Image re-ranking, Image Search engines mostly use keywords and they rely on surrounding text for searching images. Ambiguity of query images is hard to describe accurately by using keywords. Eg: Apple is query keyword then categories can be "red apple", "apple laptop" etc. In this paper, we have a tendency to propose a completely unique image re-ranking framework. Four steps: A query image is 1st classified into one in every of many predefined intention classes, and a particular similarity live is employed within every class to mix image options for re-ranking supported the query image. Query keywords are enlarged to capture user intention, through the visual content of the question image hand-picked by the user and the image agglomeration victimization fuzzy c mean algorithm, Image pool is enlarged to contain additional relevant pictures. The query image is additionally enlarged by victimization keyword growth. The Experimental analysis shows that our approach considerably improves the exactness of top-ranked pictures and conjointly the user expertise.