Novel color feature representation and matching technique for content-based image retrieval

Color features of an image are the most widely used features in content-based image retrieval (CBIR) systems. Specifically histogram-based algorithms are considered to be effective for color image indexing. Color histogram describes the global distribution of pixels of an image which is insensitive to variations in scale and easy to calculate. However, the high dimensionality of feature vectors results in high computation cost and space cost. In this paper, we mainly focus on color features and propose a novel method named color frequency sequence difference (CFSD) to express color images, which only has one numerical value in one color channel. The CFSD is combined with information entropy to realize indexing. The novel approach is described in detail and compared with color histogram method presented in the literature. The experiment is finished and shows that the method proposed in this paper is effective and efficient.