An Image Segmentation Approach Based on Log-Normal Distribution

Abstract

image thresholding has a great importance in most image processing application due to its importance and effectiveness although its simplicity; it has a big issue in estimating the optimal threshold value for obtaining better segmentation quality. The objective of this study is to develop a thresholding method based on histogram Split-Merge technique and Log-Normal distribution. Using Log-Normal distribution to model histogram modes allow for better estimation of the threshold value. (Result) The proposed method is applied on different optical images. A good segmentation result is obtained. The experiment showed that the proposed method obtained very good results but it requires more testing on different types of images.