论文信息 - An Innovative Method for Defect Detection in Digital Radiography Images

An Innovative Method for Defect Detection in Digital Radiography Images

194 Abstract—The process of detecting the weld defect plays a vital role in the Non Destructive Testing (NDT) filed. An Innovative method for extraction of defects in digital radiographic welding images using mathematical morphology is proposed in the present paper. The proposed algorithm initially uses wiener filter for enhancing the input weld image. Then the Raleigh probability distribution function is used to highlight the defects and the boundaries of the weld region. Next mathematical morphological top-hat transformation is used to extract the defects existing in the input weld image. Then, the Otsu’s global threshold method is used for binary conversion and the morphological opening followed by closing operations is used to smooth the detected defects. The experimental results clearly exhibit the efficacy of the proposed scheme.

Dr. T. Srikanth | Dr. V. Kamala

[1] Kamal Jamshidi,et al. Defect Detection of Tiles Using 2D-Wavelet Transform and Statistical Features , 2009 .

[2] Xin Wang,et al. Recognition of Welding Defects in Radiographic Images by Using Support Vector Machine Classifier , 2010 .

[3] Ramón Ruiz,et al. An adaptive-network-based fuzzy inference system for classification of welding defects , 2010 .

[4] Dimitrios Kosmopoulos,et al. Multiclass defect detection and classification in weld radiographic images using geometric and texture features , 2010, Expert Syst. Appl..

[5] Ramón Ruiz,et al. An automatic system of classification of weld defects in radiographic images , 2009 .

[6] Wu Xiaomeng,et al. Detection of Weld Line Defect for Oil-gas Pipeline Based on X-rays Image Processing , 2009 .

[7] Abdelhak Mahmoudi,et al. Welding Defect Detection by Segmentation of Radiographic Images , 2009, 2009 WRI World Congress on Computer Science and Information Engineering.