Range image registration based on 2D synthetic images

Sukwoo Jung, Seunghyun Song, Minho Chang, Sangchul Park

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Proposed in this paper is a new procedure of range image (point cloud) registration reducing the risk of the original Iterative Closest Point (ICP) algorithm which gets stuck in a local minimum. To reduce this risk, it is essential to achieve the fine initial alignment of the given range images. The proposed procedure employs a 2D image registration method to achieve the fine initial alignment instead of a 3D feature based registration method. Although the 2D image registration method has also the risk of mismatching, we minimize the risk by increasing the number of test cases and selecting the answer that gains the most votes. To generate effective test cases, multiple 2D images are synthetized from the given range images by varying the orientation and resolution. The proposed procedure is implemented and tested for various examples, with experimental results showing that it outperforms existing methods.

Original languageEnglish
Pages (from-to)16-27
Number of pages12
JournalCAD Computer Aided Design
Volume94
DOIs
Publication statusPublished - 2018 Jan 1

Fingerprint

Image registration

Keywords

  • 3D registration
  • ICP
  • Image registration
  • Initial alignment
  • Range image
  • Scanning

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Industrial and Manufacturing Engineering

Cite this

Range image registration based on 2D synthetic images. / Jung, Sukwoo; Song, Seunghyun; Chang, Minho; Park, Sangchul.

In: CAD Computer Aided Design, Vol. 94, 01.01.2018, p. 16-27.

Research output: Contribution to journalArticle

Jung, Sukwoo ; Song, Seunghyun ; Chang, Minho ; Park, Sangchul. / Range image registration based on 2D synthetic images. In: CAD Computer Aided Design. 2018 ; Vol. 94. pp. 16-27.
@article{d39d344ce32c44358bd149205061f7d0,
title = "Range image registration based on 2D synthetic images",
abstract = "Proposed in this paper is a new procedure of range image (point cloud) registration reducing the risk of the original Iterative Closest Point (ICP) algorithm which gets stuck in a local minimum. To reduce this risk, it is essential to achieve the fine initial alignment of the given range images. The proposed procedure employs a 2D image registration method to achieve the fine initial alignment instead of a 3D feature based registration method. Although the 2D image registration method has also the risk of mismatching, we minimize the risk by increasing the number of test cases and selecting the answer that gains the most votes. To generate effective test cases, multiple 2D images are synthetized from the given range images by varying the orientation and resolution. The proposed procedure is implemented and tested for various examples, with experimental results showing that it outperforms existing methods.",
keywords = "3D registration, ICP, Image registration, Initial alignment, Range image, Scanning",
author = "Sukwoo Jung and Seunghyun Song and Minho Chang and Sangchul Park",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.cad.2017.08.001",
language = "English",
volume = "94",
pages = "16--27",
journal = "CAD Computer Aided Design",
issn = "0010-4485",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Range image registration based on 2D synthetic images

AU - Jung, Sukwoo

AU - Song, Seunghyun

AU - Chang, Minho

AU - Park, Sangchul

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Proposed in this paper is a new procedure of range image (point cloud) registration reducing the risk of the original Iterative Closest Point (ICP) algorithm which gets stuck in a local minimum. To reduce this risk, it is essential to achieve the fine initial alignment of the given range images. The proposed procedure employs a 2D image registration method to achieve the fine initial alignment instead of a 3D feature based registration method. Although the 2D image registration method has also the risk of mismatching, we minimize the risk by increasing the number of test cases and selecting the answer that gains the most votes. To generate effective test cases, multiple 2D images are synthetized from the given range images by varying the orientation and resolution. The proposed procedure is implemented and tested for various examples, with experimental results showing that it outperforms existing methods.

AB - Proposed in this paper is a new procedure of range image (point cloud) registration reducing the risk of the original Iterative Closest Point (ICP) algorithm which gets stuck in a local minimum. To reduce this risk, it is essential to achieve the fine initial alignment of the given range images. The proposed procedure employs a 2D image registration method to achieve the fine initial alignment instead of a 3D feature based registration method. Although the 2D image registration method has also the risk of mismatching, we minimize the risk by increasing the number of test cases and selecting the answer that gains the most votes. To generate effective test cases, multiple 2D images are synthetized from the given range images by varying the orientation and resolution. The proposed procedure is implemented and tested for various examples, with experimental results showing that it outperforms existing methods.

KW - 3D registration

KW - ICP

KW - Image registration

KW - Initial alignment

KW - Range image

KW - Scanning

UR - http://www.scopus.com/inward/record.url?scp=85030466548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030466548&partnerID=8YFLogxK

U2 - 10.1016/j.cad.2017.08.001

DO - 10.1016/j.cad.2017.08.001

M3 - Article

AN - SCOPUS:85030466548

VL - 94

SP - 16

EP - 27

JO - CAD Computer Aided Design

JF - CAD Computer Aided Design

SN - 0010-4485

ER -