Stable position and pose estimation of industrial parts using evaluation of observability of 3D vector pairs

Shuichi Akizuki, Manabu Hashimoto

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

This paper introduces a stable 3D object detection method that can be applied to complicated scenes consisting of randomly stacked industrial parts. The proposed method uses a 3D vector pair that consists of paired 3D vectors with a shared starting point. By considering the observability of vector pairs, the proposed method has achieved high recognition performance. The observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, a few vector pairs that are effective for recognition are automatically extracted froman object model, and then used for the matching process. Experiments have confirmed that the proposed method improves the 88.5% recognition success rate of previous state-of-the-art methods to 93.1%. The processing time of the proposed method is fast enough for robotic bin-picking.

Original languageEnglish
Pages (from-to)174-181
Number of pages8
JournalJournal of Robotics and Mechatronics
Volume27
Issue number2
Publication statusPublished - 2015
Externally publishedYes

Fingerprint

Observability
Bins
Robotics
Processing
Experiments

Keywords

  • Bin-picking
  • Point cloud data
  • Position and pose estimation
  • Vector pair

ASJC Scopus subject areas

  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

Stable position and pose estimation of industrial parts using evaluation of observability of 3D vector pairs. / Akizuki, Shuichi; Hashimoto, Manabu.

In: Journal of Robotics and Mechatronics, Vol. 27, No. 2, 2015, p. 174-181.

Research output: Contribution to journalArticle

@article{cb74f2dc94a5417584b6edb3be468307,
title = "Stable position and pose estimation of industrial parts using evaluation of observability of 3D vector pairs",
abstract = "This paper introduces a stable 3D object detection method that can be applied to complicated scenes consisting of randomly stacked industrial parts. The proposed method uses a 3D vector pair that consists of paired 3D vectors with a shared starting point. By considering the observability of vector pairs, the proposed method has achieved high recognition performance. The observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, a few vector pairs that are effective for recognition are automatically extracted froman object model, and then used for the matching process. Experiments have confirmed that the proposed method improves the 88.5{\%} recognition success rate of previous state-of-the-art methods to 93.1{\%}. The processing time of the proposed method is fast enough for robotic bin-picking.",
keywords = "Bin-picking, Point cloud data, Position and pose estimation, Vector pair",
author = "Shuichi Akizuki and Manabu Hashimoto",
year = "2015",
language = "English",
volume = "27",
pages = "174--181",
journal = "Journal of Robotics and Mechatronics",
issn = "0915-3942",
publisher = "Fuji Technology Press",
number = "2",

}

TY - JOUR

T1 - Stable position and pose estimation of industrial parts using evaluation of observability of 3D vector pairs

AU - Akizuki, Shuichi

AU - Hashimoto, Manabu

PY - 2015

Y1 - 2015

N2 - This paper introduces a stable 3D object detection method that can be applied to complicated scenes consisting of randomly stacked industrial parts. The proposed method uses a 3D vector pair that consists of paired 3D vectors with a shared starting point. By considering the observability of vector pairs, the proposed method has achieved high recognition performance. The observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, a few vector pairs that are effective for recognition are automatically extracted froman object model, and then used for the matching process. Experiments have confirmed that the proposed method improves the 88.5% recognition success rate of previous state-of-the-art methods to 93.1%. The processing time of the proposed method is fast enough for robotic bin-picking.

AB - This paper introduces a stable 3D object detection method that can be applied to complicated scenes consisting of randomly stacked industrial parts. The proposed method uses a 3D vector pair that consists of paired 3D vectors with a shared starting point. By considering the observability of vector pairs, the proposed method has achieved high recognition performance. The observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, a few vector pairs that are effective for recognition are automatically extracted froman object model, and then used for the matching process. Experiments have confirmed that the proposed method improves the 88.5% recognition success rate of previous state-of-the-art methods to 93.1%. The processing time of the proposed method is fast enough for robotic bin-picking.

KW - Bin-picking

KW - Point cloud data

KW - Position and pose estimation

KW - Vector pair

UR - http://www.scopus.com/inward/record.url?scp=84927923104&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84927923104&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84927923104

VL - 27

SP - 174

EP - 181

JO - Journal of Robotics and Mechatronics

JF - Journal of Robotics and Mechatronics

SN - 0915-3942

IS - 2

ER -