An OVR-SVM Based Machine Vision Evaluation Method for Standard Component Assembly

DOI

10.2991/ammee-17.2017.150How to use a DOI?

Keywords

Standard Component Assembly; Evaluation Method; Machine Vision; One versus Rest.

Abstract

In view of the current evaluation of the assembly quality of the standard components, the artificial visual method is used to judge it. However, when people do it repeatedly and in a high-intensity, which is difficult to meet the needs of large-scale industrial production. In this paper, according to production process, the OVR-SVM machine vision method for the assembly quality of standard components is proposed. First, it analyzes the evaluation of assembly quality of standard components and puts forward evaluation method of standard components assembly. Secondly, it uses the One Versus Rest (OVR) strategy to form two sets including correct and incorrect assembly, which can evaluate the quality of assembly standard components by only twice judging. Finally, it finishes the assembly quality evaluation of standard components for the machine testing based on SVM. Experimental results show, in this process, the accuracy of SVM classifier based on OVR strategy is 100%, and it has the characteristics of generalization ability of learning unknown sample, which can meet the needs of testing the assembly quality of the standard components in highly automated assembly line.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jian Huang
AU  - Ping Jia
AU  - Guixiong Liu
PY  - 2017/06
DA  - 2017/06
TI  - An OVR-SVM Based Machine Vision Evaluation Method for Standard Component Assembly
BT  - Proceedings of the Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017)
PB  - Atlantis Press
SP  - 776
EP  - 782
SN  - 2352-5401
UR  - https://doi.org/10.2991/ammee-17.2017.150
DO  - 10.2991/ammee-17.2017.150
ID  - Huang2017/06
ER  -