Search

CN-122023723-A - Dental crown feature point extraction method, device, equipment and medium

CN122023723ACN 122023723 ACN122023723 ACN 122023723ACN-122023723-A

Abstract

The invention relates to the technical field of dental crown data processing and discloses a dental crown feature point extraction method, a device, equipment and a medium, wherein the method comprises the steps of acquiring dental jaw point cloud data of a target patient, reconstructing point cloud grids of the dental jaw point cloud data, and generating a three-dimensional dental crown digital model; the method comprises the steps of carrying out dental region segmentation on a three-dimensional dental crown digital model to obtain independent dental crown models of all teeth, constructing a local coordinate system corresponding to the independent dental crown models, carrying out spatial registration on the independent dental crown models according to the local coordinate system to obtain a target dental crown model, extracting an initial dental crown feature matrix according to the target dental crown model, carrying out self-adaptive sampling on the initial dental crown feature matrix to obtain dental crown sampling vertexes, carrying out double-branch deviation prediction according to the dental crown sampling vertexes to obtain a deviation point cluster, calculating a point cluster central point of the deviation point cluster, and determining dental crown feature points according to the point cluster central point. The invention can improve the efficiency and accuracy of extracting the characteristic points of the dental crowns.

Inventors

  • XIE SHICHEN
  • She yue

Assignees

  • 成都齿灵科技有限公司

Dates

Publication Date
20260512
Application Date
20260211

Claims (10)

  1. 1. A method for extracting dental crown feature points, the method comprising: Acquiring dental point cloud data of a target patient, and performing point cloud grid reconstruction on the dental point cloud data to generate a three-dimensional dental crown digital model; Dividing the dental region of the three-dimensional dental crown digital model to obtain independent dental crown models of all teeth, and constructing a local coordinate system corresponding to the independent dental crown models; Performing spatial registration on the independent dental crown model according to the local coordinate system to obtain a target dental crown model, and extracting an initial dental crown feature matrix according to the target dental crown model; performing self-adaptive sampling on the initial crown feature matrix to obtain a plurality of crown sampling vertexes, and performing double-branch offset prediction according to the crown sampling vertexes to obtain offset point clusters; And calculating the point cluster center point of the offset point cluster, and determining the characteristic point of the dental crown according to the point cluster center point.
  2. 2. The method for extracting dental crown feature points according to claim 1, wherein the performing point cloud mesh reconstruction on the dental jaw point cloud data to generate a three-dimensional dental crown digital model includes: performing data normalization processing on the dental point cloud data to obtain first dental point cloud data, and performing outlier removal on the first dental point cloud data to obtain target dental point cloud data; Performing normal vector analysis on the target dental point cloud data to obtain a normal direction corresponding to the target dental point cloud data; performing point cloud smoothing processing on the target dental point cloud data according to the normal direction to obtain smoothed dental point cloud data; calculating a curved surface center point of the smooth dental point cloud data, and generating a curved surface supporting domain according to the curved surface center point; performing surface reconstruction on the smooth dental point cloud data according to the surface supporting domain to obtain a three-dimensional dental crown surface; and carrying out three-dimensional modeling on the three-dimensional dental crown curved surface to obtain a three-dimensional dental crown digital model.
  3. 3. The method for extracting dental crown feature points according to claim 1, wherein the performing dental jaw region segmentation on the three-dimensional dental crown digital model to obtain an independent dental crown model of each tooth comprises: smooth filtering is carried out on the three-dimensional dental crown digital model, and a target dental crown digital model is obtained; Extracting global features of the target dental crown digital model, and carrying out initial region division on the target dental crown digital model according to the global features by utilizing a region growing algorithm to obtain a plurality of initial tooth regions; Acquiring initial region boundaries corresponding to a plurality of initial tooth regions, and extracting local features of the initial tooth regions; performing boundary optimization on the boundary of the initial region based on the local features to obtain a target tooth region; Performing morphological fitting on the target tooth area to obtain an initial independent dental crown model; and performing model post-processing on the initial independent dental crown model to obtain an independent dental crown model.
  4. 4. The method for extracting dental crown feature points according to claim 1, wherein said constructing a local coordinate system corresponding to the independent dental crown model comprises: obtaining vertex coordinate data of the independent dental crown model, and calculating the average value of the vertex coordinate data; Performing difference calculation on the vertex coordinate data according to the mean value to obtain centralized coordinate data; constructing a covariance matrix based on the centralized coordinate data, and solving eigenvalues and corresponding eigenvectors of the covariance matrix; Selecting a feature vector corresponding to the largest feature value in the feature values to determine the first initial main direction, selecting a feature vector corresponding to the next largest feature value in the feature values to determine the second initial main direction, and selecting a feature vector corresponding to the smallest feature value in the feature values to determine the third initial main direction; performing direction validity verification on the first initial main direction, the second initial main direction and the third initial main direction to obtain a target main direction, and determining the target main direction as a first approximate coordinate axis; Extracting a dental crown geometric center, a dental crown occlusal plane, a facial lingual plane and a dental crown near-far direction of the independent dental crown model, and performing least square fitting on the dental crown occlusal plane to obtain a target occlusal plane; Aligning the transverse axis of the first approximate coordinate axis with the far and middle direction of the dental crown, aligning the longitudinal axis of the first approximate coordinate axis with the facial lingual plane direction, and generating a second coordinate axis according to an alignment result, wherein the vertical axis of the first approximate coordinate axis is perpendicular to the target occlusal plane; And constructing a local coordinate system corresponding to the independent dental crown model by taking the dental crown geometric center as an origin and according to the origin and the second coordinate axis.
  5. 5. The method for extracting dental crown feature points according to claim 1, wherein spatially registering the independent dental crown model according to the local coordinate system to obtain a target dental crown model comprises: extracting characteristic points under a preset reference coordinate system and the local coordinate system, and calculating coordinate system transformation parameters according to the characteristic points; performing initial rigid matrix transformation according to the coordinate system transformation parameters to obtain a coordinate system rotation matrix and a coordinate system translation vector; carrying out rigid registration on the local coordinate system according to the coordinate system rotation matrix and the coordinate system translation vector to obtain a target coordinate system; And rotating the independent dental crown model into the target coordinate system to obtain a target dental crown model.
  6. 6. The method for extracting dental crown feature points according to claim 1, wherein said adaptively sampling the initial dental crown feature matrix to obtain a plurality of dental crown sampling vertices comprises: extracting geometric characteristic parameters and texture characteristic parameters of the initial crown characteristic matrix; carrying out parameter normalization processing on the geometric feature parameters and the texture feature parameters, and mapping the normalized feature parameters to the initial dental crown feature matrix to obtain a standardized feature matrix; Calculating importance weights of all feature parameters in the standardized feature matrix; Determining the characteristic parameters with the importance weights being larger than a preset first sampling threshold value as high sampling density parameters, and determining the characteristic parameters with the importance weights being smaller than or equal to the first sampling threshold value as low sampling density parameters; Performing layered sampling on the standardized feature matrix according to the high sampling density parameter and the low sampling density parameter to obtain a preliminary sampling top point set; and calculating Euclidean distances of adjacent vertexes in the preliminary sampling vertex set, and determining a vertex pair with the Euclidean distance smaller than a preset distance threshold value as a crown sampling vertex.
  7. 7. The method for extracting dental crown feature points according to claim 1, wherein said performing a double-branch offset prediction from said crown sampling vertices to obtain an offset point cluster comprises: constructing a vertex initial feature matrix according to the crown sampling vertexes, and carrying out multi-layer convolution on the vertex initial feature matrix to obtain a high-dimensional vertex feature matrix; performing direction offset analysis on the high-dimensional vertex feature matrix according to a preset direction branch network to obtain unit offset direction vectors of all vertexes in the high-dimensional vertex feature matrix; performing distance offset analysis on the high-dimensional vertex feature matrix according to a preset scale branch network to obtain offset distances of all vertexes in the high-dimensional vertex feature matrix; Calculating the offset of each vertex according to the unit offset direction vector and the offset distance; and obtaining initial three-dimensional coordinates of each vertex in the high-dimensional vertex feature matrix, and adding the initial three-dimensional coordinates and the offset to obtain an offset point cluster.
  8. 8. A crown feature point extraction device, the device comprising: The point cloud model generation module is used for acquiring dental point cloud data of a target patient, carrying out point cloud grid reconstruction on the dental point cloud data and generating a three-dimensional dental crown digital model; the local coordinate system construction module is used for carrying out dental region segmentation on the three-dimensional dental crown digital model to obtain independent dental crown models of all teeth, and constructing a local coordinate system corresponding to the independent dental crown models; The model space registration module is used for carrying out space registration on the independent dental crown model according to the local coordinate system to obtain a target dental crown model, and extracting an initial dental crown feature matrix according to the target dental crown model; The dental crown deviation prediction module is used for adaptively sampling the initial dental crown feature matrix to obtain a plurality of dental crown sampling vertexes, and performing double-branch deviation prediction according to the dental crown sampling vertexes to obtain a deviation point cluster; And the dental crown feature point determining module is used for calculating the point cluster center point of the offset point cluster and determining the dental crown feature point according to the point cluster center point.
  9. 9. An electronic device, the electronic device comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform a dental crown feature point extraction method according to any one of claims 1 to 7.
  10. 10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements a crown feature point extraction method according to any one of claims 1 to 7.

Description

Dental crown feature point extraction method, device, equipment and medium Technical Field The present invention relates to the field of dental crown data processing technologies, and in particular, to a method, an apparatus, a device, and a medium for extracting dental crown feature points. Background In the field of computer-aided dental diagnosis and treatment, the construction of a high-precision three-dimensional dental crown digital model is a key for realizing personalized diagnosis and treatment and accurate restoration, and dental crown characteristic points are used as the most representative anatomical marks on the tooth surface, and the accurate extraction and positioning of the dental crown characteristic points directly relate to the effects of model construction, restoration body design and orthodontic treatment. The existing common dental crown feature point extraction method mainly depends on manual labeling, but the method is not only limited by the difference of understanding of technicians on dental crown structures, so that data consistency and accuracy are difficult to ensure, and meanwhile, the manual labeling process is time-consuming and labor-consuming, cannot meet the requirements of clinical large-scale data processing, and is easy to introduce human errors, so that the accuracy and stability of the construction of a three-dimensional dental crown digital model are influenced, and the establishment and implementation of dental restoration and orthodontic treatment schemes are restricted. Therefore, how to improve the efficiency and accuracy of extracting the feature points of the dental crowns is a urgent problem to be solved. Disclosure of Invention The invention provides a method, a device, equipment and a medium for extracting dental crown feature points, which mainly aim to solve the problems of low extraction efficiency and inaccurate extraction of dental crown feature points. In order to achieve the above object, the present invention provides a method for extracting a feature point of a dental crown, comprising: Acquiring dental point cloud data of a target patient, and performing point cloud grid reconstruction on the dental point cloud data to generate a three-dimensional dental crown digital model; Dividing the dental region of the three-dimensional dental crown digital model to obtain independent dental crown models of all teeth, and constructing a local coordinate system corresponding to the independent dental crown models; Performing spatial registration on the independent dental crown model according to the local coordinate system to obtain a target dental crown model, and extracting an initial dental crown feature matrix according to the target dental crown model; performing self-adaptive sampling on the initial crown feature matrix to obtain a plurality of crown sampling vertexes, and performing double-branch offset prediction according to the crown sampling vertexes to obtain offset point clusters; And calculating the point cluster center point of the offset point cluster, and determining the characteristic point of the dental crown according to the point cluster center point. In a second aspect, the present invention also provides a device for extracting a feature point of a dental crown, including: The point cloud model generation module is used for acquiring dental point cloud data of a target patient, carrying out point cloud grid reconstruction on the dental point cloud data and generating a three-dimensional dental crown digital model; the local coordinate system construction module is used for carrying out dental region segmentation on the three-dimensional dental crown digital model to obtain independent dental crown models of all teeth, and constructing a local coordinate system corresponding to the independent dental crown models; The model space registration module is used for carrying out space registration on the independent dental crown model according to the local coordinate system to obtain a target dental crown model, and extracting an initial dental crown feature matrix according to the target dental crown model; The dental crown deviation prediction module is used for adaptively sampling the initial dental crown feature matrix to obtain a plurality of dental crown sampling vertexes, and performing double-branch deviation prediction according to the dental crown sampling vertexes to obtain a deviation point cluster; And the dental crown feature point determining module is used for calculating the point cluster center point of the offset point cluster and determining the dental crown feature point according to the point cluster center point. In a third aspect, the present invention also provides an electronic device, including: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to per