Matching unorganized data sets using multi-scale feature points

Abstract

In order to match partly overlapped data clouds measured from different view point, a multi-scale feature points detecting algorithm was proposed. A few feature points can be extracted from large number of original data quickly. This algorithm consists of three steps: discrete curvature computing, bilateral filtering process and feature points detecting. The number of feature points can be controlled by scale parameter approximately. After we got two feature point sets, an exhaustive searching process was carried out for maximal congruent triangles between two feature point sets, with which rotation and translation matrix could be computed easily to register original data sets. Although the exhaustive search is a time-consuming process, we still got high running speed by controlling the number of feature points.