Automated detection of welding defects in pipelines from radiographic images DWDI期刊界 All Journals 搜尽天下杂志传播学术成果专业期刊搜索期刊信息化学术搜索

Automated detection of welding defects in pipelines from radiographic images DWDI

Affiliation:	1. Program on Electrical Engineering and Computer Science (CPGEI), Federal University of Technology of Paraná (UTFPR), Av. Sete de Setembro, 3165, Rebouças, CEP 80230-901, Curitiba, PR, Brazil;2. Department of Electronics, Federal Institute of Santa Catarina (IFSC), Rua Pavão, 1337, Costa e Silva, CEP 89220-200, Joinville, SC, Brazil;1. Imperial College, Department of Mechanical Engineering, London, United Kingdom;2. BAM Federal Institute for Materials Research and Testing, Berlin, Germany;1. School of Computing, SASTRA University, Thanjavur, Tamil Nadu, India;2. Indira Gandhi Centre for Atomic Research, Kalpakkam, Tamil Nadu, India;3. Bhabha Atomic Research Centre, Mumbai, Maharashtra, India;1. School of Computing, SASTRA University, Thanjavur, Tamil Nadu, India;2. Indira Gandhi Centre for Atomic Research, Kalpakkam, Tamil Nadu, India;3. Bhabha Atomic Research Centre, Trombay, Maharashtra, India;1. Industrial Tomography and Instrumentation Section, Isotope and Radiation Application Division, Bhabha Atomic Research Centre (BARC), Trombay, Mumbai, 400085, India;2. Homi Bhabha National Institute, Anushaktinagar, Mumbai, 400094, India;3. Reactor Engineering Division, Bhabha Atomic Research Centre (BARC), Trombay, Mumbai, 400085, India;4. Radiological Physics and Advisory Division, Bhabha Atomic Research Centre (BARC), Trombay, Mumbai, 400085, India;5. Radiopharmaceuticals Division, Bhabha Atomic Research Centre (BARC), Trombay, Mumbai, 400085, India

Abstract:	This paper presents a method for the automatic detection and classification of defects in radiographic images of welded joints obtained by exposure technique of double wall double image (DWDI). The proposed method locates the weld bead on the DWDI radiographic images, segments discontinuities (potential defects) in the detected weld bead and extracts features of these discontinuities. These features are used in a feed-forward multilayer perceptron (MLP) with backpropagation learning algorithm to classify descontinuities in “defect and no-defect”. The classifier reached an accuracy of 88.6% and a F-score of 87.5% for the test data. A comparison of the results with the earlier studies using SWSI and DWSI radiographic images indicates that the proposed method is promising. This work contributes towards the improvement of the automatic detection of welding defects in DWDI radiographic image which results can be used by weld inspectors as a support in the preparation of technical reports.

Keywords:	Pattern recognition Non-destructive testing Descontinuities classification Artificial neural network Welding defects detection
本文献已被 ScienceDirect 等数据库收录！

设为首页 | 免责声明 | 关于勤云 | 加入收藏