Introduction to machine and deep-learning for image analysis

Abstract :

Image analysis has significantly evolved in the past decade with the rise of machine and deep learning approaches. More robust and more accurate than traditional approaches, they offer potential for significant improvement in automated phenotyping. However, their implementation often requires a certain knowledge of the underlying concepts, some programming skills and computational resources, which might discourage newcomers. In this presentation we aim at 1) demystify the theoretical background 2) give some insights on how to drive a project  involving machine learning or deep learning for automated phenotyping 3) present some user-friendly tools.