Abstract

Radiography is one of the most common and widely used non-destructive testing (NDT) technique in inspecting weld discontinuity in welded joints. Conventionally, radiography inspector is requires to do the inspection analysis manually on weld discontinuity based on visual characteristics such as location, shape, length and density. The results can be very subjective, time consuming and inconsistent. Hence, semi-automated inspection using digital image processing and segmentation technique can be applied for weld discontinuity detection. The goal of this work is to detect the weld discontinuity on digital radiographic image using Foreground Marker Controlled Watershed. It is usually implemented in image processing because it always generates closed contour for each region in the image. In this paper, image enhancement on radiographic image is aim to remove image noise and improve image contrast. Then, marker controlled watershed with foreground markers is applied on the image to detect the discontinuity. The accuracy of the technique is evaluated using Receiver Operating Characteristic (ROC) curve. The accuracy of the technique has been compared with the ground truth and the result shows that the accuracy is 67% and area under the curve is 0.7134. The application of image processing technique in detecting weld discontinuity is able to assist radiographer to improve the inconsistent results in evaluating the radiographic image.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call