An Efficient Image Retrieval Approach base on Color Clustering

A color-based image retrieval is an approach for more efficient retrieval using description of image features. In this paper, a new color cluster algorithm is presented. The algorithm was developed based on HVS color space, idea of histogram intersection and combination of two similarity values. The algorithm has been coded into a program, which runs under Windows system. The approach is evaluated using a large image databases and proved to provide successful retrieval result supported by precision estimation.