A Fast Matching Algorithm Based on Square-Block

Abstract

Image matching is a very important area of image processing. This paper presents a fast matching algorithm for image matching, based on square-block. In this novel algorithm for matching, first template is divided into certain sub templates called D-block, then eigen value of every sub template is calculated and the optimal one is chosen according to certain principle, and finally the best matching position for reference image is determined by adopting interlaced scanning during image matching processing. Experimental results show that the proposed algorithm is superior to the regular algorithm and has low computational complexity.