A novel method for detection and quantification of incomplete penetration type flaws in weldments

摘要

This paper presents an image processing approach to process incomplete penetration type flaws in radiographic images of the weld specimens considering morphological aspects of the image. In the present approach, radiographic weld image is processed to get the image with good contrast and reduced noise using image enhancement techniques like histogram equalization and noise filtering. This is followed by an effective edge detector such as Canny operator to find the edges of flaws, which are further used to segment the image to fix the boundaries. Image segmentation is performed using morphological approach implemented through dilation of similar boundaries and erosion of dissimilar and irrelevant boundaries decided on the basis of pixel characteristics. The incomplete penetration type flaws are clearly identifiable in the finally processed image, which is obtained by superimposing the segmented image over the original enhanced image. The experimental results proved that proposed approach can be used as a reliable method for automatic detection of incomplete penetration types of flaws in radiographic weld images.