Cellular neural networks, image processing.

Cellular neural networks (CNNs) are similar to artificial neural networks (ANNs) in that they are composed of many distributed processing elements, called ?cells?, which are connected in a network. CNNs were designed to operate in a two-dimensional grid, where each processing element (cell) is connected to neighboring cells in the grid. CNNs have been shown to be an adept at image processing tasks. Specifically, CNN cells maintain a state which evolves through time due to differential equations dependent on the cell's inputs and feedback. This article introduces software for graphics processing units (GPUs) by abstracting the hardware as arrays of configurable CNNs cells. Introducing an efficient CNN-based abstraction of GPU computation should encourage the study of massively parallel computation using tools and methods of dynamic systems theory and abstract automata theory.