Improved CBIR system using Edge Histogram Descriptor (EHD) and Support Vector Machine (SVM)

Abstract

Due to widespread applications found in many areas, Content Based Image Retrieval (CBIR) system is attracting attention of many researchers. Effectiveness of any CBIR system depends on the features extracted to represent an image. So feature extraction is the crucial step in design and development of any Content Based Image Retrieval system. Most commonly used features to represent images are Color, texture and shape. Recently developed CBIR system combines these features to effectively represent an image. This paper first discusses the concept and scope of content based image retrieval system. It also includes the overview of MPEG-7 edge histogram descriptor (EDH) to extract the contents from images. Further it gives the idea of Support Vector Machine (SVM) classifier. In this paper the basic CBIR system is developed by combining features like color moments, color-correlogram and Gabor texture features along with edge histogram descriptor. Further the results obtained are compared with CBIR system using SVM classifier.