Visual Image Search

Traditional search engines are unable to capture the user’s intention when the query term is ambiguous. In order to bring effectiveness in search performance, reranking with the user interactions is highly needed. The main problem in active reranking is to target the user’s intention. In this paper, we have proposed Visual Image Search which is the best method for the image search. The main goal of the Visual Image Search is to target the user’s intention efficiently and effectively. This is achieved by using the active user interaction. This paper presents the idea that reranks the images resulting from initial text-only searches based on visual features and content similarity. A query image is first categorized into one of several predefined intention categories, and a specific similarity measure is used inside each category to combine image features for reranking based on the query image.