Title:	HOMOGRAPHY USING TOPOLOGICAL INFORMATION FROM SCATTERED FEATURE POINTS
Author(s):	Rishu Gupta, Lee Byung Gook
ISBN:	978-972-8939-48-9
Editors:	Yingcai Xiao
Year:	2011
Edition:	Single
Keywords:	Homography, Correspondence Search, Delaunay triangulation, Simplex method, Normalized Correlation
Type:	Poster/Demonstration
First Page:	343
Last Page:	345
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	This paper proposes a method for finding homography between the left and the right images using correspondence points or the matching points calculated by topological information of feature points, which are extracted with the help of Harris corner detection algorithm. Delaunay triangulation is used to get information about scattered feature points and the simplex method is introduced to calculate neighborhood points of competent edges extracted from Delaunay triangulation. The competent edges are extracted using edge length as a parameter. In the final step, normalized correlation is performed by taking gray values of the image into account. The gray values of vertices that are selected using simplex method are taken and the normalized correlation is performed of vertices from the first image to the vertices of second image. The most suitable points are taken as correspondence points. And in the last step homography matrix is computed using correspondence points calculated.
	Go Back