Multi FPGA processor for gray-scale morphology

Local operations in image processing are often used, namely during the preprocessing step. On one hand, their implementation is expensive, on the other hand, they become efficient when they use a wide area neighborhood. In this paper we propose a general method for synthesis of linear and non linear filters (mathematical morphology operators). The main interest of this method is that it can be applied to various types of filters: separable or not, factorizable or not and for any size of the filter kernel (convolution) or the structuring element (mathematical morphology). From this method, one can get an architecture in a straightforward way. We propose a re-configurable architecture, based upon reprogrammable circuits (FPGA: Field Programmable Gate Array). This multi FPGA architecture allows a real time implementation of any r × c size kernel (processing at video rate).