Similarity Searching In Statistical Figures Based On Extracted Meta Data

Abstract

Similarity searching is an excellent approach for getting information from subjective materials like images or videos. Some excellent works on special domains have done. We focus on statistical images. These kinds of images have some excellent features that can be clearly extractable and useable in similarity searching. But there no significant work has been done in this area. So we have done some preliminary works in this domain. By some extensive analysis we classify images of this domain in some sub domains and also identified the nature of features those can be considered as silent. We develop a prototype based on this analysis where we store extracted features information of a statistical images as meta data. Then we devise some strategy to do similarity searching using standard query formulation.