Automated detection of welding defects in pipelines from radiographic images DWDI

详细信息    查看全文

• 作者：Neury Boaretto^a ; ^b ; ^{neury@ifsc.edu.br}Author Vitae ; Tania Mezzadri Centeno^a ; ^{mezzadri@utfpr.edu.br}Author Vitae
• 关键词：Pattern recognition ; Non-destructive testing ; Descontinuities classification ; Artificial neural network ; Welding defects detection
• 刊名：NDT & E International
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：86
• 期：Complete
• 页码：7-13
• 全文大小：813 K
• 卷排序：86

文摘

An automatic method for detection of flaws in welded joints is proposed. The experiments were performed on DWDI real field radiographic images. The proposed method uses image processing techniques and a MLP classifier. Classification of the samples in two classes reached an F-score of 87.5%. This work is an improvement of the automatic detection of welding defects.