Berlin 2024 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
O: Fachverband Oberflächenphysik
O 93: Scanning Probe Techniques: Method Development
O 93.10: Talk
Thursday, March 21, 2024, 17:15–17:30, MA 043
Total variation based image decomposition and denoising for microscopy images — •Marco Corrias1,2, Thomas Pock4, and Cesare Franchini1, 3 — 1University of Vienna, Faculty of Physics and Center for Computational Materials Science, Vienna, Austria — 2University of Vienna, Vienna Doctoral School in Physics, Vienna, Austria — 3University of Bologna, Department of Physics and Astronomy, Bologna, Italy — 4Institute of Computer Graphics and Vision, Graz University of Technology, Graz, Austria
Experimentally acquired images are affected by the ubiquitous presence of noise, which degrades their quality and hides their features. With the increase of image acquisition rate in recent years, modern denoising solutions have become necessary. This study focuses on microscopy image denoising, specifically those obtained through atomic force microscopy (AFM), scanning tunneling microscopy (STM), scanning electron microscopy (SEM), and scanning transmission electron microscopy (STEM). A total variation (TV)-based workflow is presented to automatically denoise microscopy images with different types of noise, proving that the Huber-ROF and TGV-L1 are effective to accomplish this task. Our results suggest a wider applicability of this method in microscopy, not only restricted to STM, AFM, SEM, and STEM images. The Python code used for this study will be made publicly available as part of the AiSurf package. It is designed to be integrated into experimental workflows for image acquisition or can be used to denoise previously acquired images.
Keywords: total variation; denoising; microscopy