Optimized SIFT image matching algorithm

SIFT (scale invariant feature transform) is one of the most active research subjects in the field of feature matching algorithms at present. This algorithm can dispose of matching problem with translation, rotation and affine distortion between images and to a certain extent is with more stable feature matching ability of images which are shot from random different angles. but its algorithm is complicated and computation time is long. Method of precision orientation key-pots SIFT-based and image matching algorithm are analyzed, Eulidean distance is replaced by the linear combination of city-block distance and chessboard distance in computing process in this article. Matching algorithm is proposed to reduce the algorithms time and to improve the accuracy of image matching. The result shows that the improved algorithm was effective.