Hardware Implementation of a Desktop Supercomputer for High Performance Image Processing. Time Multiplexed Color Image Processing Based on a VLSI Cellular Neural Network with Cell-State Outputs.

This report presents a Cellular Neural Network CNN VLSI implementation to be used in a time-multiplexing scheme for processing large and complex images. While efficient VLSI implementations have been reported, no research work has addressed the use of small CNN arrays for processing large complex images. Hence, the work hereby presented introduces a strategy to process large images using small CNN arrays. For practical size applications, due to hardware limitations, it is impossible to have a one-on-one mapping between the CNN hardware cells and all the pixels in the image involved. This report presents a practical solution by processing the input image block by block, with the number of pixels in a block being the same as the number of CNN cells in the array. Furthermore, unlike other implementations in which the output is taken from the feedback path of each cell, the VLSI architecture hereby described takes the outputs from the state node. While previous implementations are mostly suitable for black and white applications because of the thresholded outputs, our approach is especially suitable for applications in color gray image processing due to the analog nature of the state node.