Hue assisted registration of 3D point clouds

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

This paper presents a variant of the Iterative Closest Point (ICP) algorithm for merging multiple color point clouds generated from a mobile 3D Light Detection and Ranging (LIDAR) System. This algorithm uses hue information generated from a camera along with the coordinates of the scan points and enables high accuracy registration of point clouds. A k-d tree based nearest neighbor search associates corresponding colored points in 4-D space between data and model point clouds. Singular Value Decomposition (SVD) method solves for the rigid rotation and translation. Experimental results illustrate that 3D color point clouds accelerate the 3D map registration if the hue data and model point clouds have sufficient hue distribution and the imaging sensor robustly captures the hue.

Original languageEnglish
Title of host publicationASME 2010 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2010
Pages1075-1083
Number of pages9
EditionPARTS A AND B
DOIs
StatePublished - 2010
EventASME 2010 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2010 - Montreal, QC, Canada
Duration: 15 Aug 201018 Aug 2010

Publication series

NameProceedings of the ASME Design Engineering Technical Conference
NumberPARTS A AND B
Volume3

Conference

ConferenceASME 2010 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2010
Country/TerritoryCanada
CityMontreal, QC
Period15/08/1018/08/10

Fingerprint

Dive into the research topics of 'Hue assisted registration of 3D point clouds'. Together they form a unique fingerprint.

Cite this