Diversity in image retrieval based on inferring user image search goals

Abstract

Image retrieval is an extremely tough process to retrieve the user perception based on the customer intervention from the dynamic data set. Various technologies are there to retrieve the images with the similar technique is mining user's search goals. In this paper we summarized all the existing technique and we introduced our improved user image search goal prediction and retrieving. To make this browsing process more efficient, image summarization is often needed to address this problem. In web search applications, users submit queries (i.e., some keywords) to search engines to represent their search goals. However, in many cases, queries may not exactly represent what they want. For this problem we analyzed various existing methodologies like visual, clustering, colors/shape based, image, feature based. Each method contains different output and features. In this paper we compared these methods and we proposed our work based on text and image based retrieval.