Fuzzy Similarity Thresholding Using Gamma Distribution

Abstract

Image thresholding which is used for separating the object from the background is one of the most common applications in image processing and analysis. Thresholding is a crucial tool, for the pre-processing purposes of an image. In this paper, Gamma distribution is applied as a new thresholding method based on minimizing the measures of fuzziness. The principle can be directly extended to multilevel thresholding. The effectiveness of this method is illustrated by using the test images of having various types of histograms. The experimental results for Gamma and Gaussian distribution are presented.