Content-Based Image Retrieval Using Color Existence Features

The most popular technique for image retrieval in a database of color images is the comparison of images based on their color histogram. The color histogram describes the distribution of colors in the color space of a color image. In the most image retrieval systems, the color histogram is used to compute similarities between the query image and all the images in the database. But, small changes in the resolution, scaling, and illumination may cause important modifications of the color histogram, and so two color images may be considered to be very different from each other even though they have completely related semantics. It is a result that color features are only limited to frequencies of colors. When the color histogram is computed, color existence information is included but not used in fact. So we adapted this color existence property and could earn good results in image retrieval. In this paper, we propose two color existence representation methods and compare them with the traditional color histogram methods. The experimental results reveal that the proposed is less sensitive to small changes in the image and that achieve higher retrieval performances than the traditional color histograms.