A modified Watersheds Image Segmentation Algorithm for Blood Cell
Weixing Wang, Hao Song, Qi Zhao · 2006
In this paper, we present a new grayscale image segmentation algorithm, which combines edge- and region-based techniques through the morphological algorithm of watersheds. The method integrating top/bottom hat transformation with gray-level transformation and median filters is used as a preprocessing stage in order to reduce noise as well as augment the foreground. Then, an initial partitioning of the image into primitive regions is produced by applying the watershed transform. A region merging process based on the properties of watershed region is applied to control the oversegmentation. As additional contribution, we demonstrate a new method to establish the watershed regions based on the distance and elevation fall associated with pixels and minima and an approach to merge region to control oversegmentation according to the properties of initial watershed regions