A Robust Point-Matching Algorithm Based on Integrated Spatial Structure Constraint for Remote Sensing Image Registration

Abstract

Feature matching, which refers to finding the correct correspondences from two sets of features, is an important step in feature-based image registration. In this letter, an accurate and highly robust point-matching algorithm, which is called the integrated spatial structure constraint, is proposed. We establish a set of tentative correspondences using the scale-invariant feature transform algorithm and then focus on increasing the number of correct correspondences (inliers) and removing incorrect correspondences (outliers). First, a global structure constraint, i.e., the shape context, is constructed for each correspondence out of the tentative set to increase the number of inliers and raise the correct rate simultaneously. Then, a local structure constraint based on the triangle area representation is utilized on the neighboring points of each correspondence to remove outliers. Experimental results compared with four state-of-the-art methods demonstrate that the proposed algorithm is robust and can achieve preferable results in terms of both matching accuracy and quantity of inliers.