Automatic detection of defective welding electrode tips using color segmentation and Hough circle detection

Chisung Kim, Dong Seog Han, Jin Kyoung Kim, Byoung Ik Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

The quality of a welded spot depends on the current, voltage, welding force, the welding duration and most importantly is the quality of the electrode tip that is used. Worthy of noting is that during welding, as the area of electrode contact area changes, current and force also changes. This paper proposes major defects like burr, dirt or small chip that represent a degradation in the quality of welded spots. There exist some similar researches to assess the quality of the electrode tip. However, those all are not targeting to detect specific region of the defects, but only to measure the size of the tip. A method using color segmentation and Hough circle detection is proposed in this paper to detect whether the electrode tip has any defect. The vision system detects the faults by combining a circle detection algorithm and a morphological algorithm.

Original languageEnglish
Title of host publicationProceedings of the 2016 IEEE Region 10 Conference, TENCON 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1371-1374
Number of pages4
ISBN (Electronic)9781509025961
DOIs
StatePublished - 8 Feb 2017
Event2016 IEEE Region 10 Conference, TENCON 2016 - Singapore, Singapore
Duration: 22 Nov 201625 Nov 2016

Publication series

NameIEEE Region 10 Annual International Conference, Proceedings/TENCON
ISSN (Print)2159-3442
ISSN (Electronic)2159-3450

Conference

Conference2016 IEEE Region 10 Conference, TENCON 2016
Country/TerritorySingapore
CitySingapore
Period22/11/1625/11/16

Keywords

  • computer vision
  • detecting fault
  • electrode tip

Fingerprint

Dive into the research topics of 'Automatic detection of defective welding electrode tips using color segmentation and Hough circle detection'. Together they form a unique fingerprint.

Cite this