Efficient content based image retrieval through sector histogram

Abstract

This paper deals with a novel technique that utilizes a weighted undirected graph for each color, called GHSG, and applied in landscape images including three steps: Calculating the similarity between query image and all database images based on global color histogram method. Making a set of images includes all the images that are filtered from the image database. The similarity of query image with each image in the above image set is calculated according to a method which constructs sector graph for each color. We carried out an experiment on an image database containing 7,560 landscape images. Experimental result shows that our technique is more effective than the local color histogram and cell/color histograms.