Systolic Algorithms for Digital Image Filtering

Abstract Many low-level image processing operations, termed local operators, require access to the four or eight neighbouring intensity values of a pixel, when computing the new value for the pixel and need large amounts of computing, i.e. banded matrix operations. However, these algorithms contain explicit parallelism which can be efficiently exploited by processor arrays. The purpose of this paper is to identify a set of systolic array designs suitable for implementing low-level image processing algorithms on VLSI processing arrays. In particular we consider the sigma, inverse gradient and mean filters. To achieve high performance we have developed several models of systolic arrays.