An Area Based Image Matching Algorithm and Its Implementation

Abstract

Aiming at the problems of the global image matching, an area based on image matching algorithm is presented. In the paper, the principle of area based image matching is given firstly and the algorithm is analyzed in detail. In the algorithm, in order to diminish the influence of the segmentation to image matching, color key feature points are chosen as the feature of image and weighted distance measurement are used to achieve the coarse image matching and get candidate locations. Dominant color extraction to candidate locations according to human vision, the final result is achieved by precise matching which is measured by color measurement. A large number of experimental results demonstrate the feasibility and efficiency of the algorithm.