A novel overlapping mice macrophages images segmentation method

Abstract

Aiming at the overlapping phenomenon that exists among the cells in the actual mice macrophages image, this paper presents the method of image segmentation based on Otsu threshold and watershed of distance transform. The image is first segmented by Otsu threshold algorithm. Then erosion and dilation morphological operators are used to get rid of noises. Finally, the watershed algorithm based on distance transform is carried out to segment the image. Experimental results show this method works successfully in the segmentation of mice macrophages microscopic images.