Color image perception based on stochastic spiking neural network

Our aim is to present an interpretable algorithm for enhancing low-illuminance color image based on the principle of stochastic resonance and fundamental biophysical process human brain perceiving object color. To this end, phenomenon in a conductance-based integrate-and-fire neuronal network first explored, with effect firing threshold, synaptic weight population size signal-to-noise ratio revealed, threshold recognized as key parameter effects. And then, enhancement algorithm, where peak natural quality evaluator are adopted quantifying indexes, developed by combining spiking involved relating visual perception. Note that enhanced aperiodic, thus order optimize performance illuminance distribution strategy given us time. The numerical tests show has good stability. We wish could be applied relevant signal processing fields such military detection medical preprocessing.