Wafaa Al-Hameed - Segmentation of Radiographic Images of Weld Defect

Wafaa joined the School of Science and Technology University of Northampton as a PhD research visitor from University of Babylon on a six-moth visit. The purpose of her visit is to develop her PhD research further whilst at Northampton. 

Her research project was in the general area of machine recognition, in particular the automatic labelling of images using learning vector machines. After six month she has managed to present her work in the Graduate School Conference on May and then published her work in the Journal of Global Research in Computing.

Wafaa Al-Hameed[1], Yahya Mayali[2], and Phil Picton[3]
[1]Computer science, University of Babylon / College of Science, Babylon, Iraq
[2] Faculty of Mathematics and Computer Science, University of Kufa, Kufa, Iraq
[3] University of Northampton, Northampton, Uk
Journal of Global Research in Computer Science
Volume 4, No. 7, July 2013
ISSN 2229-371X

The first stage in the classification or identification of defects in gray-level x-ray images of welds is the segmentation of the defects. The gray levels in weld images depend on the density and thickness of the material being tested. This causes the relative contrast of the defect area to vary with its position. As a consequence, it is difficult to carry out the process of segmentation. As a result, the subsequent stages of operations such as classification or recognition are affected. In this paper, different segmentation methods are introduced which are known as “data-driven”. In this approach, only the gray-level data is used to identify an area of interest, i.e. an area of the image that contains a defect, and hence extract it. The comparison of results show that using the morphology process with local thresholding yields better results than using edge detection method such as Sobel and Canny filters.

Full text of the article can be found at: http://www.jgrcs.info/index.php/jgrcs/article/view/743/525