Image sharpening based on time–frequency characteristics of human vision using FitzHugh–Nagumo equation