An Intelligent Distributed Algorithm for Efficient Web Image Retrieval

Web image retrieval is a challenging task that requires efforts from image processing, web text retrieval, and link structure analysis. General web image retrieval engines such as Google and Yahoo retrieve images according to the meta-data or the context in which the images are found. Hence, the results search engines are far from expected regarding the semantics of the images and the user can be overwhelmed by thousands of results, of which few are valuable. In order to become an effective complement to traditional Web-scale textbased image retrieval, content-based image retrieval must address scalability and efficiency issues. This paper presents a framework for distributed content based web image retrieval.