Fast Automatic Matching Method on Multi-view Point Clouds For Rail Tanker

摘要

In this paper, a novel method is proposed for automatically matching the measured data of a rail tanker. The spindle orientation of point cloud is determined by PCA, and then any relative point pair is well matched according to the spatial topology relations as the complexity reduction point of view, finally the data is mapped to a uniform coordinate system with the Quaternion method. The experimental results demonstrated that the proposed method achieves simplicity of implementation with robustness, and especially suitable for positioning the work piece of the rail tanker and the multi-view data fusion.