Improving Limited Angle CT Reconstruction with a Robust GAN Prior
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Limited angle CT reconstruction is an under-determined linear inverse problem that requires appropriate regularization techniques to be solved. In this work we study how pre-trained generative adversarial networks (GANs) can be used to clean noisy, highly artifact laden reconstructions from conventional techniques, by effectively projecting onto the inferred image manifold. In particular, we use a robust version of the popularly used GAN prior for inverse problems, based on a recent technique called corruption mimicking, that significantly improves the reconstruction quality. The proposed approach operates in the image space directly, as a result of which it does not need to be trained or require access to the measurement model, is scanner agnostic, and can work over a wide range of sensing scenarios.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- DOE Contract Number:
- AC52-07NA27344
- OSTI ID:
- 1598955
- Report Number(s):
- LLNL-TR-789817; 987684; TRN: US2102749
- Resource Relation:
- Conference: 33.Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver (Canada), 8-14 Dec 2019
- Country of Publication:
- United States
- Language:
- English
Similar Records
Learning functional priors and posteriors from data and physics
Physics-driven learning of Wasserstein GAN for density reconstruction in dynamic tomography