Point Cloud Registration Based on ISS Feature Fusion NDT
Point cloud registration technology plays an important role in the field of three-dimensional reconstruction and automatic driv-ing.Aiming at the problem that the registration efficiency is low and the registration accuracy is greatly affected by the initial pose when the point cloud registration data of the normal distribution transform(NDT)algorithm are large,an internal shape signature(ISS)feature point fusion NDT point cloud registration method is proposed.Firstly,the ISS algorithm is used to extract the feature points of the point cloud,and the fast point feature histogram(FPFH)at the feature point is calculated to describe the local geometric information at the feature point.Then,the random sampling consistent initial registration algorithm(SAC-IA)is used for coarse registration.After coarse registration,the target point cloud and the source point cloud obtain an initial transformation pose.Finally,the initial transformation is provided to the NDT algorithm for fine registration.Through the comparative experiments on the data set under different algorithms,it is proved that the algorithm can effectively improve the registration accuracy and efficiency of point cloud.
automatic drivingISS feature pointsFPFH featuresSAC-IAnormal distribution transform algorithm
万方数据
维普
