Initial contour generation approach in level set methods for dental image segmentation

Abstract

Segmentation is challenging process in medical images especially on dental x-ray images. Level set methods have effective result on medical and dental image segmentation. Initial Contour (IC) is the essential step in level set image segmentation methods due to start the efficient process. However, the main issue with IC is how to generate the automatic technique in order to reduce the human interaction and moreover, suitable IC to have accurate result. In this paper a new region-based technique for IC generation, is proposed to overcome this issue. The idea is to generate the most suitable IC since the manual initialization of the level set function surface is a well-known drawback for accurate segmentation which has dependency on selection of IC and wrong selection will affect the result. We have utilized the statistical and morphological information inside and outside the contour to establish a region-based map function. This function is able to find the suitable IC on images to perform by level set methods. Experiments on dental x-ray images demonstrate the robustness of segmentation process using proposed method even on noisy images and with weak boundary. Furthermore, computational cost of segmentation process will be reduced.