A simplicial CNN architecture for on-chip image processing

Imagers with on-chip processing capabilities are required for high-speed and high resolution image processing systems. In this paper, we present an image processor based on a simplicial CNN (S-CNN) cell. As a result of using the S-CNN as the core processor, the architecture is endowed with a solid theoretical framework and the parallel processing capabilities of CNNs. The proposed architecture is fully digital, which is fundamental because it implies that the density of the S-CNN imager will scale directly with technology.