基于SURF和改进的RANSAC算法的医学图像配准  

Medical Image Registration Based on SURF and Improved RANSAC Algorithm

作　　者：谷宗运[1,2] 谭红春[1] 殷云霞[1] 杜春敏[1] 

机构地区：[1]安徽中医药大学医药信息工程学院,安徽合肥230031 [2]安徽省计算机中医应用研究所,安徽合肥230038

出　　处：《中国医学影像学杂志》2014年第6期470-475,480,共7页Chinese Journal of Medical Imaging

基　　金：安徽省高等学校省级优秀青年人才基金项目(2012SQRL102);安徽中医药大学青年基金项目(2013QN005)

摘　　要：为了提高医学图像配准的鲁棒性、准确性和速度,本文提出一种结合加速鲁棒性特征(SURF)和改进的随机采样一致算法(RANSAC)的医学图像配准算法,首先提取图像上的SURF特征点,完成特征点初始匹配,然后用改进的RANSAC算法剔除误匹配点对,最后根据提纯后的匹配点对估计出两幅图像间的空间几何变换参数完成图像配准。实验结果表明,与传统的几种算法相比,在图像中含有噪声、灰度不均匀及初始变换范围比较大的情况下,该算法在达到良好配准精度的同时,具有较强的鲁棒性和更快的速度。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.

关 键 词：医学图像配准 图像处理 计算机辅助 加速鲁棒性特征 随机采样一致算法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]