A novel method for detecting weld defects accurately and reliably in radiographic images

2016 ◽  
Vol 58 (1) ◽  
pp. 28-34 ◽  
Author(s):  
Changying Dang ◽  
Jianmin Gao ◽  
Zhao Wang ◽  
Yulin Xiao ◽  
Yalin Zhao
IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 180947-180964 ◽  
Author(s):  
Celia Cristina Bojarczuk Fioravanti ◽  
Tania Mezzadri Centeno ◽  
Myriam Regattieri De Biase Da Silva Delgado

2009 ◽  
Vol 42 (5) ◽  
pp. 467-476 ◽  
Author(s):  
Rafael Vilar ◽  
Juan Zapata ◽  
Ramón Ruiz

2019 ◽  
Vol 61 (10) ◽  
pp. 591-596 ◽  
Author(s):  
Yasheng Chang ◽  
Jianmin Gao ◽  
Hongquan Jiang

With the rapid development of industries such as nuclear power and shipbuilding, radiographic testing (RT) is widely used in these fields as an important means of weld inspection. It also produces a large number of radiographic films, which consume a great deal of manpower and material resources. It is therefore beneficial for the radiographic film to be digitised for storage and archiving. Text detection in RT weld images is an important prerequisite for the archiving of digitised films. This paper proposes a novel text detection method that employs mask convolution and frequency-domain filtering, which can detect text at different positions, with different fonts and of different sizes in RT weld images. The method is evaluated using 366 different images and shows significant efficacy for text detection in RT weld images, with the precision value reaching 96%. The method used in this paper is also compared with other methods that are commonly used in other fields and the results show that the proposed method gives improved results compared to state-of-the-art methods.


2013 ◽  
Vol 290 ◽  
pp. 71-77
Author(s):  
Wen Ming Guo ◽  
Yan Qin Chen

In the current industrial production, as steel weld X-ray images are low contrasted and noisy, the efficiency and precision can’t be both ensured. This paper has studied three different edge detection algorithms and found the most suitable one to detect weld defects. Combined with this edge detection algorithm, we proposed a new weld defects detection method. This method uses defect features to find the defects in edge images with morphological processing. Compared to the traditional methods, the method has ensured detection quality of weld defects detection.


2015 ◽  
Vol 28 (02) ◽  
pp. 124-130 ◽  
Author(s):  
Z. Dokic ◽  
D. Lorinson ◽  
J. P. Weigel ◽  
A. Vezzoni

SummaryObjective: To report a novel method of treating femoro-patellar instability in association with severe femoro-patellar osteoarthritis, by substituting the femoral trochlear with a patellar groove replacement prosthesis.Study design: Retrospective case series.Methods: Preoperative lameness was scored from 0–4, and radiographic studies including standard positions for patellar luxation were obtained for evidence of malalignment and femoro-patellar osteoarthritis. Cases with or without previous surgeries were included. The size of trochlear implant was determined by transparent templates and confirmed intra-operatively with trials. Radiographic images, together with clinical examinations, were reviewed immediately and at three months postoperatively and at longer term when available.Results: Thirty-five cases of patellar luxation ranging from grades II to IV were included. Eleven of these cases had prior surgical interventions which failed to stabilize the patella. Fourteen dogs required additional surgical procedures in conjunction with patellar groove replacement. Complications occurred in six patients, of which three required revision. Complete resolution of subjectively- assessed lameness was evident in 24/35 cases by the third month and in another seven of 35 patients on the longer term re-evaluations.Clinical significance: Use of a patellar groove replacement prosthesis has the potential to decrease the lameness associated with severe femoro-patellar arthritis, to improve patellar stability, and to correct the alignment of the extensor mechanism.


Sign in / Sign up

Export Citation Format

Share Document