Medical Image Registration Based on SURF and Improved RANSAC Algorithm
GU Zongyu · Zhongguo yixue yingxiangxue zazhi · 2014
To improve the medical image registration robustness, accuracy and speed, this paper proposes a medical image registration algorithm combining SURF with improved RANSAC algorithm. This algorithm first extracts SURF featured points from the images to match with similar featured points. Then the improved RANSAC algorithm is used to eliminate wrong matches. Finally the image registration process is accomplished by estimating space geometric varied parameters according to the matching point after purification. Experimental results show that compared with conventional algorithms, the proposed algorithm is more robust with higher speed while maintaining good registration accuracy when analyzing images with noise, non-uniform intensity and large scope of the initial misalignment.