A region-based image retrieval system using salient point extraction and image segmentation

Abstract

Although most image indexing schemes are based on global image features, they have limited capability because they cannot capture local variations of the image properly. In order to solve this problem, we propose a new region-based image retrieval system. Since objects are important for image search in a huge database, we first find the important region including an interesting object using image segmentation and salient point extraction. We then find color and texture features in each important region. We have demonstrated that the color and texture information in the important region is very useful for improving performance of the image retrieval system.