A new method for remote sensing image matching by integrating affine invariant feature extraction and RANSAC

A new method on remote sensing image matching by integrating affine invariant feature extraction and RANSAC is presented. The novelty of this method is a strategy on automatic optimization for affine invariant feature matching based on RANSAC. An automatic way to determine the distance threshold of RANSAC is proposed, which is a key problem to implement this RANSAC-based automatic optimization. Since affine invariant feature matching technology has been successfully applied to remote sensing image matching, we design an experiment to compare the proposed method (with optimization) with the standard affine invariant feature matching (without optimization). By using two stereo pairs with different types of imagery, the experiment indicates that the proposed method can always get much matching score compared to the standard affine invariant feature matching method.