Efficient content based image retrieval using SVM and color histogram

Abstract

The volume of the image databases have increased to such an extent that it is very difficult to retrieve the relevant images from the database efficiently. Content Based Image Retrieval methods use shape, color and texture features of the images and gives a more accurate method to search for the relevant images in the database. In this paper, we propose an approach for an efficient Content Based Image Retrieval which combines the use of Support Vector Machine (SVM) with color histogram. A two level approach is used by first filtering the relevant images using SVM followed by color based matching. The comparison of proposed image retrieval techniques is done based on the retrieval time and accuracy of the retrieved images.