A 3D dental model gum line recognition method, system, device and storage medium

Abstract

The present invention discloses a method, system, device and storage medium for recognizing the gum line of a 3D dental model, wherein the method comprises extracting multiple feature points from the 3D dental model based on a curvature geometry algorithm, preprocessing each of the feature points, and outputting a feature contour point group; matching a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters; iteratively calculating the first fitting reference line based on the feature contour point group to generate an initial fitting curve; smoothing the initial fitting curve using a dimensionality reduction algorithm, and outputting the gum line; thereby reducing the cost of orthodontic work, shortening data processing time, simplifying the automated workflow of orthodontic work, improving work efficiency, enhancing user experience, and expanding the application scenarios of the combination of dental diagnosis and treatment and 3D technology. It can be widely used in the field of orthodontic technology.

Description

Technical Field

[0001] The present invention relates to the technical field of orthodontics, and in particular to a method, system, device and storage medium for recognizing gum lines of 3D dental models. Background Art

[0002] Terminology explanation:

[0003] Invisible orthodontics: refers to invisible bracket-free orthodontics, which is a type of dental orthodontics.

[0004] With the development of science and technology information, computer technology, manufacturing technology, digital modeling technology, material science, numerical control technology and other technologies have grown rapidly and these disciplines have been integrated with each other. In particular, computer technology has increasingly penetrated into various aspects of teaching, scientific research and clinical application in various fields of medicine and has achieved more progressive mutual cooperation. With the development and popularization of measurement technology, people can easily obtain digital tooth models, which plays an important role in oral clinical diagnosis and treatment. The emergence and development of 3D printing technology has become one of the hot topics now, and the application of 3D printing technology in the medical field is becoming more and more common. 3D printing has been used in the medical field for more than 20 years and is widely used in oral implants, orthopedics, neurosurgery and other surgeries.

[0005] Dental care is an issue that most people cannot avoid. With the improvement of computer software and hardware technology, and the emergence of increasingly sophisticated dental data, people have also shifted their focus from traditional purely manual dental care to digital dental care such as invisible orthodontic treatment, using the prior knowledge provided by high-tech digital technology to improve the safety and success rate of dental treatment.

[0006] In the current application scenarios of invisible orthodontic dental diagnosis and treatment, all application scenarios require data processing of tooth models, and the data preprocessing of tooth models occupies most of the time of 3D tooth printing. Especially in the application scenarios of orthodontic correction, in addition to the need to pre-process the data of the tooth model including placement, hollowing, and Boolean operation, it is also necessary to perform lamination after printing, and after lamination, it is necessary to perform manual cutting, or manually draw the gum line and provide it to the CNC machine for cutting. This process requires huge manpower costs and a huge workload. In order to improve the efficiency of the entire diagnosis and treatment process and enhance the user experience, it is necessary to shorten the data processing time of the tooth model data.

[0007] In addition, the automated processing flow of orthodontic correction is still in a blank stage in China. Therefore, there is an urgent need to combine artificial intelligence algorithms, 3D printing, CNC machining and other technologies to realize the automated generation of processing flow for tooth correction in order to improve the work efficiency and competitiveness of the entire industry.

[0008] The technical solution of this application aims to solve the current technical problems in China of low automation, poor efficiency and poor user experience in the process of automated orthodontic treatment of tooth model data. Based on improving efficiency, this application innovatively explores the combination of the dental field and 3D printing technology, integrates digital 3D printing technology into the tooth restoration process, and introduces artificial intelligence technology and a series of algorithms to automatically identify gum line data, reduce manpower, deepen digital production, and fill the vacant technology of low automation, poor efficiency and poor user experience in the process of automated orthodontic treatment of tooth model data, laying a solid foundation for subsequent application scenarios combining dentistry and 3D technology. Summary of the invention

[0009] In order to solve the above technical problems, the purpose of the present invention is to provide a 3D dental model gum line recognition method, system, device and storage medium.

[0010] The first technical solution adopted by the present invention is:

[0011] A method for identifying a gum line of a 3D dental model comprises the following steps:

[0012] Extracting multiple feature points from the 3D dental model based on a curvature geometry algorithm, preprocessing each of the feature points, and outputting a feature contour point group;

[0013] Matching a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters;

[0014] Performing an iterative operation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve;

[0015] The initial fitting curve is smoothed by using a dimensionality reduction algorithm to output the gum line.

[0016] Optionally, the preprocessing includes filtering or denoising, and the step of extracting multiple feature points from the 3D dental model based on a curvature geometry algorithm, preprocessing each of the feature points, and outputting a feature contour point group specifically includes the following steps:

[0017] After identifying the morphological parameters and the maximum bottom surface of the 3D dental model, the 3D dental model is aligned to the target position;

[0018] A curvature geometry algorithm is used to extract a plurality of feature points from the 3D dental model at the target position, wherein the feature points are distributed in the concave and convex areas of the 3D dental model;

[0019] After filtering or denoising the extracted feature points, a feature contour point group is generated.

[0020] Optionally, the morphological parameters include the shape of the 3D dental model, and the step of matching the first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters specifically includes the following steps:

[0021] Determining the direction of the 3D dental model according to the shape of the 3D dental model;

[0022] Matching characteristic contour points in the same direction as the characteristic contour of the current dental model from a pre-stored fitting reference line pool based on the direction of the 3D dental model;

[0023] A first fitting reference line is selected from each of the characteristic contour points in the same direction in combination with a pre-built coordinate system and a first threshold.

[0024] Optionally, the step of performing iterative operation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve specifically includes the following steps:

[0025] Geometrically superimposing the centroid of the first fitting reference line with the centroid of the characteristic contour point group based on a pre-built coordinate system;

[0026] An approximate iterative algorithm is used to iteratively fit the first fitting reference line to each feature point of the feature contour point group to generate an initial fitting curve.

[0027] Optionally, the dimension reduction algorithm includes a principal component analysis method, and the step of smoothing the initial fitting curve using the dimension reduction algorithm and outputting the gum line specifically includes the following steps:

[0028] The initial fitting curve is smoothed by using a principal component analysis method to determine the main direction points of the initial fitting curve;

[0029] The main direction points of the initial fitting curve are smoothly connected according to the interpolation spline to the smoothed initial fitting curve, and the gum line is output.

[0030] Optionally, after identifying the morphological parameters and the maximum bottom surface of the 3D dental model, the step of aligning the 3D dental model to a target position specifically includes the following steps:

[0031] Acquire a 3D dental model, identify the maximum bottom surface and morphological parameters of the 3D dental model, and output a normal vector of the maximum bottom surface;

[0032] Combining the normal vector of the maximum bottom surface and the preset first normal vector, rotating the 3D dental model to a target plane;

[0033] Obtaining the contour of the 3D dental model on the target plane, determining the tooth skeleton curve equation according to the contour, and outputting the direction vector of the 3D dental model;

[0034] The 3D dental model is rotated to a target position by combining the direction vector of the 3D dental model and a preset second normal vector.

[0035] The second technical solution adopted by the present invention is:

[0036] A 3D dental model gum line recognition system, comprising:

[0037] An extraction module is used to extract multiple feature points from the 3D dental model based on a curvature geometry algorithm, pre-process each of the feature points, and output a feature contour point group;

[0038] A matching module, used for matching a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters;

[0039] An iterative module, used for performing an iterative operation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve;

[0040] The output module is used to smooth the initial fitting curve using a dimensionality reduction algorithm and output the gum line.

[0041] Optionally, the extraction module includes:

[0042] An identification unit, used for identifying the morphological parameters and the maximum bottom surface of the 3D dental model, and then aligning the 3D dental model to a target position;

[0043] An extraction unit, used for extracting a plurality of feature points from the 3D dental model at a target position by using a curvature geometry algorithm, wherein the feature points are distributed in a concave and convex area of ​​the 3D dental model;

[0044] The generating unit is used to generate a characteristic contour line point group after filtering or denoising the extracted characteristic points.

[0045] Optionally, the identification unit includes:

[0046] An acquisition subunit is used to acquire a 3D dental model, identify the maximum bottom surface and morphological parameters of the 3D dental model, and output a normal vector of the maximum bottom surface;

[0047] A first rotation subunit, configured to rotate the 3D dental model to a target plane by combining the normal vector of the maximum bottom surface and a preset first normal vector;

[0048] An output subunit, used to obtain the contour of the 3D dental model on the target plane, determine the tooth skeleton curve equation according to the contour, and output the direction vector of the 3D dental model;

[0049] The second rotation subunit is used to rotate the 3D dental model to a target position by combining the direction vector of the 3D dental model and a preset second normal vector.

[0050] Optionally, the matching module includes:

[0051] A determination unit, used for determining the direction of the 3D dental model according to the shape of the 3D dental model;

[0052] A matching unit, used for matching characteristic contour points in the same direction as the characteristic contour of the current dental model from a pre-stored fitting reference line pool based on the direction of the 3D dental model;

[0053] The screening unit is used to screen out a first fitting reference line from each of the characteristic contour points in the same direction by combining a pre-built coordinate system and a first threshold value.

[0054] Optionally, the iteration module includes:

[0055] A superposition unit, used for geometrically superimposing the centroid of the first fitting reference line with the centroid of the characteristic contour point group based on a pre-built coordinate system;

[0056] The iteration unit is used to iteratively fit the first fitting reference line with each feature point of the feature contour point group by using an approximate iteration algorithm to generate an initial fitting curve.

[0057] Optionally, the output module includes:

[0058] A smoothing unit, used for smoothing the initial fitting curve by using a principal component analysis method to determine the main direction points of the initial fitting curve;

[0059] A smoothing unit is used to smoothly connect the smoothed initial fitting curve according to the interpolation spline for the main direction points of the initial fitting curve, and output the gum line.

[0060] The third technical solution adopted by the present invention is:

[0061] A device, wherein the memory is used to store at least one program, and the processor is used to load the at least one program to execute the above method.

[0062] The fourth technical solution adopted by the present invention is:

[0063] A storage medium stores a program executable by a processor, wherein the program executable by the processor is used to execute the method as described above when executed by the processor.

[0064] The beneficial effects of the present invention are: by adopting a curvature geometry algorithm, multiple feature points are extracted from a 3D dental model at a target position and each feature point is pre-processed to generate a feature contour line; a first fitting reference line is matched by combining the identified morphological parameters of the 3D dental model and a pre-stored fitting reference line pool; the first fitting reference line is iteratively calculated based on the extracted feature points to generate an initial fitting curve; finally, the initial fitting curve is smoothed by the algorithm to output the gum line; thereby greatly simplifying the automatic generation and processing flow of orthodontic treatment, reducing work costs, shortening the data processing time of tooth model data, improving work efficiency, enhancing user experience, and laying the foundation for the application scenarios of the combination of dental diagnosis and treatment with 3D technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a flow chart of steps of a method for identifying gum lines in a 3D dental model of the present invention;

[0066] Figure 2 This is a structural block diagram of a 3D dental model gum line recognition system of the present invention;

[0067] Figure 3 It is a schematic diagram of the overall method flow of the 3D dental model gum line provided by the present invention. DETAILED DESCRIPTION

[0068] like Figure 1 As shown, this embodiment provides a method for identifying the gum line of a 3D dental model, comprising the following steps:

[0069] S1. Extract multiple feature points from the 3D dental model at the target position based on the curvature geometry algorithm, and pre-process each feature point to generate a feature contour line;

[0070] S2. matching a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters;

[0071] S3, performing iterative calculation on the first fitting reference line based on each feature point to generate an initial fitting curve;

[0072] S4. Use the dimension reduction algorithm to smooth the initial fitting curve and output the gum line.

[0073] In this embodiment, the 3D dental model, i.e., the 3D tooth model, is made based on 3D printing technology and computer program processing and repair and is provided with a corresponding plane tooth model. The curvature geometry algorithm is the rotation rate of the tangent direction angle to the arc length of a certain face on the 3D dental model, indicating the degree of concavity and convexity of the face. The first fitting reference line is the same as the direction of the feature point (the direction of the 3D dental model), ensuring that the range with the most contour lines can cover the feature point area. Specifically, after identifying the morphological parameters and the maximum bottom surface of the imported 3D dental model, the 3D dental model is aligned to the target position; then, a plurality of feature points are extracted from the 3D dental model at the target position according to the curvature geometry algorithm; according to the morphological parameters of the identified 3D dental model, the feature contour line with the smallest deviation is matched from the pre-stored fitting reference line pool as the first fitting reference line; according to the plurality of feature points, the first fitting reference line is iteratively calculated to generate an initial fitting curve; finally, the dimensionality reduction algorithm is used to smooth the initial fitting curve, and the final gum line is output to complete the pre-processing of the 3D dental model data. The technical solution of this application aims at the current invisible orthodontic application scenarios and other problems such as large workload, high labor costs, low work efficiency and poor user experience in the tooth model data processing process, and proposes a 3D dental model gum line recognition method that can greatly simplify the automatic generation and processing process of dental correction, reduce work costs, shorten the data processing time of tooth model data, improve work efficiency, and enhance user experience. At the same time, it lays the foundation for the application scenario of combining dental diagnosis and treatment with 3D technology, filling the gap in the domestic orthodontic automation process. In addition, the 3D dental model gum line recognition method provided in this embodiment is also applicable to tooth models with arbitrary planes and directions, and any other type of tooth model with planes does not affect the implementation of the present invention.

[0074] Optionally, the preprocessing includes filtering or denoising, and step S1 specifically includes the following steps:

[0075] S11, obtaining a 3D dental model, identifying the maximum bottom surface and morphological parameters of the 3D dental model, and outputting a normal vector of the maximum bottom surface;

[0076] S12, extracting a plurality of feature points from the 3D dental model at the target position using a curvature geometry algorithm, wherein each of the feature points is distributed in a concave and convex area of ​​the 3D dental model;

[0077] S13, after filtering or denoising the extracted feature points, generating feature contour lines and storing them in a feature contour line pool.

[0078] In this embodiment, the obtained initial 3D dental model feature points are denoised or filtered to eliminate the influence of feature points in non-gingival line areas on the fitting of subsequent contour lines; the curvature geometry algorithm refers to the rotation rate of the tangent direction angle of a certain surface on the dental model to the arc length, indicating the degree of concavity of the surface, which is called a feature in this application scheme, and feature points can be extracted from the concave and convex areas of the 3D dental model through the curvature method.

[0079] Optionally, the step S11 includes the following steps:

[0080] S111, obtaining a 3D dental model, identifying the maximum bottom surface and morphological parameters of the 3D dental model, and outputting a normal vector of the maximum bottom surface;

[0081] S112, combining the normal vector of the maximum bottom surface and the preset first normal vector, rotating the 3D dental model to the target plane;

[0082] S113, obtaining the contour of the 3D dental model on the target plane, determining the tooth skeleton curve equation according to the contour, and outputting the direction vector of the 3D dental model;

[0083] S114. Rotate the 3D dental model to a target position by combining the direction vector of the 3D dental model and the preset second normal vector.

[0084] In this embodiment, the preset first normal vector is the normal vector corresponding to the rotation of the 3D dental model to the target plane; the preset second normal vector is the normal vector corresponding to the rotation of the 3D dental model to the target position. Specifically, first, the imported 3D dental model is obtained, and the normal vector of the bottom surface is obtained by obtaining a dental model in any direction with a bottom surface, and the 3D dental model is rotated to the target plane based on the cross multiplication method and the first normal vector; secondly, when the 3D model is in the target plane, the 3D dental model is projected to obtain the tooth contour curve; the skeleton is extracted from the tooth contour curve shape, and the extracted skeleton shape is processed for noise to obtain a smooth and continuous skeleton curve; the skeleton curve equation is solved to obtain the second normal vector of the target position; according to the obtained second normal vector of the target position, the rotation angle and the rotation axis are solved according to the cross multiplication method, and then the dental model is rotated to the target position; thereby completing the automatic placement of the 3D dental model in any direction to the target position.

[0085] Optionally, the morphological parameters include the shape of the 3D dental model, and step S2 specifically includes the following steps:

[0086] S21, determining the direction of the 3D dental model according to the shape of the 3D dental model;

[0087] S22, matching characteristic contour points in the same direction as the characteristic contour of the current dental model from a pre-stored fitting reference line pool based on the direction of the 3D dental model;

[0088] S23 , selecting a first fitting reference line from characteristic contour points in the same direction by combining a pre-built coordinate system and a first threshold value.

[0089] In this embodiment, the coordinate system is a rectangular coordinate system with X, Y, and Z axes. Other coordinate systems can also be selected according to the application scenario, which will not be described here. The first threshold refers to the value that minimizes the deviation between the matched same-direction feature contour line and the 3D dental model in the established coordinate system. The feature contour line refers to the gum line file of a tooth, which is composed of point coordinates; the fitting reference pool is actually the smoothed historical gum lines, which are generated based on other dental models.

[0090] Optionally, step S3 specifically includes the following steps:

[0091] S31, geometrically superimposing the centroid of the first fitting reference line and the centroid of the characteristic contour point group based on a pre-built coordinate system;

[0092] S32, using an approximate iterative algorithm to iteratively fit the first fitting reference line to each feature point of the feature contour point group to generate an initial fitting curve.

[0093] In this embodiment, the approximate iterative algorithm refers to the projection of the first fitting reference line, and the cyclic iteration refers to the projection error being less than the set threshold. By simultaneously acquiring the feature points and the first fitting reference line, the center of gravity of the first fitting reference line is combined and superimposed with the center of gravity of the line connecting the feature points in the established coordinate system. After the cyclic projection iteration, the projected points are connected to form an initial fitting line to ensure that the optimal feature contour line can cover the feature point area.

[0094] Optionally, the dimension reduction algorithm includes a principal component analysis method, and step S4 specifically includes the following steps:

[0095] S41, using principal component analysis to smooth the initial fitting curve and determine the main direction points of the initial fitting curve;

[0096] S42, smoothly connecting the smoothed initial fitting curve according to the interpolation spline for the main direction points of the initial fitting curve, and outputting the gum line.

[0097] In this embodiment, the generated initial fitting line is not a smooth curve, and there will be local folding and deviation. To address this phenomenon, the principal component analysis (PCA) method is used to obtain its main contour shape, determine the points of the main direction of the contour, and use interpolation splines such as the Kochanek-Bartels style to smoothly reconnect them point by point to avoid uneven line segment areas and obtain the final smooth gum line.

[0098] like Figure 2As shown, this embodiment provides a 3D dental model gum line recognition system, including:

[0099] An extraction module is used to extract multiple feature points from the 3D dental model based on a curvature geometry algorithm, pre-process each of the feature points, and output a feature contour point group;

[0100] A matching module, used for matching a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters;

[0101] An iterative module, used for performing an iterative operation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve;

[0102] The output module is used to smooth the initial fitting curve using a dimensionality reduction algorithm and output the gum line.

[0103] Optionally, the extraction module includes:

[0104] An identification unit, used for identifying the morphological parameters and the maximum bottom surface of the 3D dental model, and then aligning the 3D dental model to a target position;

[0105] An extraction unit, used for extracting a plurality of feature points from the 3D dental model at a target position by using a curvature geometry algorithm, wherein the feature points are distributed in a concave and convex area of ​​the 3D dental model;

[0106] The generating unit is used to generate a characteristic contour line point group after filtering or denoising the extracted characteristic points.

[0107] Optionally, the identification unit includes:

[0108] An acquisition subunit is used to acquire a 3D dental model, identify the maximum bottom surface and morphological parameters of the 3D dental model, and output a normal vector of the maximum bottom surface;

[0109] A first rotation subunit, configured to rotate the 3D dental model to a target plane by combining the normal vector of the maximum bottom surface and a preset first normal vector;

[0110] An output subunit, used to obtain the contour of the 3D dental model on the target plane, determine the tooth skeleton curve equation according to the contour, and output the direction vector of the 3D dental model;

[0111] The second rotation subunit is used to rotate the 3D dental model to a target position by combining the direction vector of the 3D dental model and a preset second normal vector.

[0112] Optionally, the matching module includes:

[0113] A determination unit, used for determining the direction of the 3D dental model according to the shape of the 3D dental model;

[0114] A matching unit, used for matching characteristic contour points in the same direction as the characteristic contour of the current dental model from a pre-stored fitting reference line pool based on the direction of the 3D dental model;

[0115] The screening unit is used to screen out a first fitting reference line from each of the characteristic contour points in the same direction by combining a pre-built coordinate system and a first threshold value.

[0116] Optionally, the iteration module includes:

[0117] A superposition unit, used for geometrically superimposing the centroid of the first fitting reference line with the centroid of the characteristic contour point group based on a pre-built coordinate system;

[0118] The iteration unit is used to iteratively fit the first fitting reference line with each feature point of the feature contour point group by using an approximate iteration algorithm to generate an initial fitting curve.

[0119] Optionally, the output module includes:

[0120] A smoothing unit, used for smoothing the initial fitting curve by using a principal component analysis method to determine the main direction points of the initial fitting curve;

[0121] A smoothing unit is used to smoothly connect the smoothed initial fitting curve according to the interpolation spline for the main direction points of the initial fitting curve, and output the gum line.

[0122] A device, a memory is used to store at least one program, and a processor is used to load at least one program to execute a method embodiment method.

[0123] A device of this embodiment can execute a 3D dental model gum line recognition method provided by the method embodiment of the present invention, can execute any combination of implementation steps of the method embodiment, and has the corresponding functions and beneficial effects of the method.

[0124] A storage medium stores a program executable by a processor, and the program executable by the processor is used to execute a method embodiment method when executed by the processor.

[0125] A storage medium of this embodiment can execute a 3D dental model gum line recognition method provided in the first embodiment of the method of the present invention, can execute any combination of implementation steps of the method embodiment, and has the corresponding functions and beneficial effects of the method. Specific embodiments

[0127] Reference Figure 3 The overall method flow diagram of the 3D dental model gum line provided by the present invention includes the following:

[0128] Start importing 3D dental models and establishing a rectangular coordinate system.

[0129] Acquire the 3D dental model and automatically adjust the 3D dental model to the target position.

[0130] Specifically: (1) For obtaining the imported 3D tooth model, a tooth model in any direction with a bottom surface is obtained, and the tooth model is a digital three-dimensional body composed of a series of triangular facets.

[0131] (2) Detect the maximum plane of the tooth model. The method for detecting the maximum plane of the tooth is: set a certain triangular facet, superimpose the set triangular facet and the 3D tooth model composed of triangular facets, set the error threshold e, and when e is greater than a certain value, it is considered that the set triangular facet and the facet on the 3D tooth model are not level; otherwise, they are considered to be in the same plane. When the set triangular facet and a facet of the tooth model are in the same plane, superimpose them together, continue to look for the next triangular facet and determine the error threshold. Repeat the above steps until the largest plane of the tooth model is obtained. In the process of detecting the largest plane of the tooth model, the equation of its plane can be obtained.

[0132] (3) According to the cross product method, the rotation angle and rotation axis are solved according to the vector values ​​before and after the rotation. The cross product method is a binary operation of vectors in the vector space, and its operation result is a vector rather than a scalar. Based on the above rotation angle and rotation axis, any model can be rotated to the desired spatial position.

[0133] (4) After the rotation is realized based on (3), the tooth model is obtained and projected into a tooth contour curve. The tooth contour curve is to project the object in 3D space onto a 2D plane to obtain the plane contour of the object. The contour skeleton curve of the tooth model is obtained, which can be represented by a quadratic curve equation. By solving the curve equation, the direction vector of the tooth model can be obtained. According to the cross multiplication method, as in (3), the rotation angle and rotation axis are solved according to the vector values ​​before and after the rotation. The above rotation angle and rotation axis are rotated to the desired direction position based on a specific plane.

[0134] Extract feature points of 3D dental models based on curvature geometry algorithm.

[0135] Specifically: after placing the model at the specified position, the feature points extracted from the tooth model are filtered and denoised according to the geometric calculation method of curvature to obtain the characteristic contour of the 3D dental model; the curvature geometric calculation method is the rotation rate of the tangent direction angle of a certain surface on the dental model to the arc length, which indicates the degree of concavity and convexity of the surface, which is also called a feature in this scheme. The curvature method can be used to obtain the feature points of the concave and convex areas of the dental model (the real gum line on the dental model is also reflected by the concavity and convexity).

[0136] Optimize the feature points to obtain the feature contour point group of the 3D dental model.

[0137] Specifically, for the initial dental model feature points obtained in (2), it is necessary to remove noise. The reason for removing noise is that in addition to the gum line area, other parts of the teeth also have bumps and depressions, which are also considered as features. This part will affect the subsequent fitting with the contour line, so it is necessary to filter or remove it, and finally obtain the optimal feature points; wherein the feature contour point group contains multiple feature points remaining after optimization, and the distribution shape of the feature points in the feature contour point group (referred to as the feature contour point group shape) will be similar to the gum line contour of the dental model. It should be noted that the centroid of the feature contour point group can be understood as the centroid of the feature contour point group shape.

[0138] Automatically select the optimal fitting reference line based on the shape and contour of the 3D dental model.

[0139] Specifically: (1) According to the shape and contour of the imported dental model, the optimal fitting reference line is automatically selected through the placed dental model. The fitting reference line refers to the gum line file of a tooth, which is composed of point coordinates; (2) The main method for searching the optimal fitting reference line is: first, match the fitting reference line data of the same batch of teeth with the same direction according to the direction of the dental model; secondly, from the screened data with the same direction, determine the matching degree between the fitting reference line and the dental model according to the three directions of X, Y, and Z, and select the fitting reference line with the smallest deviation, that is, the optimal fitting reference line.

[0140] The feature points are fitted with the fitting reference line to obtain the initial fitting line.

[0141] Specifically, after obtaining the feature points and the optimal fitting reference line at the same time, the fitting reference line needs to be overlapped or brought closer to the feature points as much as possible, that is, the center of gravity of the fitting reference line is translated in the X, Y, and Z directions to coincide with the center of gravity of the dental model. Preferably, the center of gravity of the fitting reference line is made to coincide with the center of gravity of the shape of the feature contour point group, while ensuring that the fitting reference line can cover the feature point area; (2) using an approximate iterative algorithm, the feature points of the fitting reference line and the feature contour point group are fitted to form an initial fitting line, where the fitting is to project the points of the fitting reference line (essentially a series of coordinate points) to the nearest feature point; approximate iteration means that after the contour line is projected, the iteration continues until the projection error is less than the set threshold; (3) after the iteration is completed, the projected points are connected to form an initial fitting line.

[0142] Smooth the initial fitted line to obtain the final gum line.

[0143] Specifically: obtain the initial fitting line. Since the initial fitting line is not a smooth curve, there will be local folding and deviation. In response to this situation, this solution uses the principal component analysis method to obtain its main contour shape, determine the points of the main direction of the contour, and reconnect them smoothly using the Kochanek-Bartels style to avoid the uneven area of ​​the line segment and obtain the final smooth gum line.

[0144] Finally, output the final gum line of the 3D dental model.

[0145] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for identifying the gum line of a 3D dental model, characterized in that: The following steps are involved: Extract multiple feature points from the 3D dental model based on the curvature geometry algorithm, pre-process each feature point, and output a feature contour point group, including: - Identify the morphological parameters and maximum bottom surface of the 3D dental model, and adjust the 3D dental model to the target position; - extracting multiple feature points from the 3D dental model at the target position using a curvature geometry algorithm, wherein the feature points are distributed in the concave and convex areas of the 3D dental model; - filtering or denoising the extracted feature points to generate a feature contour point group; Matching a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters includes: -determining the direction of the 3D dental model according to the shape of the 3D dental model; - matching characteristic contour points in the same direction as the characteristic contour of the current dental model from a pre-stored fitting reference line pool based on the direction of the 3D dental model; - selecting a first fitting reference line from each of the characteristic contour points in the same direction by combining a pre-built coordinate system and a first threshold; performing an iterative operation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve; The initial fitting curve is smoothed by using a dimensionality reduction algorithm to output the gum line.

2. A method for identifying the gum line of a 3D dental model according to claim 1, characterized in that: The step of performing iterative calculation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve specifically comprises the following steps: Geometrically superimposing the centroid of the first fitting reference line with the centroid of the characteristic contour point group based on a pre-built coordinate system; An approximate iterative algorithm is used to iteratively fit the first fitting reference line to each feature point of the feature contour point group to generate an initial fitting curve.

3. The method for identifying the gum line of a 3D dental model according to claim 1, characterized in that: The dimension reduction algorithm includes a principal component analysis method. The step of using the dimension reduction algorithm to smooth the initial fitting curve and output the gum line specifically includes the following steps: The initial fitting curve is smoothed by using a principal component analysis method to determine the main direction points of the initial fitting curve; The main direction points of the initial fitting curve are smoothly connected according to the interpolation spline to the smoothed initial fitting curve, and the gum line is output.

4. The method for identifying the gum line of a 3D dental model according to claim 1, characterized in that: After identifying the morphological parameters and the maximum bottom surface of the 3D dental model, the step of aligning the 3D dental model to the target position specifically includes the following steps: Acquire a 3D dental model, identify the maximum bottom surface and morphological parameters of the 3D dental model, and output a normal vector of the maximum bottom surface; Combining the normal vector of the maximum bottom surface and the preset first normal vector, rotating the 3D dental model to a target plane; Obtaining the contour of the 3D dental model on the target plane, determining the tooth skeleton curve equation according to the contour, and outputting the direction vector of the 3D dental model; The 3D dental model is rotated to a target position by combining the direction vector of the 3D dental model and a preset second normal vector.

5. A 3D dental model gum line recognition system, characterized in that: include: The extraction module is used to extract multiple feature points from the 3D dental model based on the curvature geometry algorithm, pre-process each of the feature points, and output a feature contour point group. The extraction module includes: - an identification unit, used to identify the morphological parameters and the maximum bottom surface of the 3D dental model, and to adjust the 3D dental model to a target position; - an extraction unit, used to extract a plurality of feature points from the 3D dental model at the target position by using a curvature geometry algorithm, wherein the feature points are distributed in the concave and convex areas of the 3D dental model; - a generating unit, used for filtering or denoising the extracted characteristic points to generate a characteristic contour line point group; A matching module is used to match a first fitting reference line from a pre-stored fitting reference line pool according to the morphological parameters, and the matching module includes: - a determination unit, used to determine the direction of the 3D dental model according to the shape of the 3D dental model; - A matching unit, used for matching characteristic contour points in the same direction as the characteristic contour of the current dental model from a pre-stored fitting reference line pool based on the direction of the 3D dental model; - a screening unit, configured to screen out a first fitting reference line from each of the characteristic contour points in the same direction by combining a pre-built coordinate system and a first threshold value; An iterative module, used for performing an iterative operation on the first fitting reference line based on the characteristic contour point group to generate an initial fitting curve; The output module is used to smooth the initial fitting curve using a dimensionality reduction algorithm and output the gum line.

6. A 3D dental model gum line recognition device, characterized in that: The invention comprises a memory and a processor, wherein the memory is used to store at least one program, and the processor is used to load the at least one program to execute the method according to any one of claims 1 to 4.

7. A storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to perform the method according to any one of claims 1 to 4 when executed by the processor.

Images