Title :
Application of neighborhood feature in point clouds registration
Author :
He Yong-Xing ; Ou Xin-Liang ; Kuang Xiao-Lan
Author_Institution :
Coll. of Comput. & Commun., Hunan Univ. of Technol., Zhuzhou, China
Abstract :
A point clouds registration method based on the invariant features of neighborhood points is proposed. Firstly, the target point set are weighted. Secondly, all the corresponding point-pairs are searched by comparing the feature value and then the preliminary registration is completed. Finally, to register accurately by using the ICP algorithm, this has improved the invariant feature. Experiments indicate the good results of the preliminary registration and the better results of the fine registration, which have met the requirement of registering point clouds from different viewpoints.
Keywords :
feature extraction; image registration; iterative methods; ICP algorithm; feature value; fine registration; invariant features; neighborhood feature; neighborhood points; point clouds registration method; point-pairs; preliminary registration; Accuracy; Algorithm design and analysis; Gravity; Iterative closest point algorithm; Noise; Registers; Vectors; Features of neighborhood; ICP algorithm; Point clouds; Registration;
Conference_Titel :
Computer Science and Network Technology (ICCSNT), 2011 International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4577-1586-0
DOI :
10.1109/ICCSNT.2011.6182093