Deep learning for image restoration
Microscopists must reach a compromise between spatial resolution, image depth, and temporal resolution. A balance of these compromises is needed to get a good enough image to complete quantitative work and gain new scientific insight. However, it is a dream of microscopists to stretch the rules, so fewer compromises are needed. Image deconvolution based on the system’s point spread function has been employed to this end. Recent advances in deep learning and machine learning are improving the options for image restoration and resolution enhancement.
In this webinar, Dr. Luciano Lucas, Director Leica Microsystems, demonstrates 3DRCAN (3D residual channel attention networks) for restoring and enhancing volumetric multi-channel time-lapse (4D) fluorescence microscopy data.
Watch this webinar to learn how to extract higher quality insights from light microscopy images, courtesy of Leica Microsystems.