Abstract
Moonlit night scenes have a tinge of blue (blue-shift). Earlier work to model this perceptual effect has been statistical in nature; often based on unreliable measurements of blueness in paintings of moonlit night scenes. Needless to say there is a need of a more reliable and accurate model. We present a model based on the physiological functioning of the retina and how the rod and cone cells in the retina interact in moonlight conditions to generate the perception of blue. Blue-shift in purely moonlit scenes is a perceptual illusion. The light itself is not blue; moonlight is merely sunlight reflected off the grayish surface of the moon. This perception of blue is even more surprising when we consider that the light intensity at this level is below cone cell detection threshold (cells in retina responsible for color perception). Recent findings in the anatomy of the retina suggest rod cells (sensitive photoreceptors responsible for monochromatic vision) synapse onto cone circuitry under moonlight conditions (Soucy et al. Neuron, 1998, 21:481–493) (Tsukamoto et al. The Journal of Neuroscience, November 1, 2001, 21(21):8616–8623). Thus some of the rod signal is processed by the visual cortex as though it was coming from the cones. We hypothesize that the rod cells predominantly synapse onto the S-cone (cone cells sensitive to bluish light) circuitry resulting in the visual cortex perceiving a tinge of blue. To mathematically model this behavior we use R.W.G Hunt's photoreceptor response and adaptation functions (R.W.G.Hunt, Measuring Color, Fountain Press 1998). Our model produces perceptually accurate blue-shift in images of moonlit night scenes, reflecting how we would perceive the scene if we are physically present in it. Results of the model are compared with artistic renditions of night scenes by various photographers.