An efficient method for content based image retrieval using histogram graph

In this paper, we present a new technique to compare image similarity, called HISG (histogram graph), that utilizes a weighted undirected graph for each color, and its each vertex is a bin of histogram. As a result, the technique is able to take advantage of color location, and smaller space overhead but is not sensitive to rotation and translation. We carried out an experiment on an image database containing 12,960 landscape images. Experimental result shows that our technique is more effective than the local color histogram and cell/color histograms.