SIFT特征在图像配准过程中的应用研究

针对现有基于特征的图像配准方法所存在的特征提取的多样性和相似度计算的复杂性等问题，提出了一种基于SIFT特征的图像配准方法．首先利用SIFT算法提取出图像的特征点，用欧式距离比进行特征匹配，然后利用图像位置的先验条件，采用RANSAC算法去除误匹配，最后计算出待配准图像和基准图像间的变换关系参数．实验结果证明了该算法的有效性．

The Image Registration Based on SIFT Features

For the diversity of feature extraction and the complexity of similarity calculation in the feature-based image registration methods, a Scale Invariant Feature Transform （SIFF） feature-based approach for image registration is proposed. First of all , by using the SIFT, the feature points of the images are extracted, and the feature points are matched according to the ratio of Euclidean distance. Then, by using the priory condition and the RANSAC algorithm the mismatch points are removed. At last, the relation parameters between the base image and the matching images are calculated. The experiments show that this method is very reliable.