Imaging Lunch: Image Reconstruction 101: Computational Methods & Tools
Our in-house experts Sepand Kashani will give a workshop on Image Reconstruction 101: Computational Methods & Tools during our next imaging lunch. Open to all EPFL PhD students and Postdocs!
In many scientific and engineering fields, direct observation of the physical processes of interest is not possible. Instead, we rely on sensors to capture indirect measurements which are then used to reconstruct the quantity of interest. This reconstruction process combines sensor data with an understanding of the physics underlying the measurement process to "image" or visualize what cannot be directly observed.
Despite the diversity of measurement apparatuses and modalities (e.g., radio waves, visible light, X-rays), the underlying physical models in imaging problems fall into a few common categories.
This means that many imaging problems share common mathematical principles and can be addressed with a set of core computational techniques.
The goal of this lecture is to distill the basics of image reconstruction to PhD students and postdocs so that they can tackle their own imaging tasks.
We will introduce some computational tools to perform image reconstruction and demonstrate their application through examples in different modalities, including CT, visible light, and radio wave imaging. Time permitting, we will also give a glimpse of more advanced topics such as large-scale imaging.Learning
Objectives
- Understand the main categories of image reconstruction methods.
- Learn how to formulate an imaging task as a reconstruction problem and solve it effectively using computational tools.
Pre-requisites
- This is a beginner-level lecture.
- Familiarity with Python/NumPy.
- Basic knowledge of linear algebra.
About the Imaging Lunches: Once per month, the EPFL Center for Imaging organises an event dedicated to all PhD students and postdocs working with/in imaging. Discuss the latest advances in imaging. Connect with imaging peers. Learn about popular imaging tools!
