A fuzzy approach for retrieving images in databases using dominant color and texture descriptors

In this paper, a fuzzy approach for image retrieval on the basis of color and texture features is presented. For this purpose, two new descriptors ”Fuzzy Dominant Color Descriptor” and ”Fuzzy Dominant Texture Descriptor” are proposed. Each descriptor will be defined as a fuzzy set over a finite universe of color/texture fuzzy sets (with membership degrees according to its dominance degrees). The color/texture fuzzy sets are defined taking into account the relationship between the computational feature and its human perception. In addition, fuzzy operators over the new descriptors are proposed to define conditions in image retrieval queries. The proposed framework makes database systems able to answer queries using color-based and texture-based linguistic labels in natural language.