Texture Segmentation by Fuzzy Clustering of Spatial Patterns
An approach to perceptual segmentation of textured images by fuzzy clustering of spatial patterns is proposed in this paper. The dissimilarity between a texture feature, which is modeled as a spatial pattern, and each cluster is calculated as a combination of the Euclidean distance in the feature space and the spatial dissimilarity, which reflects how much of the pattern’s neighborhood is occupied by other clusters. The proposed algorithm has been applied to the segmentation of texture mosaics. The results of comparative experiments demonstrate that the proposed approach can segment textured images more effectively and provide more robust segmentations.
KeywordsFuzzy Cluster Synthetic Aperture Radar Texture Image Markov Random Field Texture Segmentation
Unable to display preview. Download preview PDF.
- 5.Brodatz, P.: Texture: A Photographic Album for Artists and Designers. Dover, New York (1966)Google Scholar