Study on an Improved Robust Algorithm for Feature Point Matching

Image matching is one of the keystones of computer vision. Feature point matching is the most common one among all kinds of image matching. Its matching result is affected greatly by many factors, such as object occlusions, lighting conditions and noises. In this paper, it uses a combined corner detection algorithm to detect corners, and equably selects parts of feature point in the first image to implement pre-matching, and based on the matching result to find the correspondences points of the other detected points. Experiments show that the results of the algorithm in the paper had high right matching rate and it has less computation demand. (C) 2012 Published by Elsevier B. V. Selection and/or peer review under responsibility of ICMPBE International Committee.