Euclidean Distance Based Color Image Segmentation Algorithm for Dimensional Characterization of Lack of Penetration from Weld Thermographs for on-Line Weld Monitoring in GTAW

Conventional Non Destructive Testing (NDT) techniques for assessing the weld quality are applied after welding is completed. It results in wastage of time, material and manpower. These inherent limitations of the conventional welding processes can be overcome with an automated adaptive welding system to correct the deviation in the welding current and torch speed to provide defect free welds. This system requires an on-line weld-monitoring sensor, efficient image processing algorithm for defect identification and neurofuzzy control software for correlating the defect characteristics with deviations in physical parameters. Infrared Thermography is the best-suited sensor for on-line weld monitoring. It monitors the surface temperature distributions of the plates being welded and produces thermal maps called thermographs. The image processing algorithm for hot spot identification must be a generalized and standardized algorithm that works well for all different frames depicting different percentages of lack of penetration. Moreover time consumed by the algorithm must also be less to suit for on-line weld monitoring. This paper proposes an image processing algorithm that effectively identifies and quantifies the hotspot. The hot spot is then characterized using statistical moments, major axis length, minor axis length and area.