Adaptive Image Retrieval Technique using Texture and Color Features

Abstract

CBIR is the area in image mining system which performs retrieval based on the similarity defined in terms of extracted features with more objectiveness. It aims at searching image databases for specific images that are similar to a given query image. In CBIR the features of the query image alone are considered and this is the drawback of this technique. Thus, a novel method based on clusters is emerged that improves user interaction with image retrieval systems by fully exploiting the similarity information. In order to reduce the searching time of images from the image database, the query image will be classified in this method. The target image is selected optimally according to the color characteristics with feature vectors which represent typical color distributions. The proposed technique aims to reduce the searching time of image retrieval and hence it improves the performance of image retrieval system.