Tooth segmentation on dental meshes using morphologic skeleton

Tooth segmentation has an important role in computer-aided orthodontics. However, fine segmentation results remain difficult to obtain because of various tooth shapes, complex tooth arrangements, and especially, tooth-crowding problems. Most published approaches or commercial solutions in this area are either interaction-intensive or inaccurate, and thus, we propose a novel tooth segmentation approach based on morphologic skeleton for scanned dental meshes. Strict single-vertex width boundaries are obtained through improved morphologic skeleton technique. The skeleton describes the topological relationship among different dental parts on meshes and is exploited by automatic adjacent teeth separation. The morphologic skeleton technique eliminates dependence on a complex, precise mesh feature estimation and is implemented efficiently. The characteristics of the skeleton also facilitate effective teeth separation. Our techniques significantly reduce user interactions and are robust to various levels of tooth-crowding problems. We have conducted experiments on clinical dental models, thus demonstrating the effectiveness of the proposed approach. & 2013 Elsevier Ltd. All rights reserved.