Abstract:

Image matching is a key technology of image processing applied to many fields. The image matching based on invariant features is the hot spot in image match research recently. SIFT is one of the most effective scale, rotation and illumination invariant features. However there are a lot of wrong match pairs produced by the original SIFT algorithm. Two specific types of wrong matches are analyzed and the corresponding eliminating methods are given. Aiming at general wrong match elimination, a method based on similar triangles is proposed. Experiments are carried out and the results confirm that: the proposed method is faster than the commonlyused two algorithms and it can eliminate wrong matches and preserve correct matches as well.

Key words: SIFT;wrong match pairs;elimination;similar triangles