Development of a spatter index for automated welding inspection using computer vision

Spatter is a cosmetic defect in the welding process, caused by small globules of the weld metal, which solidify on the workpiece, outside the desired weld area. Visual inspection of spatter is very tedious and time consuming. This paper describes the use of a computer vision system to estimate spatt...

Full description

Saved in:
Bibliographic Details
Published in:Computers & industrial engineering 1989, Vol.16 (2), p.215-224
Main Authors: Bidanda, B., Rubinovitz, J., Raman, S.
Format: Article
Language:English
Subjects:
Inspections
Mathematical models
Quality control
Vision systems
Welding
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Spatter is a cosmetic defect in the welding process, caused by small globules of the weld metal, which solidify on the workpiece, outside the desired weld area. Visual inspection of spatter is very tedious and time consuming. This paper describes the use of a computer vision system to estimate spatter in but welds produced by gas metal arc welding (GMAW). Accordingly, an index was used to quantify this spatter. Tests were run on HR1020 steel plates welded by a GMAW process. Based on the data acquired, it was concluded that this spatter index is relatively insensitive to changes in vision system threshold values. It was also proved that the method for quantification of spatter is highly reliable. Different illumination arrangements for the spatter index evaluation were tested, with best results being achieved when using an incandescent light at low angle.
ISSN:0360-8352
1879-0550
DOI:10.1016/0360-8352(89)90140-X