Image matching method based on improved SURF algorithm

In recent years, SURF (Speeded Up Robust Feature) algorithm has gained great interest in image matching and self-localization or self-navigation of robots noted by its affine invariant property as well as its low computational complexity. Image matching method based on SURF algorithm has been widely used in many fields, such as computer vision, medical diagnosis, and treatment and image mosaic. In the process of matching, the efficiency of the traditional linear algorithm is low, we can use the method which constructing the k-d tree and using improved BBF algorithms to replace the linear algorithm to accelerate the speed of matching. The final results and analysis of error show that this method is simple and effective, BBF algorithm based on the k-d tree is obviously much faster than the conventional linear algorithm.