Dynamical systems like neural networks based on lateral inhibition have a large field of applications in image processing. We will propose some examples of dynamical flows used in image contrasting and contouring. First we recall the physiological basis of the retina function by showing the role of the lateral inhibition in the optical illusions thanks to a serious game for the early detection of the diabetic retinitis. Then, based on these biomedical considerations about the real vision mechanisms, we present an enhancement method for contrasting medical images, using either a discrete neural network approach, or its continuous version, i.e. a non-isotropic diffusion reaction partial differential system. Eventually, we introduce the notion of mixed potential Hamiltonian flow method, we compare it with the watershed method and we use it for contouring.

This presentation is part of Minisymposium “MS53 - Dimensionality Reduction Algorithms for Large-Scale Images (2 parts)”
organized by: Salvatore Cuomo (Dept. Mathematics and Applications "Renato Caccioppoli", University of Naples) , Costantinos Siettos (National Technical University of Athens ) , Lucia Russo (Consiglio Nazionale delle Ricerche, Istituto di Ricerche sulla Combustione) .

Authors:

Jacques Demongeot (University of Grenoble Alpes (UGA))

Keywords: