An efficient Multiscale Morphological Watershed Segmentation using Gradient and Marker Extraction

A novel segmentation algorithm is proposed in this paper for the segmentation of digital images with either bright or dark or both type of features corrupted at very high impulse noise ratios. The specialized pre-processing set-up completely restores the impulse corrupted image and generates all the information required for perfect segmentation. The algorithm works on the marker extractions of the gradient images obtained by multiscale morphological reconstruction to avoid oversegmentation vivid in watershed algorithm. Experimental results add to the computational efficiency of the algorithm, its shape maintaining, edge preserving and scale-calibrating features. The performance is also superior to most other standard segmentation algorithms