About this Abstract |
Meeting |
6th International Congress on 3D Materials Science (3DMS 2022)
|
Symposium
|
6th International Congress on 3D Materials Science (3DMS 2022)
|
Presentation Title |
Noise, Sampling, and Phase: Understanding Spurious X-ray Signal |
Author(s) |
Matthew Andrew, Stephen T. Kelly, Andriy Andreyev, Ravikumar Sanapala, Robin White, William Harris, Hrishikesh Bale, William Fadgen |
On-Site Speaker (Planned) |
Stephen T. Kelly |
Abstract Scope |
We introduce two new technologies which significantly improve X-ray microscopy image quality and contrast through the removal of spurious or undesired reconstructed signal typical for traditional reconstruction algorithms. DeepRecon Pro is the first technique for the fully automated training of deep learning networks for X-ray reconstruction. The resulting networks can be used to remove or greatly reduce two major contributors to spurious reconstructed image signal; random noise and sparse sampling artefacts. We also introduce a technique which allows for the complete removal of propagation phase contrast artefacts in the volume domain, which can become the dominant contrast mechanism when imaging at high resolution or at low kV. The removal of this effect reveals the inherent reconstructed material contrast which can then be much more effectively denoised, segmented or analyzed. We demonstrate the utility of these technologies on samples including low-contrast features like graphite anode of a commercial lithium-ion battery. |
Proceedings Inclusion? |
Definite: Other |