Image Segmentation via Gradient Watershed Hierarchies and Fast Region Merging
Thangavelu Saravanan, Venkatesh Srinivasan, R. Udayakumar · 2013
The watershed algorithm from mathematical morphology has recently become powerful tool for image segmentation. Image analysis has been used successfully in a number of applications to classify different features according to their relative scales. In this paper we present the multistage behavior of gradient watershed regions. Boundaries of gradient watershed regions correspond to the edges of objects in an image. Then an initial partitioning of the image into primitive regions is produced by applying the watershed transform on the image gradient magnitude. This initial segmentation is the input to computationally efficient hierarchical region merging process that produces the final segmentation.