Enhancement for Low-Contrast Images with Dynamical Saturating Nonlinearity and Adaptive Stochastic Resonance

In recent years, image processing based on stochastic resonance has received more and more attention. In this paper, a dynamic saturated nonlinear stochastic pooling network model based on adaptive stochastic resonance is proposed. And it combines the dynamical saturating nonlinearity with the stochastic resonance, which can realize adaptive iteration and enhance low-contrast images. At the same time, a new method to solve the optimal parameters of nonlinear systems is proposed, which is used to solve the optimal parameter values of dynamical saturating nonlinearity in this paper. This new image enhancement method is tested on low-contrast images in the LOL dataset. By comparing the new model and some other image enhancement algorithms, it demonstrates that the results not only have good visual perception, but also obtain more excellent evaluation indicators value.