line.

In the present embodiment, interpolation operation is performed on the target feature points, loss of feature points due to determination of the target feature points from the first feature points may be supplemented, and the target feature points may be combined into a closed curve by the interpolation operation. The closed curve is the gum line.

When the target characteristic points are interpolated, B spline curve interpolation can be carried out on the adjacent target characteristic points; all points are connected in sequence, and the formed curve is the gum line. In the interpolation of the target feature points, the interpolation is not limited to the B-spline curve, and other spline curves may be used, and are not particularly limited here.

As an alternative example, acquiring the target tooth model includes: obtaining an initial tooth model, wherein the initial tooth model comprises a plurality of polygonal patches, and the number of edges in the initial tooth model is greater than the number of edges in the target tooth model; and reducing the number of edges in the initial tooth model to a preset number according to a preset mode to obtain a target tooth model.

Alternatively, in this embodiment, the preset manner may be an edge contraction operation or an edge collapse operation. The initial tooth model is subjected to edge contraction operation, and the aim is to reduce the number of edges in the initial tooth model, so that a target tooth model is obtained, the tooth model is simplified, and the recognition rate of a gum line is improved.

There are several ways to reduce the number of edges of the initial tooth model. The edges or batches of edges may be reduced, such as one at a time, until a predetermined number or a predetermined number of times is reduced. Alternatively, the batch is reduced one at a time until the number is reduced to a preset number or the preset number of times is reduced.

Taking the edge collapse operation as an example, the initial tooth model can be a tooth model in any direction, and the initial tooth model is a digital three-dimensional body consisting of a series of vertexes and polygonal patches. After the initial tooth model is imported, the tooth model is downsampled in a side collapse mode, the polygonal surface pieces of the tooth model after downsampling are sparse, the number of the polygonal surface pieces is reduced, the number of sides is reduced, and downsampling is not stopped until the number of the designated sides is provided.

The imported initial tooth model often has a large number of edges and the number of edges is not uniform, but the number of edges needed is often specific when classified by the classification network in a subsequent step, so that the imported initial tooth model needs to be downsampled, some edges are deleted, and a specified number of edges are reserved. The determination of the number of sides may take into account the classification speed (the speed is faster with fewer sides) and the accuracy of the resulting gum line (the accuracy is higher with more sides, the loss of information from the original model is less with more sides). The main flow is as follows:

(1) Recording the sum of the distances from the vertex of each side of the initial tooth model to the adjacent surface sheets as an error value, wherein the initial value is 0; wherein, the adjacent surface piece refers to the surface piece where the vertex is located; an initial error value of 0 means that the sum of the distances of the vertex to its adjacent patches is 0 when edge contraction has not yet been performed.

(2) Calculating error values (the sum of distances between the vertexes and adjacent patches when the edges are not deleted) corresponding to the vertexes after the edges are deleted, and firstly deleting the edges with small errors according to the arrangement from small to large of the error values; when one edge is contracted into one vertex, 2 adjacent face sheets are deleted, and 1 vertex and 3 edges are deleted; there are various situations where an edge is condensed into a point, for example, two vertices of the edge may be moved to any point on the edge for deletion, and the specific position of the point is selected in this step according to the minimum error value.

(3) Repeating the steps, continuously calculating the error value and deleting the corresponding edges until the number of the edges reaches the number of the appointed edges. Thus, the adjustment of the number of edges of the initial tooth model is completed.

As an alternative example, reducing the number of edges in the initial tooth model to a preset number in a preset manner, the obtaining the target tooth model includes: calculating the sum of distances from the vertex corresponding to each side to the adjacent polygonal surface sheets of the vertex corresponding to each side in the tooth model to be processed after each side is subjected to side shrinkage processing according to each of a plurality of shrinkage modes; wherein, in the first treatment, the tooth model to be treated is an initial tooth model; determining the edge and the shrinkage mode corresponding to the minimum value in the sum of the calculated distances; and performing edge contraction treatment on the edge corresponding to the minimum value according to a contraction mode corresponding to the minimum value so as to obtain the target tooth model.

As an alternative example, performing edge contraction processing on the edge corresponding to the minimum value according to the contraction mode corresponding to the minimum value to obtain the target tooth model includes: after performing edge contraction treatment, if the number of edges in the obtained tooth model is not greater than the preset number, the obtained tooth model is a target tooth model; if the number of edges is larger than the preset number, the obtained tooth model is determined to be the tooth model to be processed, so that edge contraction processing is continued.

Optionally, in this embodiment, when performing edge shrinking processing on the initial tooth model, it is also determined which edge is to be shrunk. The process may be a cyclic process. I.e. one edge per time of the shrinkage and cycle or one batch of edges per time of the shrinkage and cycle until the dental model to be treated after the shrinkage is satisfactory. Thus, at each shrink, it is determined which edge or batch of edges to shrink at the bottom.

In this embodiment, the initial model may be used as a model to be processed, and the edges to be shrunk may be determined for the model to be processed. The determination method can be that simulation or calculation is performed first, and if a certain edge is contracted, the sum of the distances from the vertex corresponding to each edge in the contracted model to the adjacent polygonal surface patch of the vertex corresponding to the edge in the tooth model to be processed is calculated, and if the sum is minimum, the edge is contracted according to the strategy. And repeating the calculation every time the teeth are contracted, and repeating the contraction of the edges until the number of the edges of the tooth model to be processed meets the requirement of the preset number. And (5) after the edge contraction is finished, obtaining the target tooth model.

According to another aspect of embodiments of the present application, there is also provided a method of manufacturing a dental appliance, the method of manufacturing a dental appliance may include the steps of:

Step one, a digitized tooth model is obtained.

The tooth model may be the target tooth model in the aforementioned gum line extraction method embodiment. Accordingly, the dental model may be obtained in a manner consistent with that of the target dental model, i.e., may be obtained by oral scanning or by conventional impression taking, which is not limited herein.

And step two, preprocessing the digitized tooth model.

In one embodiment, the preprocessing action may include the steps of a three-dimensional tooth model segmentation method to obtain the target segmented tooth. The pre-treatment operation may also include steps in a gum line extraction method to identify the gum line, and may further include converting the gum line into a cut line for use in subsequent steps to cut the initial instrument for use with a dental appliance, or in some application scenarios may also be applied to automatic tooth segmentation, gingival crown separation, and the like. The gum line extraction method may be the same as that in the foregoing embodiments of the gum line extraction method, and details thereof are described above, which are not repeated herein.

In one embodiment, the three-dimensional tooth model segmentation method in this preprocessing may further employ the following steps:

Step S242, a two-dimensional projection image of the three-dimensional tooth model is obtained, and the two-dimensional projection image is identified to obtain a plurality of tooth areas;

step S244, confirming original seed points corresponding to a plurality of tooth areas in the three-dimensional tooth model;

step S246, expanding the original seed points in a preset range to obtain target seed points of teeth in the three-dimensional tooth model;

step S248, dividing the three-dimensional tooth model based on the target seed points of each tooth to obtain divided teeth.

The two-dimensional projection image may be an image obtained by projecting a three-dimensional tooth model onto a plane. The two-dimensional projection image includes dental regions and non-dental regions. By identifying the two-dimensional projection image, a plurality of tooth areas can be identified, thereby dividing the teeth on the three-dimensional tooth model. The method for identifying can identify the two-dimensional projection image in a machine vision or artificial intelligence mode to obtain the tooth area.

The surface of the three-dimensional tooth model is composed of polygonal facets, such as triangular facets, one triangular facet containing 3 vertices, all of which can be regarded as seed points. The original seed points are seed points corresponding to tooth areas in the two-dimensional projection image among all seed points of the three-dimensional tooth model. That is, by the primordial seed points, the area of the tooth can be initially determined on the three-dimensional tooth model.

Since the original seed point may not determine the region of the tooth accurately, the original seed point may be expanded by expanding the original seed point to obtain the target seed point. The area covered by the target seed points can be regarded as teeth on the three-dimensional tooth model, and the three-dimensional tooth model can be accurately divided by the target seed points.

After the tooth region is identified, it is understood that the location and region of the tooth has been determined on the two-dimensional projection image. It is then mapped onto the three-dimensional tooth model, and the original seed points on the three-dimensional tooth model are determined. Since the seed points are the vertices of a triangular patch of the surface of the three-dimensional tooth model, the primordial seed points can be understood as covering the area of the tooth on the three-dimensional tooth model. In order to ensure accuracy, the original seed points are further expanded to obtain target seed points, the three-dimensional tooth model is segmented according to the target seed points to obtain segmented teeth, and the teeth in the three-dimensional tooth model can be divided into single teeth.

According to the method, the original seed points on the three-dimensional tooth model are determined by identifying the two-dimensional projection image of the three-dimensional tooth model, and the original seed points are expanded to obtain the target seed points, so that the range of the teeth on the three-dimensional tooth model is marked through the target seed points, the three-dimensional tooth model can be further segmented according to the target seed points, segmented teeth are obtained, and the effect of accurately dividing the three-dimensional tooth model is achieved.

Optionally, in this embodiment, after the original seed point is obtained, the original seed point may be expanded to obtain the target seed point. The process of expanding the primordial seed points may be divided into one or more stages.

For example, in one stage, the original seed point may be expanded according to a preset curvature threshold to obtain the target seed point. The preset curvature threshold may be understood as a constraint used when expanding the primordial seed points, avoiding that the primordial seed points are expanded beyond a limit. The preset curvature threshold may include one or more curvature values, and if the preset curvature threshold includes one curvature value, the original seed point may be expanded according to the one curvature value to obtain the target seed point, and if the preset curvature threshold includes a plurality of curvature values, the original seed point may be expanded by using the first curvature value, then the expansion result of the first curvature value may be expanded by using the second curvature value, and the expansion result of the second curvature value may be expanded by using the third curvature value until all curvature values are used once.

As an alternative example, expanding the original seed point by a preset curvature threshold value to obtain a target seed point includes: expanding the original seed points according to the initial curvature threshold value to obtain first seed points; and expanding the first seed point according to a target curvature threshold value to obtain a target seed point, wherein the target curvature threshold value is obtained according to the initial curvature threshold value.

Taking a preset curvature threshold value as an example, an initial curvature threshold value and a target curvature threshold value are used for forming, firstly, the initial curvature threshold value is used for expanding the original seed point to obtain a first seed point, and then the target curvature threshold value is used for expanding the first seed point to obtain a target seed point. The initial curvature threshold value and the target curvature threshold value can be the same or different, and the target seed point is obtained through two successive expansions.

As an alternative example, expanding the original seed point by the initial curvature threshold value to obtain the first seed point includes: taking a seed point adjacent to the original seed point as a current seed point; in the case where the curvature of the current seed point is less than or equal to the initial curvature threshold, the current seed point and the original seed point are taken as the first seed point.

In this embodiment, when the original seed point is expanded using the initial curvature threshold value, the curvature of the seed point adjacent to the original seed point may be acquired. The curvature is defined by differentiation for the rotation rate of the tangential angle to the arc length at a point on the curve, indicating the extent to which the curve deviates from a straight line. A value indicating the degree of bending of the curve at a certain point. Each seed point corresponds to a curvature. By comparing the magnitude relation of the curvature and the initial curvature threshold value, whether the seed point adjacent to the original seed point can be the first seed point is determined. The original seed point may be the first seed point without alignment of the curvatures.

Here, the adjacency with reference to the following is understood to mean two vertices that form the same edge of the same triangular patch as the seed point. Such as two vertices adjacent to the original seed point, i.e., forming the same edge of the same triangular patch as the original seed point.

As an optional example, expanding the first seed point according to the target curvature threshold value, obtaining the target seed point includes: taking a seed point adjacent to the first seed point as a current seed point; and taking the first seed point and the current seed point as target seed points under the condition that the curvature of the current seed point is smaller than or equal to the target curvature threshold value.

After expanding the original seed point with the initial curvature threshold to obtain a first seed point, the first seed point may be expanded with the target curvature threshold to obtain a target seed point. And determining the curvature of the seed point adjacent to the first seed point, comparing the curvature with a target curvature threshold value, and determining whether the seed point adjacent to the first seed point is a target seed point or not through comparison of the size relationship. And expanding the original seed point by using the initial curvature threshold value and the target curvature threshold value to obtain a target seed point.

As an optional example, before expanding the first seed point according to the target curvature threshold value to obtain the target seed point, the method further includes: and taking the sum of the initial curvature threshold value and a preset value as a target curvature threshold value, wherein the preset value is a positive number.

Alternatively, in this embodiment, the initial curvature threshold value and the target curvature threshold value may be empirical values, or the initial curvature threshold value is an empirical value, and the target curvature threshold value is determined according to the initial curvature threshold value. For example, the sum of the initial curvature threshold and the preset value is taken as a target curvature threshold, namely, the target curvature threshold is obtained according to the initial curvature threshold, and the target curvature threshold is larger than the initial curvature threshold. The preset value is a preset value and can be modified according to different three-dimensional tooth models.

As an optional example, after expanding the original seed point according to the initial curvature threshold to obtain the first seed point, or expanding the first seed point according to the target curvature threshold to obtain the target seed point, the method further includes: adjusting the first seed point to be a non-first seed point under the condition that the corresponding point of the first seed point on the two-dimensional projection image does not fall into the corresponding tooth area; or adjusting the target seed point to a non-target seed point in case the corresponding point of the target seed point on the two-dimensional projection image does not fall into the corresponding dental region.

Optionally, in this embodiment, when the original seed point is expanded according to the initial curvature threshold or the first seed point is expanded according to the target curvature threshold, it is further required to check whether the expanded seed point meets the requirement, that is, whether the expansion exceeds the range, and whether the seed point of the non-tooth area is used as the first seed point or the seed point of the non-tooth area is used as the target seed point. If the curvature of the seed point adjacent to the original seed point is smaller than or equal to the initial curvature threshold value when the original seed point is expanded according to the initial curvature threshold value, the area where the corresponding point of the seed point on the two-dimensional projection image is located is also judged. If not located in the tooth region, it is indicated that the seed point has been removed from the region of the tooth of the three-dimensional tooth model, and therefore the seed point is to be taken as a non-first seed point. When the first seed point is expanded according to the target curvature threshold value, if the curvature of the seed point adjacent to the first seed point is smaller than or equal to the target curvature threshold value, the area where the corresponding point of the seed point on the two-dimensional projection image is located is also judged. If not located in the tooth region, it is indicated that the seed point has been removed from the region of the tooth of the three-dimensional tooth model, and therefore the seed point is to be considered a non-target seed point.

As an alternative example, after expanding the original seed points according to the curvature threshold value to obtain the target seed points, the method further includes, before dividing the three-dimensional tooth model based on the target seed points of each tooth to obtain the divided teeth: the method comprises the steps of obtaining a target area by expanding a plurality of tooth areas; in the target region, the target seed point is expanded by the initial curvature threshold and the height.

After the original seed points are expanded to obtain target seed points, the teeth of the three-dimensional tooth model may be segmented according to the target seed points. In addition, the target seed point may be expanded again before segmenting the teeth. That is, the second stage expansion may be performed in addition to the first stage expansion of the original seed point using the initial curvature threshold and the target curvature threshold to obtain the target seed point.

In the second stage of expansion, the tooth area on the two-dimensional projection image can be adjusted first, and the tooth area is expanded to obtain the target area. Then, the target seed point is expanded in a second stage using the initial curvature threshold and height with the target region as a constraint. The purpose of expanding the tooth area into the target area is to ensure that seed points on the teeth of the three-dimensional tooth model are marked as target seed points, so that omission is avoided.

As an alternative example, when the target seed point is expanded by the initial curvature threshold and the height, a seed point adjacent to the target seed point may be taken as the current seed point; and taking the current seed point as a target seed point under the condition that the height of the current seed point is larger than a preset standard height and the curvature of the current seed point is smaller than or equal to a target curvature threshold value.

The height may be a value of the seed point in a predetermined direction of the three-dimensional tooth model. For example, the preset direction of the three-dimensional tooth model may be taken as the Z axis, and the coordinate value of the seed point on the Z axis may be taken as the height of the seed point.

The predetermined direction may be any direction. After the three-dimensional tooth model is obtained, the orientation of the three-dimensional tooth model can be adjusted to a preset direction, so that the directions of all the obtained three-dimensional tooth models can be unified.

When the initial curvature threshold and the height are used for expanding the target seed point, the seed point adjacent to the target seed point can be used as the current seed point, and if the curvature and the height value of the current seed point meet the requirements of the initial curvature threshold and the height, the current seed point can be used as the target seed point, so that the expansion of the target seed point is completed.

As an optional example, in a case where the height of the current seed point is greater than the preset standard height and the curvature of the current seed point is less than or equal to the target curvature threshold value, the step of taking the current seed point as the target seed point includes: when the height of the current seed point is larger than the standard height, the curvature of the current seed point is smaller than or equal to the target curvature threshold value, and the corresponding point of the current seed point on the two-dimensional projection image is located in a target area, the current seed point is taken as the target seed point, wherein the target area is an area obtained by expanding a plurality of tooth areas; and taking the current seed point as a non-target seed point under the condition that the corresponding point of the current seed point on the two-dimensional projection image is positioned outside the target area.

Optionally, when the initial curvature threshold and the height are used to expand the target seed point, it is ensured that the expanded target seed point does not exceed the target area. The target area is an area obtained by enlarging the tooth area. The purpose of expanding the dental region is to also include a portion of the non-dental region in the vicinity of the dental region within the target region, which allows the target seed point to expand to the non-dental region in the vicinity of the dental region when the target seed point is expanded using the initial curvature threshold and height. Doing so may cause the expanded target seed point to cover all of the tooth area.

As an alternative example, after expanding the target seed point by the initial curvature threshold and the height, the method further includes: taking a seed point adjacent to the expanded target seed point as a current seed point; the current seed point is also taken as the target seed point.

After the target seed point is expanded by using the initial curvature threshold and the height, the expanded target seed point may be expanded again, and a seed point adjacent to the target seed point may be used as the target seed point. The purpose of this secondary expansion is also to allow the expanded target seed point to cover all of the tooth area so that the tooth is intact when the tooth can be segmented according to the expanded target seed point.

As an optional example, before expanding the original seed point within the preset range, the method further includes: expanding a plurality of tooth areas to obtain a target area; marking points of the target area outside the points on the three-dimensional tooth model as third seed points; and expanding the third seed point according to the curvature threshold.

In this embodiment, the target area is first enlarged to include all of the tooth areas and the non-tooth areas near the tooth areas. Thus, when points of the target area outside the points on the three-dimensional tooth model are marked as third seed points, none of the third seed points are points on the tooth. By expanding the third seed point, a point of the non-tooth portion (gum portion or space between teeth) near the tooth portion can be marked as the third seed point, and the boundary between the outer portion of the tooth and the tooth portion can be brought closer to the tooth portion, so that the range of the target area can be reduced, and the target area includes less non-tooth area although the target area includes non-tooth area.

After expanding the third seed point, the parting line between the tooth portion and the extradental portion on the three-dimensional dental model becomes "thin".

As an alternative example, expanding the third seed point by the curvature threshold includes: taking a seed point adjacent to the third seed point as a current seed point; and taking the current seed point as a third seed point when the curvature of the current seed point is larger than the curvature threshold value.

In this embodiment, when the third seed point is expanded, the third seed point may be expanded according to the curvature and the curvature threshold. The curvature of the third seed point may be calculated by differentiating the rotation rate of the tangential angle of the point with respect to the arc length. The curvature threshold may be a preset threshold, which may be different for different three-dimensional tooth models.

As an alternative example, after taking the seed point adjacent to the third seed point as the current seed point, the method further includes: and taking the current seed point as a non-third seed point under the condition that the current seed point is positioned in the target area in the corresponding point in the two-dimensional projection image.

After expanding the third seed point, it is also checked whether the third seed point has been expanded into the target area. Because the target area contains a dental region, if the third seed point extends into the target area, it may extend into the dental region, and therefore the third seed point located in the target area is to be retracted as a non-third seed point.

As an optional example, after expanding the third seed point by the curvature threshold, the method further includes: determining a first region composed of the third seed points; determining a subarea from the first area; in the case that the sub-region is surrounded by the target seed point, the seed point in the sub-region is determined as the target seed point.

Optionally, in this embodiment, after expanding the third seed point, the third seed point forms the first area. The first region may be divided into a plurality of sub-regions. Each sub-region of the plurality of sub-regions may be understood as a region of the plurality of third seed points. If a sub-region is surrounded by a target seed point, it is indicated that the sub-region is located within a tooth portion on the three-dimensional tooth model, but the sub-region may not be a tooth, i.e. a hole portion on a tooth. Therefore, the seed point of the sub-region is to be divided into tooth portions as a target seed point.

As an alternative example, acquiring a two-dimensional projection image of a three-dimensional tooth model includes: adjusting the orientation of the three-dimensional tooth model from an initial orientation to a target orientation; and projecting the three-dimensional tooth model with the target orientation onto a target surface to obtain a two-dimensional projection image.

Alternatively, in this embodiment, when the three-dimensional tooth model is projected as a two-dimensional projection image, the orientation of the three-dimensional tooth model may be adjusted first, and the orientation may be adjusted to be the target orientation. The target orientation may be a predetermined orientation. The purpose of adjusting the orientation of the three-dimensional tooth model is to project the three-dimensional tooth model into a two-dimensional projection image, so that the tooth area on the two-dimensional projection image is more complete and less occlusion occurs. For example, the X-axis and the Y-axis of the three-dimensional coordinates may be the horizontal plane, the upward direction of the horizontal plane may be the target direction, and after the three-dimensional tooth model is acquired, the tooth direction of the three-dimensional tooth model may be oriented toward the target direction.

After adjusting the orientation of the three-dimensional tooth model, the three-dimensional tooth model may be projected onto a target surface, which may be a horizontal surface.

As an alternative example, identifying the two-dimensional projection image to obtain a plurality of tooth regions includes: the two-dimensional projection image is input into the recognition model, and a plurality of tooth regions are marked on the two-dimensional projection image by the recognition model.

Alternatively, in this embodiment, the two-dimensional projection image may be identified by the identification model, thereby allowing the tooth region to be identified. After the two-dimensional projection image is input into the recognition model, features are extracted from the recognition model and recognized, and a plurality of tooth regions are output.

As an alternative example, in a three-dimensional tooth model, identifying primordial seed points corresponding to a plurality of tooth regions includes: taking the vertex of each triangular surface in the three-dimensional tooth model as a current vertex; in the case where the corresponding point of the current vertex in the two-dimensional projection image is located in a plurality of tooth areas, the current vertex is taken as one primitive seed point.

Alternatively, in this embodiment, the surface of the three-dimensional dental model is covered by a polygonal patch, which may be a triangular patch, for example, each triangular patch having 3 vertices. Two adjacent triangular patches share one edge. In the vertices of each triangular patch, which correspond to points in the two-dimensional projection image that are located within the tooth area, the vertices may be used as primordial seed points.

As an alternative example, after segmenting the three-dimensional tooth model based on the target seed points of each tooth, the method further includes: the segmented teeth are ranked.

As an alternative example, ordering the segmented teeth includes: taking the average value of the tooth midpoints of all the teeth as a starting point, and taking the tooth midpoint of each tooth as an end point to form a vector of each tooth; taking the straight line of two teeth with the largest distance between tooth midpoints as a target straight line; and sequencing all teeth according to the included angle between the vector and the target straight line.

Alternatively, the teeth in this embodiment may be sorted after being segmented. The connection line between the midpoint of each tooth and the midpoint of each tooth is used as a vector, and the teeth can be ordered by the included angle between the connection line and a target straight line formed by the teeth with the largest distance. That is, the sequence may be from the first tooth on one side of the teeth to the last tooth on the other side.

As an alternative example, after segmenting the three-dimensional tooth model based on the target seed points of the tooth, the method further comprises: smoothing the edge of the tooth to obtain a smoothed edge.

As an alternative example, smoothing the edges of the teeth to obtain smoothed edges includes: ordering vertices on the edges of teeth; taking the second vertex as the current vertex in the ordered vertices, and executing the following operation on the current vertex until the current vertex does not comprise the rear vertex: taking the center points of the front vertex and the rear vertex of the current vertex as smooth points; taking the rear vertex of the current vertex as a new current vertex, and taking the obtained smooth point as the front vertex of the new current vertex; and connecting the first vertex on the edge of the tooth with the obtained smooth point according to the sequence, so as to obtain the edge of the tooth after smoothing.

In this embodiment, the purpose of smoothing the edge of the tooth is to make the angle of the tooth and the parting line portion of the portion outside the tooth smaller, so as to avoid the gum discomfort of the user caused by the too sharp tooth model to be cut later.

In smoothing, a series of seed points form a curve among the seed points through which the edges of the teeth pass. Starting from the initial seed point, taking an average value with a second seed point to obtain a first midpoint, taking an average value of the first midpoint and a third seed point to obtain a second midpoint, and taking an average value of the second midpoint and a fourth seed point to obtain a third midpoint. The above steps are repeated until the last seed point at the edge of the tooth. All midpoints are connected in sequence to obtain the smooth edge of the tooth.

After the user's mouth scan data is obtained, the direction of the three-dimensional tooth model can be adjusted. The direction of the three-dimensional tooth model is adjusted in such a way that, when a two-dimensional projection image is projected, as many tooth parts of the three-dimensional tooth model as possible appear on the two-dimensional projection image.

When the direction is adjusted, the three-dimensional tooth model can be input, and the three-dimensional tooth model is aligned towards the positive direction of the Z axis. The alignment method may be any alignment method in the art, and is not limited herein. Taking the example that the teeth of the three-dimensional tooth model are upwards placed on the tabletop, the tabletop can be an X axis and a Y axis of a three-dimensional rectangular coordinate system, and the normal line of one side of the tabletop facing the teeth is a Z axis. And (3) aligning the three-dimensional tooth model according to the direction, and then enabling the teeth of the three-dimensional tooth model to face upwards, namely, projecting the three-dimensional tooth model onto a table to obtain a two-dimensional projection image.

After the two-dimensional projection image is obtained by projecting the three-dimensional tooth model onto the plane, the two-dimensional projection image may be identified using the identification model, so that the tooth region on the two-dimensional projection image may be identified. The identified tooth areas may be marked using boxes. The box may be a minimum axial bounding box containing the mask region or may be a beveled box. The identified tooth region may be a regular shape or an irregular shape, such as block 402 in fig. 4, that is, the identified tooth region. The dental region includes dental and partially non-dental regions. After the two-dimensional projection image is obtained, the actual area of the tooth in the three-dimensional model can be determined. The tooth region on the two-dimensional projection image corresponds to a three-dimensional tooth model and may correspond to a tooth portion on the three-dimensional tooth model. All points on the tooth part are filtered, points with intersection between the normal direction and the original model are removed, and the points remained after the filtering are used as original seed points.

For example, the surfaces of a three-dimensional dental model (dental and non-dental parts, surfaces of the entire model) are combined from triangular patches (polygonal patches such as quadrilaterals, pentagons … …, etc. are also possible). The triangular patches are not all on the same plane and have included angles with each other. The apex of the triangular patch is the seed point. Alternatively, the points left after the vertices of the triangular patch have been noise-removed are used as seed points. If the seed point on the three-dimensional tooth model is located at the corresponding point on the two-dimensional projection image in the tooth region, the seed point is taken as the original seed point.

In this embodiment, the original seed point may be expanded to obtain the target seed point. The purpose of expanding the primordial seed points is that the primordial seed points may not contain all of the tooth parts on the three-dimensional tooth model, and thus, the tooth parts of the three-dimensional tooth model are covered by the expansion.

The expansion of the original seed points is divided into multiple stages.

In the first stage, the original seed points are expanded by curvature. The original seed point is expanded, in effect, to see if there is a seed point that can be used as a target seed point with the original seed point among seed points adjacent to the original seed point.

Judging whether the seed points are adjacent or not, and judging whether the seed points are positioned on the same straight line or not.

When the original seed point is expanded by curvature, two subsequent expansions may be performed. The expansion may be performed by an initial curvature threshold and then by a target curvature threshold. The target curvature threshold is the sum of the initial curvature threshold and a preset value. Thus, for seed points adjacent to the original seed point, it is first determined whether the curvature is less than the initial curvature threshold. If the curvature is less than the initial curvature threshold, then the original seed point is taken as the first seed point. Then, a seed point adjacent to the first seed point is used as the target seed point along with the first seed point if the curvature is less than or equal to the target curvature threshold. Thereby completing the two successive expansions of the original seed points. At this point, the expansion of the first stage has not yet ended. Because the first stage of expansion, the original seed points are expanded twice in succession, there are too many points that may be expanded beyond the tooth portion. At this time, it is necessary to determine whether or not the corresponding position of the expanded point on the two-dimensional projection image falls within the tooth region. If the tooth region is dropped, it is indicated that the original seed point is not expanded beyond the tooth portion on the three-dimensional tooth model. If not, the corresponding seed point is no longer the target seed point. So far, the first stage expansion ends.

In the second stage, the target seed point of the first stage may be expanded by an initial curvature threshold and height. In the second stage, the seed points adjacent to the target seed points obtained after the expansion in the first stage can be used as seed points to be expanded, the curvature of the seed points is smaller than or equal to the initial curvature threshold value, and the height is larger than the preset standard height. Both match, then the partial seed point is also the target seed point. If one of the conditions is not met, it is not the target seed point. In one specific example, the preset standard height may be the height of the current point.

In addition, the partial seed points are located in the target region on the two-dimensional projection image in addition to the curvature and the height. The target area is an enlarged area of the tooth area. That is, on the two-dimensional projection image, the tooth area is slightly enlarged to obtain the target area. Then, in the second stage of expansion, if the expanded seed point corresponds to the target area beyond on the two-dimensional projection image, the seed point beyond the target area is not used as the target seed point, namely, the expansion is retracted. The second stage expansion is thus completed.

That is, each tooth seed point is expanded according to a given curvature threshold value, and the range of the tooth area on the two-dimensional projection image is ensured not to be exceeded, wherein the points serve as initial points of each tooth; expanding again according to the initial point according to a given (curvature threshold +0.2 to obtain a target curvature threshold), and ensuring that the range of the bounding box of the tooth area is not exceeded;

expansion is performed according to a given curvature threshold and height, and no more than the bounding box of the tooth region + the expanded range (i.e., target region) is guaranteed; each tooth boundary point extends outwardly once (to near the crevices and gum line) by a predetermined distance.

The initial seed points are not all the areas of the final teeth, and need to be spread continuously, i.e. the solution forms a spread range from a two-dimensional projection. The bounding box of the tooth region+the enlarged region refers to a range to which the seed point can be diffused, forms a diffusion range according to the two-dimensional projection image, and can control the diffusion range through algorithm parameters.

The dental region expands time-division multiple times for: the separation between adjacent teeth of partial dental model is not obvious. The final height limit when the tooth area expands is to reduce the likelihood of teeth from expanding down to the gums. The tooth boundary extends outwards once, because after being divided directly according to the curvature threshold, the divided teeth can be smaller than the original teeth by one circle, and the teeth are required to extend because of the small curvature near the gum line.

In addition to the first stage and second stage expansion, the extradental portion may also be expanded. That is, points on the three-dimensional tooth model corresponding to the outside of the target area on the two-dimensional projection image are taken as third seed points, and the third seed points are points outside of the tooth parts, such as points of gum parts or gaps between teeth. The partial seed points are to be expanded towards the tooth portion. When expanding, the expansion can be performed according to the curvature. If the curvature of the seed point adjacent to the third seed point is greater than the required curvature threshold, then the portion of the seed point together with the third seed point acts as the third seed point, i.e., the point of the portion outside the tooth. It should be noted, however, that if the expanded third seed point is located in the target area on the two-dimensional projection image, the seed point may be expanded to the tooth portion, and therefore, to be retracted, the third seed point located in the target area is regarded as a non-third seed point.

For the first stage and the second stage of expansion, a target seed point is obtained, and a third seed point is obtained after expansion, wherein the boundary of the target seed point and the third seed point can be used as the boundary of the teeth.

If a hole is included in the tooth, then an enclosure may be included between the target seed point and the third seed point, such as where the target seed point encloses a portion of the third seed point, which is identified as the third seed point due to the hole in the tooth. In this case, the point in the region surrounded by the target seed point is also taken as the target seed point.

After the expansion is completed, the teeth are segmented according to the target seed point and the third seed point, so that single teeth are obtained. Alternatively, the tooth boundaries may be smoothed prior to dividing the teeth, with the smoothed boundaries being gentler.

After dividing the teeth to obtain individual teeth, the teeth may be ordered. The mean value of the tooth midpoints of the three-dimensional tooth model, which is the midpoint of each tooth, may be connected to a plurality of vectors. Taking the straight line of the two teeth with the largest middle points as a target straight line, checking the included angles formed by the vectors and the target straight line, and sorting according to the included angles.

And thirdly, performing shaping treatment based on the pre-treated digital tooth model to obtain the solid dental model.

In an embodiment, a model of the shaped tooth is printed based on the target segmented tooth, wherein the model of the shaped tooth is used to obtain the dental appliance. I.e. the target segmented teeth are arranged together to obtain a digitized tooth model and a solid dental model is obtained based on a shaping process (also called printing). The 3D printing may be photo-cured 3D printing, such as SLA, DLP, LCD, or other 3D printing methods, such as 3DP, MJF, FDM, polyjet.

In some embodiments, the resulting solid dental model may have hollowed-out floors, fixed attachments, identification information, etc., based on the foregoing preprocessing operations.

Further, the 3D printed product further comprises a post-processing step, and the post-processing step can be selected according to the adopted 3D printing mode. For example, when a photo-curing 3D printing technique is used, the post-processing step may be selected from post-curing, cleaning, and the like.

Of course, in other embodiments, the molding method is not limited to 3D printing, and other molding methods, such as injection molding, may be used, which is not limited herein.

And fourthly, performing film pressing treatment on the solid dental model.

After the solid dental model is obtained, the preheated polymer material membrane can be subjected to membrane pressing treatment on the solid dental model to obtain a shell-shaped membrane covering the solid dental model, namely, an initial device which is not cut yet. In particular, the shell-like membrane covers at least the dental portion of the solid dental model.

In an embodiment, an image of the solid dental model may be acquired by an image acquisition terminal, such as a CCD image sensor, and the identification information may be identified by an OCR identification technology, and then a corresponding film pressing instruction may be called out in a database based on the identification information, and film pressing processing may be performed based on the dental model instruction, where the film pressing instruction may include a pressure parameter, a film preheating time, a film pressing temperature, and the like.

Fifthly, marking the shell-shaped membrane.

In the step, marking operation is mainly performed on the shell-shaped film based on the identification information added in the pretreatment operation, so that the identification is formed on the shell-shaped film, and the initial equipment with the identification is obtained.

Specifically, similar to the foregoing steps, an image of the solid dental cast may be acquired by an image acquisition terminal, such as a CCD image sensor, and then the identification information is identified by an OCR or other recognition technique, and then a corresponding marking instruction is called based on the identification information, so as to mark the shell-like film based on parameter information included in the marking instruction.

In the method for manufacturing the dental instrument, the execution of this step is not limited, that is, whether or not the marking operation is required may be selected according to actual needs.

And step six, cutting the initial equipment to obtain the dental instrument.

The initial instrument is cut in this step, primarily based on the cut line resulting from the extracted gum line transition, to cut away unwanted portions of the initial instrument, leaving only useful portions to obtain the dental instrument. In particular, the dental appliance may be a concealed appliance.

It should be noted that, in this step, the obtaining of the cutting line may be the same as that in the embodiment of the manufacturing method of the dental apparatus, and the detailed description thereof will be omitted herein. In addition, the sequence of the fifth step and the sixth step is not limited in this embodiment.

In an embodiment, similar to the previous steps, the image of the solid dental cast may be acquired by an image acquisition terminal, such as a CCD image sensor, and then the identification information is identified by an OCR or other recognition technique, and then a corresponding cutting instruction is called out based on the identification information, so as to cut the initial fixture based on the parameter information included in the cutting instruction.

According to still another aspect of the embodiments of the present application, as shown in fig. 12, there is provided a three-dimensional tooth model segmentation apparatus including: a memory 801 and a processor 803, the memory storing a computer program which, when executed by the processor, performs the three-dimensional tooth model segmentation method described above.

It should be noted that the above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments. It should be noted that the above modules may be implemented in software or hardware as a part of the apparatus in the hardware environment shown in fig. 1. The device may also include other parts such as a communication interface 805, a connection wire 807, and the like.

There is also provided, in accordance with yet another aspect of embodiments of the present application, a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the steps of any of the embodiments described above.

According to still another aspect of the embodiments of the present application, there is provided an electronic device, including a memory, a processor, a communication interface, and a communication bus, where the memory stores a computer program executable on the processor, and the memory, the processor, and the communication interface communicate through the communication bus, and the processor implements the steps of the three-dimensional tooth model segmentation method when executing the computer program.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.

In specific implementation, the embodiments of the present application may refer to the above embodiments, which have corresponding technical effects.

It is to be understood that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For a hardware implementation, the processing units may be implemented within one or more application specific integrated circuits (Application Specific Integrated Circuits, ASIC), digital signal processors (Digital Signal Processing, DSP), digital signal processing devices (DSP devices, DSPD), programmable logic devices (Programmable Logic Device, PLD), field programmable gate arrays (Field-Programmable Gate Array, FPGA), general purpose processors, controllers, microcontrollers, microprocessors, other electronic units configured to perform the functions described herein, or a combination thereof.

For a software implementation, the techniques described herein may be implemented by means of units that perform the functions described herein. The software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially or, what contributes to the prior art, or part of the technical solutions, may be embodied in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc. It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The foregoing is merely a specific embodiment of the application to enable one skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of three-dimensional tooth model segmentation, comprising:

acquiring a gum line of a three-dimensional tooth model and a tooth region of the three-dimensional tooth model;

extracting a peak point of the gum line, and matching the peak points to obtain a peak point combination;

determining a segmentation path between teeth in the tooth region based on the peak-point combination;

and carrying out segmentation processing on the tooth region according to the segmentation path to obtain the target segmented tooth.

2. The method of three-dimensional tooth model segmentation according to claim 1, wherein acquiring a tooth region of the three-dimensional tooth model comprises:

And dividing the three-dimensional tooth model based on the gum line to obtain a tooth area of the three-dimensional tooth model.

3. The method of three-dimensional tooth model segmentation according to claim 1, wherein prior to acquiring the gum line of the three-dimensional tooth model, the method further comprises:

determining an initial orientation of the three-dimensional tooth model;

and adjusting the three-dimensional tooth model from the initial orientation to a target orientation.

4. A method of three-dimensional tooth model segmentation as defined in claim 3, wherein the determining an initial orientation of the three-dimensional tooth model comprises:

determining a minimum directional bounding box of the three-dimensional tooth model in the case that the three-dimensional tooth model is a non-closed model;

determining the initial orientation of the three-dimensional tooth model from the minimum directional bounding box.

5. The method of three-dimensional tooth model segmentation according to claim 4, wherein the determining the initial orientation of the three-dimensional tooth model from the minimum directional bounding box comprises:

confirming a preset shaft direction according to the two surfaces with the largest areas of the minimum directional bounding boxes;

acquiring a first average coordinate value of a boundary edge of the three-dimensional tooth model and a second average coordinate value of a vertex of a polygonal surface patch of the three-dimensional tooth model based on the preset axis direction;

And determining the initial orientation according to the first average coordinate value and the second average coordinate value.

6. A method of three-dimensional tooth model segmentation as defined in claim 3, wherein the determining an initial orientation of the three-dimensional tooth model comprises:

determining a polygon surface patch group to which the polygon surface patch on the surface of the three-dimensional tooth model belongs under the condition that the three-dimensional tooth model is a closed model, wherein the plane included angle of any two polygon surface patches in the same polygon surface patch group is smaller than a first threshold value;

the initial orientation of the three-dimensional tooth model is determined from a sum of areas of the polygonal patches in the set of polygonal patches.

7. The method according to claim 6, wherein determining a polygon patch group to which a polygon patch of the three-dimensional tooth model surface belongs in the case where the three-dimensional tooth model is a closed model includes:

selecting any first panel from all the panels which are not divided into the polygonal panel groups as a panel in one polygonal panel group, and determining the polygonal panel with the plane normal included angle smaller than a first threshold value with the first panel as a panel which is positioned in the same polygonal panel group with the first panel, wherein in the initial condition, all the polygonal panels on the three-dimensional tooth model are not divided into the polygonal panel groups;

And selecting any one target patch from all patches which are not divided into the polygonal patch groups as the patch in the other polygonal patch group, and determining the polygonal patch with the plane included angle smaller than the first threshold value with the target patch as the patch which is positioned in the same polygonal patch group with the target patch until all the polygonal patches are divided into the polygonal patch groups.

8. The method according to claim 6, wherein determining a polygon patch group to which a polygon patch of the three-dimensional tooth model surface belongs in the case where the three-dimensional tooth model is a closed model includes:

selecting any first panel from all the panels which are not divided into the polygonal panel groups as a panel in one polygonal panel group, and determining the polygonal panel with the normal line and the normal line of the first panel being smaller than a first threshold value as a panel which is positioned in the same polygonal panel group with the first panel, wherein in the initial condition, all the polygonal panels on the three-dimensional tooth model are not divided into the polygonal panel groups;

selecting any one target patch from all patches which are not divided into the polygonal patch groups as the patch in the other polygonal patch group, and determining the polygonal patch with the normal line and the normal line of the target patch smaller than a first threshold value as the patch which is positioned in the same polygonal patch group with the target patch until all the polygonal patches are divided into the polygonal patch groups.

9. The method of claim 1, wherein the acquiring a gum line of the three-dimensional tooth model comprises:

extracting features of the three-dimensional tooth model;

and identifying the characteristic to obtain the gum line.

10. The method of three-dimensional tooth model segmentation according to claim 3, wherein extracting the peak points of the gum line comprises:

determining a coordinate value for each point on the gum line based on the target orientation;

and determining the peak point from the gum line according to the coordinate value, wherein the coordinate value of the peak point is larger than the coordinate value of an adjacent point of the peak point on the gum line.

11. The method of claim 1, wherein pairing the peak points to obtain the peak point combination comprises:

dividing the peak points into a first peak point group and a second peak point group according to the position relation between the peak points and the three-dimensional tooth model;

and combining each peak point in the first peak point group with a second peak point in the second peak point group as a group of peak points, wherein the second peak point is the closest peak point to the first peak point in the second peak point group, and the angle between the connecting line of the first peak point and the second peak point and the dental centerline of the three-dimensional tooth model is larger than a third threshold value.

12. The method of three-dimensional tooth model segmentation according to claim 11, wherein after combining each peak point in the first set of peak points as a first peak point and a second peak point in the second set of peak points as a set of peak points, the method further comprises:

deleting successfully matched peak points from the first peak point group and the second peak point group;

and combining each peak point in the second peak point group with a fourth peak point in the first peak point group as a group of peak points, wherein the fourth peak point is the closest peak point to the third peak point in the first peak point group, and the angle between the connecting line of the third peak point and the fourth peak point and the dental centerline of the three-dimensional tooth model is larger than a third threshold value.

13. The method of three-dimensional tooth model segmentation according to claim 1, wherein the determining a segmentation path between teeth in the tooth region based on the peak-point combination comprises:

determining each peak point combination as a current combination;

Determining the shortest distance between two peak points in the current combination and the three-dimensional tooth model as one segmentation path.

14. The method of three-dimensional tooth model segmentation according to claim 1, wherein the determining a segmentation path between teeth in the tooth region based on the peak-point combination comprises:

determining each peak point combination as a current combination;

determining the shortest distance between two peak points in the current combination and the three-dimensional tooth model as a first path;

correcting the first path in a curvature superposition mode to obtain a target path;

and determining the target path as the segmentation path.

15. The method of claim 14, wherein modifying the first path by overlaying curvatures to obtain a target path comprises:

taking each point on the first path as a current point, and determining a replacement point of the current point for the current point in a preset range on the three-dimensional tooth model; wherein the replacement point is a point meeting the curvature threshold requirement;

and taking the connecting line of the replacement point as the target path.

16. The method of three-dimensional dental model segmentation according to any one of claims 1-15, wherein the step of acquiring a gum line of the three-dimensional dental model comprises:

obtaining a target tooth model, wherein the target tooth model comprises a plurality of polygonal patches;

determining edge categories of edges of the plurality of polygonal patches, wherein the edge categories include tooth edges and gum edges;

determining a gum line of the target tooth model based on the edge class of the edge.

17. The method of three-dimensional tooth model segmentation according to claim 16, wherein the determining an edge class of an edge of the plurality of polygonal patches comprises:

acquiring target features of edges of the plurality of polygonal patches, wherein the target features comprise geometric features;

and determining the edge category of the edge according to the target characteristics.

18. The method of three-dimensional tooth model segmentation according to claim 17, wherein the determining the edge class of the edge from the target feature comprises:

carrying out dimension lifting on the target features of the initial dimension of the edge to obtain first features of a first dimension;

performing dimension reduction on the first feature to obtain a second feature of the target dimension;

And determining that the edge is a tooth edge or a gum edge according to the magnitude of the value of the second characteristic.

19. The method of three-dimensional tooth model segmentation according to claim 16, wherein the determining a gum line of the target tooth model according to the edge class of the edge comprises:

determining a patch class of the polygonal patch according to the edge class, wherein the patch class comprises a tooth surface patch and a gum surface patch;

the gum line is extracted based on common edges of the different and adjacent polygonal patches of the patch categories.

20. A method of making a dental appliance, comprising:

obtaining a target segmented tooth according to the method of any one of claims 1-19;

21. The method of making a dental instrument of claim 20, further comprising:

converting the gum line into a cut line;

cutting the initial instrument based on the cutting line to obtain the dental instrument; wherein the initial fixture has an association with the model of the shaped tooth.

22. The method of claim 21, further comprising, prior to converting the gum line into a cut line:

at least one of a smoothing process and a position adjustment process is performed on the gum line.

23. The method of making a dental instrument of claim 20, further comprising:

performing film pressing treatment on the molded tooth model to obtain a shell-shaped film covering the solid tooth model;

and marking the shell-shaped membrane to obtain the initial equipment with the mark.

24. A three-dimensional dental model segmentation apparatus, comprising:

a memory and a processor, the memory storing a computer program which, when executed by the processor, performs the method of any one of claims 1 to 15.

25. A computer-readable storage medium, having a computer program stored thereon, characterized in that the computer program, when executed by a processor, performs the method of any of claims 1-19 or 19-23.