Parameter-free global hybrid point based range image registration

Linh Tao, Tinh Nguyen, Hiroshi Hasegawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In order to tackle one of the most difficult tasks in 3D computer vision, range image registration (RIR), this paper proposes a new approach to register two range images of the same object or scenario from different scanning angles. The new method of two dimensional point based boundary searching replaces conventional six dimension methods. This point based approach is applied into hybrid registration paradigm which integrates global searching algorithms with a local alignment tool, Iterative Closest Point (ICP). With this approach, searching algorithms are able to find global optima more efficiently with significantly fewer searching dimensions. Because of using data points as searching variables and taking all of them into consideration, registration algorithms are able to explore the whole space to find the global solution without any parameter constraint. Point based searching approach is successfully implemented on hybrid registration algorithms which use state-of-the-art global searching tools including Simulated Annealing (SA), Differential Evolution (DE) and Particle Swarm Optimization (PSO). The new approach is evaluated in terms of both accuracy and robustness in various experiments on different datasets to prove its superior over the conventional approach.

Original languageEnglish
Title of host publicationProceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017
PublisherAssociation for Computing Machinery
Pages108-112
Number of pages5
VolumePart F131200
ISBN (Electronic)9781450352956
DOIs
Publication statusPublished - 2017 Aug 25
Event2017 International Conference on Advances in Image Processing, ICAIP 2017 - Bangkok, Thailand
Duration: 2017 Aug 252017 Aug 27

Other

Other2017 International Conference on Advances in Image Processing, ICAIP 2017
CountryThailand
CityBangkok
Period17/8/2517/8/27

Fingerprint

Image registration
Simulated annealing
Particle swarm optimization (PSO)
Computer vision
Scanning
Experiments

Keywords

  • 3D registration
  • Global registration
  • Hybrid registration
  • ICP
  • Parameter free

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Tao, L., Nguyen, T., & Hasegawa, H. (2017). Parameter-free global hybrid point based range image registration. In Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017 (Vol. Part F131200, pp. 108-112). Association for Computing Machinery. https://doi.org/10.1145/3133264.3133285

Parameter-free global hybrid point based range image registration. / Tao, Linh; Nguyen, Tinh; Hasegawa, Hiroshi.

Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017. Vol. Part F131200 Association for Computing Machinery, 2017. p. 108-112.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tao, L, Nguyen, T & Hasegawa, H 2017, Parameter-free global hybrid point based range image registration. in Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017. vol. Part F131200, Association for Computing Machinery, pp. 108-112, 2017 International Conference on Advances in Image Processing, ICAIP 2017, Bangkok, Thailand, 17/8/25. https://doi.org/10.1145/3133264.3133285
Tao L, Nguyen T, Hasegawa H. Parameter-free global hybrid point based range image registration. In Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017. Vol. Part F131200. Association for Computing Machinery. 2017. p. 108-112 https://doi.org/10.1145/3133264.3133285
Tao, Linh ; Nguyen, Tinh ; Hasegawa, Hiroshi. / Parameter-free global hybrid point based range image registration. Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017. Vol. Part F131200 Association for Computing Machinery, 2017. pp. 108-112
@inproceedings{57f56e5be53a4d728c34c26c5f01f935,
title = "Parameter-free global hybrid point based range image registration",
abstract = "In order to tackle one of the most difficult tasks in 3D computer vision, range image registration (RIR), this paper proposes a new approach to register two range images of the same object or scenario from different scanning angles. The new method of two dimensional point based boundary searching replaces conventional six dimension methods. This point based approach is applied into hybrid registration paradigm which integrates global searching algorithms with a local alignment tool, Iterative Closest Point (ICP). With this approach, searching algorithms are able to find global optima more efficiently with significantly fewer searching dimensions. Because of using data points as searching variables and taking all of them into consideration, registration algorithms are able to explore the whole space to find the global solution without any parameter constraint. Point based searching approach is successfully implemented on hybrid registration algorithms which use state-of-the-art global searching tools including Simulated Annealing (SA), Differential Evolution (DE) and Particle Swarm Optimization (PSO). The new approach is evaluated in terms of both accuracy and robustness in various experiments on different datasets to prove its superior over the conventional approach.",
keywords = "3D registration, Global registration, Hybrid registration, ICP, Parameter free",
author = "Linh Tao and Tinh Nguyen and Hiroshi Hasegawa",
year = "2017",
month = "8",
day = "25",
doi = "10.1145/3133264.3133285",
language = "English",
volume = "Part F131200",
pages = "108--112",
booktitle = "Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Parameter-free global hybrid point based range image registration

AU - Tao, Linh

AU - Nguyen, Tinh

AU - Hasegawa, Hiroshi

PY - 2017/8/25

Y1 - 2017/8/25

N2 - In order to tackle one of the most difficult tasks in 3D computer vision, range image registration (RIR), this paper proposes a new approach to register two range images of the same object or scenario from different scanning angles. The new method of two dimensional point based boundary searching replaces conventional six dimension methods. This point based approach is applied into hybrid registration paradigm which integrates global searching algorithms with a local alignment tool, Iterative Closest Point (ICP). With this approach, searching algorithms are able to find global optima more efficiently with significantly fewer searching dimensions. Because of using data points as searching variables and taking all of them into consideration, registration algorithms are able to explore the whole space to find the global solution without any parameter constraint. Point based searching approach is successfully implemented on hybrid registration algorithms which use state-of-the-art global searching tools including Simulated Annealing (SA), Differential Evolution (DE) and Particle Swarm Optimization (PSO). The new approach is evaluated in terms of both accuracy and robustness in various experiments on different datasets to prove its superior over the conventional approach.

AB - In order to tackle one of the most difficult tasks in 3D computer vision, range image registration (RIR), this paper proposes a new approach to register two range images of the same object or scenario from different scanning angles. The new method of two dimensional point based boundary searching replaces conventional six dimension methods. This point based approach is applied into hybrid registration paradigm which integrates global searching algorithms with a local alignment tool, Iterative Closest Point (ICP). With this approach, searching algorithms are able to find global optima more efficiently with significantly fewer searching dimensions. Because of using data points as searching variables and taking all of them into consideration, registration algorithms are able to explore the whole space to find the global solution without any parameter constraint. Point based searching approach is successfully implemented on hybrid registration algorithms which use state-of-the-art global searching tools including Simulated Annealing (SA), Differential Evolution (DE) and Particle Swarm Optimization (PSO). The new approach is evaluated in terms of both accuracy and robustness in various experiments on different datasets to prove its superior over the conventional approach.

KW - 3D registration

KW - Global registration

KW - Hybrid registration

KW - ICP

KW - Parameter free

UR - http://www.scopus.com/inward/record.url?scp=85033579780&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85033579780&partnerID=8YFLogxK

U2 - 10.1145/3133264.3133285

DO - 10.1145/3133264.3133285

M3 - Conference contribution

VL - Part F131200

SP - 108

EP - 112

BT - Proceedings of 2017 International Conference on Advances in Image Processing, ICAIP 2017

PB - Association for Computing Machinery

ER -