Comparison of Color Features for Image Retrieval

Content based image retrieval (CBIR) systems are used for automatic indexing, searching, retrieving and browsing of image databases. Color is one of the important features used in CBIR systems. An experimental comparison of a number of different color descriptors for content-based image retrieval is presented in this paper. Color histograms, color moments and color coherent vector (CCV) are considered for retrieval. The primary goal is to determine which color descriptor or combination of color descriptors is most efficient in representing the similarity of color images. In this paper, we first present the comparison of individual color descriptors and then the comparison of combined color descriptors. For the experiments, five different publicly available image databases are used and the retrieval performance of the features is analyzed in detail. This allows for a direct comparison of all features considered in this work and furthermore it will allow a comparison of newly proposed features to these in the future. The article is concluded by stating which features perform well for what type of data.