We will introduce two different variational methods for haze removal. The first method extends a perception-inspired variational framework in two ways: i) replacing the grey-world hypothesis by an estimation of the mean of the clean image, ii) adding a set of new terms to the energy functional for maximizing the inter-channel contrast. The second method extends the previous approach by including the computation of saturation maps, to avoid the overenhancement of near-by objects.
This presentation is part of Minisymposium “MS45 - Mathematical techniques for bad visibility restoration”
organized by: Javier Vazquez-Corral (Information and Communication Technologies Department, Universitat Pompeu Fabra) .