An Automatic Feature Based Registration Algorithm for Medical Images

Image registration involves the identification of corresponding control points in the images. As the manual identification of control points may be time-consuming and tedious, several automatic techniques have been developed. This paper proposes a new feature based automatic registration algorithm based on robust scale invariant feature transform (SIFT) features and Moving Least Squares (MLS). First, a robust SIFT algorithm is developed to find rotation and scale invariant features from the input images. Corresponding feature points are then matched using evidence accumulation. Finally the transformation parameters are estimated using Moving Least Squares (MLS) transformation. The effectiveness of the proposed method is demonstrated by the experimental results.