Title :
Registration method for scattered cloud data based on feature points
Author :
Peng, Yuhui ; Gao, Chenghui ; Cheng, Jie
Author_Institution :
Coll. of Mech. Eng. & Autom., Univ. of Fuzhou, Fuzhou, China
Abstract :
Registration of scattered cloud data is an important topic in model construction. Firstly, local geometric characteristic of point clouds is expressed by point feature. Then, coarse registration is implemented based on feature points selected. By reducing search scope of matching points and accelerating search speed based on k-d tree, the conventional ICP algorithm is improved. Experiment results show that it is valid.
Keywords :
computational geometry; iterative methods; search problems; solid modelling; trees (mathematics); ICP algorithm; coarse registration; feature points; k-d tree; local geometric point cloud characteristics; model construction; scattered cloud data; search scope reduction; Genetic algorithms; Image reconstruction; Iterative closest point algorithm; Mice; Pattern analysis; Surface reconstruction; Three dimensional displays; feature point; registration; scattered cloud data;
Conference_Titel :
Mechanic Automation and Control Engineering (MACE), 2011 Second International Conference on
Conference_Location :
Hohhot
Print_ISBN :
978-1-4244-9436-1
DOI :
10.1109/MACE.2011.5988232
