Affiliations: Department of Engineering Physics, Tsinghua
University, Beijing, 100084, China. E-mail: gujianwei99@mails.tsinghua.edu.cn,
{zhangli,chenzhiqiang}@nuctech.com, xingyx@mail.tsinghua.edu.cn | Department of Application Science, Nuctech Company
Limited, Beijing, 100084, China. E-mail: yuguoqiang@nuctech.com
Abstract: We present a sinogram inpainting approach for X-ray computed
tomography (CT) with the intent of reducing metal artifacts, where
"inpainting" is a synonym for "image interpolation". Metal artifacts arise
in CT when rays traverse the high attenuating objects such as metal bodies and
portions of projection become unavailable. In this paper, we propose a sinogram
inpainting method based on Euler's elastica and curvature for metal artifacts
reduction, in which the unavailable data are regarded as an occlusion and can
be inpainted inside the inpainting domain based on elastica interpolants.
Numerical simulations demonstrate that, compared to conventional interpolation
methods, the algorithm proposed connects the unavailable projection region more
smoothly and accurately, and thus better reduces metal artifacts and more
accurately reveals cross section structures, especially in the immediate
neighborhood of the metallic objects.
Keywords: Curvature, elastica, inpainting, metal artifacts, X-ray computed tomography (CT)
