An improved feature matching method base on gradient constraint

This article presents an improved feature matching method based on gradient constraint. SIFT is regarded as one of the most powerful features because of the conspicuous invariance of image rotation, noise and illumination. However, there will still be many mismatches when SIFT features are matched across images, especially when the amount of features is very large. In order to reduce the number of mismatches, we filter the initial matches based on the so-called gradient constraint, which requires the difference in dominant directors of corresponding features to be smaller than a threshold. The best threshold will be got by experience. Since the gradient of features can be obtained in the SIFT extraction process, no additional computation is involved and much time is saved. We then use RANSAC to remove false matches further based on the homography between two images. Experimental results indicate that our method is able to remove false matches effectively, and it is shown to be robust to noise.