An Efficient Perceptual Color Indexing Method for Content-based Image Retrieval Using Uniform Color Space

摘要

Dominant Color Descriptor (DCD) is one of the famous descriptors in Content-based image retrieval (CBIR). Sequential search is one of the common drawbacks of most color descriptors especially in large databases. In this paper, dominant colors of an image are indexed to avoid sequential search in the database where uniform RGB color space is used to index images in LUV perceptual color space. Proposed indexing method will speed up the retrieval process where the dominant colors in query image are used to reduce the search space. Additionally, the accuracy of color descriptors is improved due to this space reduction. Experimental results show effectiveness of the proposed color indexing method in reducing search space to less than 25% without degradation the accuracy.