An Image Mosaic Method Based on Harris Corner Feature
Jie Cao · Informatization Research · 2010
This paper improved the stitching algorithm in feature point matching by analyzing the existing image mosaic algorithm.First,feature points are extracted by Harris corner detector in two images.Thenusing Normalized Cross Correlation(NCC) method to extract the initial matches,and the introduction of Mahalanobis distance achieves the precise image matching.Finally,the weighted average method is used to carry out the image fusion.The experimental results show that the method can effectively remove the false matching point pair of feature,reducing the probability of false matches,so it is an effective method of image mosaic.