Comparative Performance Evaluation of Edge Histogram Descriptors and Color Structure Descriptors in Content based Image Retrieval

Content based image retrieval (CBIR) system is broadly used for searching and browsing images from a large database by extracting the visual content of the images. Image database is build by feature vectors corresponding to texture, color, shape and spatial features. The MPEG-7 standards provide standardized tools to describe and search audio and video contents. In this paper we apply the edge histogram descriptors (EHD) and color structure descriptors (CSD) standardized by MPEG-7 standard to different kind of images for content based image retrieval. We have used a database of 1000images provided by Wang.et.al. Experimental results are shown by calculating and plotting Precision and Recall performance parameters for three approaches which include the retrieval by EHD and CSD method individually as well as a combined similarity measures.