Color Image Segmentation Based on Watershed Transform and Feature Clustering

Image segmentation is a hot topic of image processing. The challenging problem is how to preserve the weak edges as well as suppress the over-segmentation. In this paper, we proposed a novel image segmentation algorithm by combining the watershed transform and feature clustering. Firstly, the input image is pre-processed to suppress the noise and smooth the fin details. Secondly, a marker-based watershed transform is applied to segment the image into watershed regions. Finally, mean shift algorithm is exploited to cluster the watershed regions into the fina segmentation. The experiment results indicate that the proposed algorithm can obtain a relative better segmentation results compared with the state of the art works.