Effective image search and retrieval through automatic image annotation

Effective and efficient image search tools are urgently needed to make use of ever increasing amount of images being captured and stored. Current image search engines are not adequate because they are either based on text by assuming the existence of text description for all images, or based on low level image features which do not reflect the true semantic contents of images. This project proposes and develops a semantic based image retrieval system to overcome the problems of existing search engines by automatically annotating and indexing image semantics. The developed system will be useful in all industries where images are heavily used, including libraries, museums, education, entertainment, media, medicine and design industries.