Object template matching using topological models

Abstract

Object extraction from complex scene of image is important for computer vision and pattern recognition. However, it is difficult to measure visual change due to environment or different viewpoint in an image. This work concerned with very simple and efficient method to measure the similarity contents between object from query image and database images. In this method first desired object is extracted from query image, after extracted object is matched with the images in the database on the basis of colour and texture. In this paper an interactive topological model based object extraction and matching are proposed, which is based on two main important properties (i) initially low level segmentation and (ii) similarity measures among query and database images. The extensive experiment has been perform to the efficiency and effectiveness of proposed method.