Prof.
Martin Burger
(Helmholtz Imaging, DESY)
9/16/24, 11:45 AM
Course 10 (Helmholtz Imaging): Regularization in Image Reconstruction: From Model to Data Driven Methods
Lukas Weigand
(Helmholtz Imaging, DESY),
Samira Kabri
(Helmholtz Imaging, DESY),
Tim Roith
(Helmholtz Imaging, DESY)
9/16/24, 1:30 PM
Course 10 (Helmholtz Imaging): Regularization in Image Reconstruction: From Model to Data Driven Methods
In the first part of the tutorial we briefly introduce the world of inverse problems. Most importantly we learn about the radon transform and regularization for inverse problems. Small hands on tasks throughout the tutorial challenge your understanding of the material.
Lukas Weigand
(Helmholtz Imaging, DESY),
Samira Kabri
(Helmholtz Imaging, DESY),
Tim Roith
(Helmholtz Imaging, DESY)
9/17/24, 9:30 AM
Course 10 (Helmholtz Imaging): Regularization in Image Reconstruction: From Model to Data Driven Methods
In the second part of the tutorial we learn about deep learning for inverse problems. We start with an end-to-end approach emplyoing a U-Net to then introduce the concept of plug-and-play regularization. We finally introduce the basics of sampling and uncertainty quantification. Again, these insights are applied to CT reconstruction in small hands on tasks.
