A Comparative Analysis of Various Segmentation Techniques on Dental Images

Abstract

Image Segmentation has a very important role in medical imaging. It is a process of dividing the image into multiple parts, which are used for identifying objects and other relevant information. Image segmentation bridges a gap between the low-level image details and high-level image components. The role of segmentation is crucial for the tasks that require image analysis. The goal of this paper is to review the image segmentation techniques on dental images. An exhaustive survey of four widely used segmentation techniques has been carried out and the performance comparison of each method within Edge detection, Thresholding, Deformable model and Clustering segmentation technique is provided. The segmentation techniques are Edge Detection, Thresholding, Deformable Model and Clustering. The accuracy of the segmentation process determines the success or failure of the final analysis process. With the help of image quality metrics such as root mean square error, mean square error, average difference, peak signal to noise ratio, normalized cross section, structural content and normalized absolute error, we compare the methods of Edge Detection, Thresholding, Deformable Model and Clustering techniques on dental images.