Skip to Main Content
A fast corner matching algorithm based on circular region features is proposed. For each corner, the features are the gray values of the pixels in a certain radius of circle around the corner. After the translation registration of the features, the normalized cross correlation coefficient is calculated to match the corners. Then, the bidirectional consistency constraint, the gray values constraint and the clustering constraint of difference in position are applied to the matching corners in turn, which are used to eliminate the mismatching corners, in order to improve the accuracy of the corner matching. Experimental results on a mobile robot verify the effectiveness of the proposed algorithm.