International Journal of Theoretical and Applied Mechanics

Tungsten Inclusion is the most commonly occurring weld defect in Gas Tungsten Arc Welding (GTAW). It is mainly due to high welding current and or the electrode momentarily touching the weld plates. Monitoring and controlling weld current can avoid the defect or if the defect is already intolerable the welding process can be stopped there to save time and money. It is thus necessary to develop an automated on-line welding system to make the correct decision. Weld thermographs are acquired on-line with IR camera. Effective feature extraction algorithms are to be developed to isolate and quantify the included Tungsten from thermographs. This paper compares the effectiveness and suitability of three different feature extraction algorithms namely conventional image processing, region growing and Euclidean distance based color image segmentation developed for on-line monitoring and control. Online weld monitoring necessitates a standardized feature extraction technique that works well irrespective of the size and shape of Tungsten inclusion. Hence comparison is based on the accuracy of the results, parameter independency and image independency. It is found that feature extraction by Euclidean distance based segmentation is best suited for on-line weld monitoring as it is parameter independent and can be standardized for a defect.