An Efficient Local Fuzzy Color and Global Color-texture Representation for Image Retrieval

An efficient local fuzzy color and global color-texture descriptor is proposed for content-based image retrieval. In our retrieval system, an image is represented by a set of color-clustering-based segmented regions and global color-texture descriptors. Each segmented region corresponds to an object or parts of an object and is represented by a fuzzified color feature. A fuzzy region matching scheme is then incorporated to address the issues associated with the color-related inaccuracies and segmentation-related uncertainties. The global colortexture features are also incorporated into our representation since they do not depend on the segmentation results. They are calculated by energydistribution in a dimensionality reduced P1P2 color space. The overall similarity measure is defined as a weighted combination between the regional and global image level similarity measures incorporating all features. Our proposed retrieval approach demonstrates a promising retrieval performance for an image database of 1000 general-purpose images from COREL, as compared with some peer systems in the literature.