Sinogram Inpainting with Generative Adversarial Networks and Shape Priors

Abstract

X-ray computed tomography is a widely used, non-destructive imaging technique that computes cross-sectional images of an object from a set of X-ray absorption profiles (the so-called sinogram). The computation of the image from the sinogram is an ill-posed inverse problem, which becomes underdetermined when we are only able to collect insufficiently many X-ray measurements. We are here interested in solving X-ray tomography image reconstruction problems where we are unable to scan the object from all directions, but where we have prior information about the object’s shape. We thus propose a method that reduces image artefacts due to limited tomographic measurements by inferring missing measurements using shape priors. Our method uses a Generative Adversarial Network that combines limited acquisition data and shape information. While most existing methods focus on evenly spaced missing scanning angles, we propose an approach that infers a substantial number of consecutive missing acquisitions. We show that our method consistently improves image quality compared to images reconstructed using the previous state-of-the-art sinogram-inpainting techniques. In particular, we demonstrate a 7 dB Peak Signal-to-Noise Ratio improvement compared to other methods.

Sinogram Upsampling via Sub-Riemannian Diffusion with Adaptive Weighting

“…To fill in large missing regions, an auto-encoder type called the context encoder was proposed for image inpainting [16] and this context encoder was used in [17] for sinogram inpainting. Valat et al [18] incorporated the shape prior information into the GAN framework. One disadvantage of using a GAN is that it requires large computational resources and so it is difficult to scale to large data.…”