Retina simulator challenges, image processing with a varying resolution sensor

Abstract:
The fields of image processing and neurobiology have an interesting convergence. Neurobiological findings motivate new algorithms while the methods applied to solve diverse image processing tasks can provide the rationale for the functional organization of neurobiological circuits. Visual capabilities of a human still outperform supercomputers, while the resolution of digital image sensors and the processing capacity already reached that of the human fovea and the retina respectively. The primate retina has spatially varying resolution, and the recorded visual stimulus is processed multiple output channels with specific functions. This paper introduces a retina simulator in which some of these channels is analyzed. The system is aimed to investigate the effects of bifocal intraocular lenses on retina output.