A new color descriptor for content-based image retrieval: Application to COIL-100

In this study, a new technique for image retrieval based on color feature is presented. This new algorithm is based on two defined equations, the standard deviation and the mean of the image. The performance of the proposed approach is tested on a large image collection of the COIL-100 database. Experimental results show that the proposed approach can achieve significant precision (up to 95% better) and, at the same time, the system is fast.