Imaging Lunch: Deconvolution: Restore Sharper Images by Inverting Acquisition

Our in-house experts Dr Vasiliki Sergiopoulou and Dr Daniel Sage will give a workshop on Deconvolution: Restore Sharper Images by Inverting Acquisition during our next imaging lunch. Open to all EPFL PhD students and Postdocs!

Image acquisition systems, such as lenses, face unavoidable degradations that produce blurred, noisy, out-of-focus images or aberrations. To mitigate this, deconvolution algorithms can correct these degradations when the image formation model can be mathematically expressed as a convolution. Deconvolution computationally restores image quality and image detail by making use of the point spread function (PSF), the impulse responses of the imaging system. It is applied to many scientific images, including light microscopy ones, yielding remarkable results with 3D data.

In this workshop, we will first study the PSF and then provide an overview of various deconvolution strategies, including inverse filtering, iterative optimization with image-prior regularization, and deep learning approaches in both supervised and self-supervised learning. Additionally, participants will have the opportunity to practice classic deconvolution algorithms, model and/or data based, using open-source software.

Audience: Scientists and engineers with an interest in imaging. No programming is required.

Registration required

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!