会议专题

Image processing techniques for ultrasonic inspection

Manual inspection of ultrasonic evaluation images can not only be a time-consuming and therefore expensive task, it can also suffer from operator performance. In this work we apply image processing techniques to automatically inspect ultrasonic C-scan images. Our method is based on the comparison of a C-scan image with a reconstructed reference image. The method was tested on a set of C-scan images with known defects and was able to detect all defects. The umber of false detections is still high and will be addressed in future research.

image processing ultrasonic inspection C-scan images automated interpretation ROC parameters

Hartmut KIECKHOEFER Jan BAAN Arjan MAST Arno W.F. VOLKER

TNO Industry & Technology, PO Box 155, 2600 AD Delft, The Netherlands

国际会议

第十七届世界无损检测会议(17th World Conference on Nondestructive Testing)

上海

英文

3027-3036

2008-10-25(万方平台首次上网日期,不代表论文的发表时间)