Detecting Defects in Digital Radiographic Images


  • W. Al-Hameed Information Technology, Software Department, University of Babylon, Iraq
  • P. D. Picton School of Science & Technology, The University of Northampton, United Kingdom


Defect Detection, Flaws in the weld, NDT, X-ray image,


It has been noticed that digital x-ray images of faulty welds in pipes tend to be darker than the rest of the image. Rather than simple thresholding, in this work a light pixel is converted to white if there are light pixels within its immediate neighborhood. The effect is that the flaw appears black and the background appears white, this enabling the flaw to be easily detected. However, this method will have the effect of eroding any rough edges on the flaw i.e. black pixels that stick out from the main body of the flaw. This method works well for large flaws, while not with fine cracks.




How to Cite

Al-Hameed, W., & Picton, P. D. (2017). Detecting Defects in Digital Radiographic Images. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 9(2-12), 151–154. Retrieved from