Cellular mixed signal pixel array for real time image processing

Contemporary computing platforms fail to deliver the computational density required for many real-time image processing tasks. On the other hand, even the simplest of living systems are able to perceive and interpret their environment effortlessly using a conglomerate of slow and inaccurate neurons in parallel. Motivated by this observation as well as the cellular neural network paradigm, this paper presents an integrated sensor processor architecture that captures the local connectivity patterns of the vertebrate retina in silicon to perform parallel programmable iconic image operations. Results are presented from simulation of this new cellular network paradigm.