Watershed-based textural image segmentation

The watershed transform is a well-established tool for image segmentation. However, watershed segmentation is often not effective for textural images. In this paper, we describe an improved watershed segmentation algorithm combined with texture features. The aim of this study is to improve the generalization of watershed techniques and to construct a well segmentation of textural images. The method includes two stages. The first stage is standard watershed algorithm. The second stage is processed by a clustering algorithm, fuzzy c-means (FCM). Watershed algorithm provides small homogenous patches which are merged by clustering algorithm based on texture features. The experimental results demonstrate that the combined algorithm is effective for textural image segmentation.