Automatic Detection of Internal Weld Defects on X-ray Radiographic Films.

• DOI: 10.2493/jjspe.64.573
• Publication Date: 2011-07-20T05:13:15Z
• Authors: Shinya SATO; Katsuhisa SHIBUKAWA; Takeshi KISHINAMI; Akira ISHII
• Keywords: Robustness

This report proposes a new method which can detect internal weld defect areas appearing in X-ray radiographic films, and verifies its efficiency using actual images. The has robustness against noise blurred images, because each pixel of image votes for an area decided by the direction gradient vector pixel. And classify into defective indefective thresholding voted image. results images are as follows, (1) more than 90% defects 125 detected, (2) 85% their characteristic shape found.