A Comparative Study of Different Segmentation Techniques for Detection of Flaws in NDE Weld Images

For detecting the flaws and their type in radiographic images (having different types of flaws), three segmentation techniques have been applied here. Every segmentation technique has its own advantages and disadvantages i.e. results obtained are of varying quality. Taking this into account, comparative study has been performed here to find the best segmentation technique for a particular type of flaw. These methodologies are compared and concluded to be effective for all possible nine types of weld flaws detection (Slag Inclusion, Worm Hole, Porosity Incomplete penetration, Under cuts, Cracks, Lack of fusion, Weaving fault Slag line), and feature extraction, after being successfully tested on more than 80 radiographic images obtained from EURECTEST, International scientific Association Brussels, Belgium, and 24 radiographs of ship weld provided by Technic Control Co. (Poland) were used, obtained from Ioannis Valavanis Greece, and 67 radiographs of actual weld test samples obtained from Bharat Heavy Electrical Limited (BHEL) India. The comparison with other NDT methods with application of image processing on the radiographic images of weld defects is aimed to detect the defects reliably and make accept/reject decision as per the international standard.