Real-time X-ray Image Processing Based on Information Fusion for Weld Defects Detection
Real-time X-ray imaging and its image processing techniques are often adopted in on-line automatic inspection of defects. Due to high noises, low contrast and moving blur, it is difficult to realize automatic inspection using those image processing techniques. Furthermore, most of the existing automatic inspection systems only use single defect segment method, which leads to the outstanding conflict between false alarm and missed detection. As a result, those systems are hard to be applied in practice. In this paper, the redundancy and complementarity are analyzed in two defect segmentation methods of weld X-ray image, i.e., background subtraction and gray-level wave analysis. Then, the information fusion method of them is proposed. Results show that the false alarm and missed detection are decreased effectively by the information fusion method.
weld defects automatic inspection X-ray dynamic imaging information fusion
Dong DU Runshi HOU Jiaxin SHAO Li WANG Baohua CHANG
Key Laboratory for Advanced Materials Processing Technology,Ministry of Education, P.R.China,Department of Mechanical Engineering, Tsinghua University, Beijing 100084, China
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
158-163
2008-10-25（万方平台首次上网日期，不代表论文的发表时间）