Depth-based Object Detection using Hierarchical Fragment Matching Method

Reza Haghighi, Mahdi Rasouli, Syeda Mariam Ahmed, Kim Pong Tan, Mohamed Almamun, Chee Meng Chew

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

Abstract

Identifying a workpiece in industrial processes using depth sensors has received increasing attention over the past few years. However, this is a challenging task particularly when the object is large or cluttered. In these scenarios, captured point clouds do not provide sufficient information to detect the object. To address this issue, we present a hierarchical fragment matching method for 3D object detection and pose estimation. We build a library of object fragments by scanning the object from different viewpoints. A descriptor, named Clustered Centerpoint Feature Histogram (CCFH), is proposed to compute the features for each fragment. The proposed method aims to enhance the robustness of the existing Clustered Viewpoint Feature Histogram (CVFH) descriptor. Subsequently, an Extreme Learning Machine (ELM) classifier is applied to identify the matched segments between the scene and the library of fragments. Finally, the pose of the object in the scene is estimated using the matched segments. Unlike existing approaches that require the CAD model of the object or pre-registration process, the proposed method directly use the scanned point clouds of the object. The experimental results are presented to illustrate the performance of the proposed method.

Original languageEnglish
Title of host publication2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018
PublisherIEEE Computer Society
Pages780-785
Number of pages6
Volume2018-August
ISBN (Electronic)9781538635933
DOIs
Publication statusPublished - Dec 4 2018
Event14th IEEE International Conference on Automation Science and Engineering, CASE 2018 - Munich, Germany
Duration: Aug 20 2018Aug 24 2018

Other

Other14th IEEE International Conference on Automation Science and Engineering, CASE 2018
CountryGermany
CityMunich
Period8/20/188/24/18

Fingerprint

Learning systems
Computer aided design
Classifiers
Scanning
Sensors
Object detection

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Haghighi, R., Rasouli, M., Ahmed, S. M., Tan, K. P., Almamun, M., & Chew, C. M. (2018). Depth-based Object Detection using Hierarchical Fragment Matching Method. In 2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018 (Vol. 2018-August, pp. 780-785). [8560427] IEEE Computer Society. https://doi.org/10.1109/COASE.2018.8560427

Depth-based Object Detection using Hierarchical Fragment Matching Method. / Haghighi, Reza; Rasouli, Mahdi; Ahmed, Syeda Mariam; Tan, Kim Pong; Almamun, Mohamed; Chew, Chee Meng.

2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018. Vol. 2018-August IEEE Computer Society, 2018. p. 780-785 8560427.

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

Haghighi, R, Rasouli, M, Ahmed, SM, Tan, KP, Almamun, M & Chew, CM 2018, Depth-based Object Detection using Hierarchical Fragment Matching Method. in 2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018. vol. 2018-August, 8560427, IEEE Computer Society, pp. 780-785, 14th IEEE International Conference on Automation Science and Engineering, CASE 2018, Munich, Germany, 8/20/18. https://doi.org/10.1109/COASE.2018.8560427
Haghighi R, Rasouli M, Ahmed SM, Tan KP, Almamun M, Chew CM. Depth-based Object Detection using Hierarchical Fragment Matching Method. In 2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018. Vol. 2018-August. IEEE Computer Society. 2018. p. 780-785. 8560427 https://doi.org/10.1109/COASE.2018.8560427
Haghighi, Reza ; Rasouli, Mahdi ; Ahmed, Syeda Mariam ; Tan, Kim Pong ; Almamun, Mohamed ; Chew, Chee Meng. / Depth-based Object Detection using Hierarchical Fragment Matching Method. 2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018. Vol. 2018-August IEEE Computer Society, 2018. pp. 780-785
@inproceedings{e104d2851a0e44c6a95dbdeff482582e,
title = "Depth-based Object Detection using Hierarchical Fragment Matching Method",
abstract = "Identifying a workpiece in industrial processes using depth sensors has received increasing attention over the past few years. However, this is a challenging task particularly when the object is large or cluttered. In these scenarios, captured point clouds do not provide sufficient information to detect the object. To address this issue, we present a hierarchical fragment matching method for 3D object detection and pose estimation. We build a library of object fragments by scanning the object from different viewpoints. A descriptor, named Clustered Centerpoint Feature Histogram (CCFH), is proposed to compute the features for each fragment. The proposed method aims to enhance the robustness of the existing Clustered Viewpoint Feature Histogram (CVFH) descriptor. Subsequently, an Extreme Learning Machine (ELM) classifier is applied to identify the matched segments between the scene and the library of fragments. Finally, the pose of the object in the scene is estimated using the matched segments. Unlike existing approaches that require the CAD model of the object or pre-registration process, the proposed method directly use the scanned point clouds of the object. The experimental results are presented to illustrate the performance of the proposed method.",
author = "Reza Haghighi and Mahdi Rasouli and Ahmed, {Syeda Mariam} and Tan, {Kim Pong} and Mohamed Almamun and Chew, {Chee Meng}",
year = "2018",
month = "12",
day = "4",
doi = "10.1109/COASE.2018.8560427",
language = "English",
volume = "2018-August",
pages = "780--785",
booktitle = "2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Depth-based Object Detection using Hierarchical Fragment Matching Method

AU - Haghighi, Reza

AU - Rasouli, Mahdi

AU - Ahmed, Syeda Mariam

AU - Tan, Kim Pong

AU - Almamun, Mohamed

AU - Chew, Chee Meng

PY - 2018/12/4

Y1 - 2018/12/4

N2 - Identifying a workpiece in industrial processes using depth sensors has received increasing attention over the past few years. However, this is a challenging task particularly when the object is large or cluttered. In these scenarios, captured point clouds do not provide sufficient information to detect the object. To address this issue, we present a hierarchical fragment matching method for 3D object detection and pose estimation. We build a library of object fragments by scanning the object from different viewpoints. A descriptor, named Clustered Centerpoint Feature Histogram (CCFH), is proposed to compute the features for each fragment. The proposed method aims to enhance the robustness of the existing Clustered Viewpoint Feature Histogram (CVFH) descriptor. Subsequently, an Extreme Learning Machine (ELM) classifier is applied to identify the matched segments between the scene and the library of fragments. Finally, the pose of the object in the scene is estimated using the matched segments. Unlike existing approaches that require the CAD model of the object or pre-registration process, the proposed method directly use the scanned point clouds of the object. The experimental results are presented to illustrate the performance of the proposed method.

AB - Identifying a workpiece in industrial processes using depth sensors has received increasing attention over the past few years. However, this is a challenging task particularly when the object is large or cluttered. In these scenarios, captured point clouds do not provide sufficient information to detect the object. To address this issue, we present a hierarchical fragment matching method for 3D object detection and pose estimation. We build a library of object fragments by scanning the object from different viewpoints. A descriptor, named Clustered Centerpoint Feature Histogram (CCFH), is proposed to compute the features for each fragment. The proposed method aims to enhance the robustness of the existing Clustered Viewpoint Feature Histogram (CVFH) descriptor. Subsequently, an Extreme Learning Machine (ELM) classifier is applied to identify the matched segments between the scene and the library of fragments. Finally, the pose of the object in the scene is estimated using the matched segments. Unlike existing approaches that require the CAD model of the object or pre-registration process, the proposed method directly use the scanned point clouds of the object. The experimental results are presented to illustrate the performance of the proposed method.

UR - http://www.scopus.com/inward/record.url?scp=85059987277&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059987277&partnerID=8YFLogxK

U2 - 10.1109/COASE.2018.8560427

DO - 10.1109/COASE.2018.8560427

M3 - Conference contribution

VL - 2018-August

SP - 780

EP - 785

BT - 2018 IEEE 14th International Conference on Automation Science and Engineering, CASE 2018

PB - IEEE Computer Society

ER -