Electron tomography limited angular image reconstruction using directional-TV

Electron tomography (ET) is a powerful method for the characterization of materials at the nanoscale. However, ET reconstructed images often suffer from significant artifacts due to data often covering only a limited angular range, causing disruptive artifacts when using traditional reconstruction algorithms. Previously, the problem of limited angular reconstruction has been formulated as a convex optimization problem and the directional total-variation (DTV) algorithm has shown promising results for image reconstruction from constrained data in X-ray computed tomography (CT). In this work we apply the DTV algorithm to simulated and real ET data over various limited angular ranges. An overview of the DTV algorithm is provided before. The performance of the DTV algorithm is compared to a standard filtered backprojection (FBP) reconstruction. The results show the DTV algorithm is able to accurately recover images at significantly limited angular ranges and considerably reduces artifacts otherwise found in reconstructions using the FBP algorithm.