Research on image matching based on improved RANSAC-SIFT algorithm
Mingfu Zhao, Haijun Chen, Tao Song, Sixing Deng · 2017
In order to reduce the false matching rate and matching time, an improved algorithm based on RANSAC-SIFT was proposed. The feature points were extracted by SIFT algorithm firstly. Then most of the mismatching points were eliminated according to the constraints that matching distances tend to be consistent. Finally the remaining points were regarded as pre matching points for achieve fine matching with RANSAC algorithm. The experimental results show that the improved RANSAC algorithm has much improved the matching accuracy and compression matching time.