Line Matching of Point Clouds Based on Encoded Asymmetric Boundary

Abstract

Feature extraction and matching of point clouds is an important issue of 3-D data processing. In this letter, the matching of the corresponding encoded asymmetric boundaries (EABs) is proposed, which is based on linear features, and the boundary line of planes is attained by Freeman differential coding and Hough transform, with two-level constructed indexing structure; EAB is not affected by rotation, translation, and scaling factors within the threshold of distance and angle means the achievement of the corresponding boundary lines matching simultaneously. A group of progressive experiments is undertaken to evaluate the robustness and efficiency of the proposed method. The experimental results have shown that the proposed method greatly improves the line matching efficiency and accuracy of point clouds, and it can be directly applied in the subsequent registration.