A Content-Based Image Meta-Search Engine using Relevance Feedback

Search engines are the most powerful resources for finding information on the rapidly expanding World-Wide Web. Finding the desired search engines and learning how to use them, however, can be very time consuming. Metasearch engines, which integrate a group of such search tools, enable users to access information across the world in a transparent and more efficient manner. The recent emergence of visual information retrieval (VIR) systems on the Web is leading to the same efficiency problem. This paper describes MetaSEEk, a meta-search engine used for retrieving images based on their visual content on the Web. MetaSEEk is designed to intelligently select and interface with multiple on-line image search engines by ranking their performance for different classes of user queries. User feedback is also integrated in the ranking refinement. MetaSEEk has been developed to explore the issues involved in querying large, distributed, on-line visual information system sources. We compare MetaSEEk with the previous version of the system and a base line meta-search engine. The base line system does not use the feedback of previous searches in recommending target search engines for future queries.