Defect sizing of post-irradiated nuclear fuels using grayscale thresholding in their radiographic images

Quantification of different types of material defects in a number of reference standard post-irradiated nuclear fuel image samples have been carried out by virtue of developing a computer program that takes radiographic images of the fuel as input. The program is based on user adjustable grayscale thresholding in the regime of image segmentation whereby it selects and counts the pixels having graylevel values less than or equal to the computed threshold. It can size the defects due to chipping in nuclear fuel, cracks, voids, melting, deformation, inclusion of foreign materials, heavy isotope accumulation, non-uniformity, etc. The classes of fuel range from those of research and power reactors to fast breeders and from pellets to annular and vibro-compacted fuel. The program has been validated against ground truth realities of some locally fabricated metallic plates having drilled holes of known sizes simulated as defects in them in which the results indicate that it either correctly selects and quantifies at least 94% of the actual required regions of interest in a given image or it gives less than 8.1% false alarm rate. Also, the developed program is independent of image size.