engleski

Brightness estimation in a neural network model with presynaptic inhibition

Recent psychophysical and neurophysiological investigations showed that visual system encodes luminance and use it to estimate illumination and surface brightness. We proposed a novel neural model for luminance coding based on recurrent inhibition, from the retinal ganglion cells to the axons of the bipolar cells, which modulates the amount of sensory input that ganglion cells receive (Sagdullaev et al., 2006). Extended version of the model, where the amount of presynaptic inhibition is made proportional to the maximum luminance in the visual scene, implements gain control mechanism which adjusts the raw luminance into a measure of brightness of surface. Computer simulations showed that the model scales brightness estimates consistent with the highest-luminance-as-white anchoring rule (Gilchist et al., 2004). Simulations also showed that the model is able to act as a change detector when the presynaptic inhibition temporally lags behind the excitatory input to the ganglion cell.

brightness; perception; neural network

nije evidentirano