期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2014
卷号：3
期号：12
页码：4268-4271
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Due to increased use of social sharing websites like Facebook, Flickr and Picasa, effective and efficient mechanism is needed to retrieve relevant images from large database. The paper presents a review on different ways of image retrieval techniques. The text-based approach is a simple keyword based search. Content Based image retrieval is a system by w hich vari ous images are retrieved from an image database. Content Based image retrieval system uses various visual features like color, shape, texture which are extracted using different techniques. Proposed approach contains fusion of Content and link based similarities. Also personalized image search approach is proposed w hich helps to improve searching experience in the terms of query relevance and user preference by considering images annotations. Proposed method provides better results in terms of relevance and speed in image retrieval process over traditional approaches.
关键词：Image mining; feature extra ction; image ; retrieval and Content based image retrieval
