An Efficient Approach to Improve Retrieval Rate in Content Based Image Retrieval Using MPEG-7 Features

Abstract

Content Based Image Retrieval is a technique of automatic indexing and retrieving of images .In order to improve the retrieval performance of images, this paper proposes an efficient approach for extracting and retrieving color images. The block diagram of our proposed approach to content-based image retrieval (CBIR) is given first, and then we introduce Histogram Euclidean distance (L2 distance) ,Cosine Distance and Histogram Intersection which are used to measure the image level similarity . The combined features of finding a set of weights of different distances represented by mini-max algorithm is used to declare the best match to the query image. Query images are used to retrieve images similar to the query image .Comparison and analysis of performance evaluation for features and similarity measures for the retrieved images with ground truth are shown. This proposed retrieval approach demonstrates a promising performance. Experiment shows that combined features are superior to any one of the three.