Semi-automatic teeth segmentation in 3D models of dental casts using a hybrid methodology - Emerging Trends in Image Processing, Computer Vision, and Pattern Recognition - page 361

CHAPTER 28

Semi-automatic teeth
segmentation in 3D models of
dental casts using a hybrid
methodology

J.D. Tamayo-Quintero 1 ; S. Arboleda-Duque 1 , 2 ; J.B. Gómez-Mendoza 1

1 Department of Electric, Electronic and Computer Engin-

eering, Universidad Nacional de Colombia, Manizales, Caldas, Colombia
2 Department of Telecommunications Engineering, Universidad Católica de Manizales, Manizales, Caldas, Colombia

Abstract
This document is an extension of the chapter “Image Segmentation Techniques Applied to Point Clouds
of Dental Models with an Improvement in Semi-Automatic Teeth Segmentation,” presented in the Inter-
national Conference on Image Processing and Computer Vision IPCV 2014. The work arises from an ex-
ploratory study on the application of a combination of different segmentation techniques to point clouds
of dental models. Results from a previous work—where the subject of study are point clouds repres-
enting urban landscapes—suggested that hybridization of current methods for point cloud segmenta-
tion perform beter if compared to the methods themselves. Base techniques used in this work comprise
geometric primitive model approximation (e.g., RANSAC), Region Growing segmentation, and a graph
theory-based approach (particularly the “Min-Cut” algorithm). In our application, the techniques were
tested using dental 3D point clouds. Data were acquired using a Konica Minolta VIVID 9i laser range
scanner.
Also, a semi-automatic segmentation methodology is presented. Results of teeth segmentation using
testing data suggest that it is possible to automatically segment teeth from digital 3D models.

Keywords

Point cloud
3D dental models
Segmentation
Region Growing
Min-Cut