Automatic denture neck edge line extraction method based on DBSCAN clustering

Abstract

The invention discloses an automatic denture neck edge line extraction method based on DBSCAN clustering. The method comprises steps: feature extraction is carried out on a to-be-extracted object, anda feature area of the to-be-extracted object is obtained; a DBSCAN clustering method is adopted in the feature area of the to-be-extracted object to automatically recognize a neck edge feature area;an improved morphology method is adopted to process the neck edge feature area to obtain a neck edge line, wherein the improved morphology method enables broken neck edge feature areas to be communicated and enables an open loop feature line in the neck edge feature area to be closed. The neck edge feature area is recognized automatically through DBSCAN clustering, thereby eliminating interferencefrom a top feature area and other noise; the improved morphology method is adopted to process the neck edge feature area to obtain the neck edge line, and the method can be ensured to reliably extract a reasonable neck edge line; and the whole extraction process does not need interaction operation, the intelligent degree is high, and the robustness is good. The method of the invention can be widely applied to a prosthodontics field.

Description

technical field [0001] The invention relates to the field of oral restoration, in particular to a method for automatically extracting denture cervical margin lines based on DBSCAN clustering. Background technique [0002] The wide application of CAD / CAM technology in the field of prosthodontics has enabled the prosthodontic technology to realize a leap from manual operation to automation and intelligence. The extraction of the cervical margin line of the preparation is a key link in the marginal sealing of the oral CAD / CAM restoration process, and the marginal sealing of the restoration crown will have an important impact on the patient's gingival health (in the denture restoration CAD / CAM system, the cervical margin The extraction of feature lines is the primary processing task of crown restoration design, and its extraction quality directly affects the modeling accuracy of the restoration and the final restoration effect). [0003] In the paper titled "Research and Realiz...

AI Technical Summary

Problems solved by technology

[0005] To sum up, most of the current neckline extraction methods have the following shortcomings: (1) The degree of intelligence is not high, and one or more feature points need to be manually picked; (2) The robustness is not good, and it is sensitive to noise Or it is difficult to deal with the situation where the local features of the cervical margin are not obvious

Images