Quantitative characterization of radiographic weld defect based on the ground truth radiographs made on a stainless steel plates

Abstract

This paper presents a new approach for quantification of radiographic defects. This approach is based on calculating the size of the pixel using the known image quality indicator present in the radiographic image. This method is first applied on the ground truth realities of different shapes whose size is known in advance. The proposed method is then validated with the defect (porosity) where the defect is quantified accurately. The image processing techniques applied on the radiographic image are contrast enhancement, noise reduction and image segmentation to quantify the defects present in the radiographic image. The image processing algorithms are validated using image quality parameter Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR).