Stabilizing deep tomographic reconstruction: Part B. Convergence analysis and adversarial attacks

Abstract

Due to lack of the kernel awareness, some popular deep image reconstruction networks are unstable. To address this problem, here we introduce the bounded relative error norm (BREN) property, which is a special case of the Lipschitz continuity. Then, we perform a convergence study consisting of two parts, (1)a heuristic analysis on the convergence of the analytic compressed iterative deep (ACID) scheme (with the simplification that the CS module achieves a perfect sparsification), and (2)a mathematically denser analysis (with the two approximations).

Weiwen Wu
Weiwen Wu
Associate Professor of Biomedical Engineering

My research interests include computed tomography, image reconstruction and deep learning.