A design technique for variable two-dimensional recursive digital filters

Abstract

In some applications the frequency characteristics of a filter may be required to change during the course of signal processing. This requirement can be satisfied by filters with coefficients that are directly computable from the specified spectral parameters. Such filters are referred to as “variable filters”. Most known approaches are based on a frequency transformation, which has certain limitations especially in 2-D digital filters. In this paper, an alternative technique for 2-D variable digital filters is proposed. The filter coefficients are expressed as analytic functions with their independent variables being the frequency specifications of interest. This is achieved by designing a set of constant-coefficients filters that correspond to reasonably spaced points in the frequency-specifications space. This is followed by using a curve fitting algorithm to fit the analytic functions to the coefficient values. This approach provides the designer with an extremely short design time and hence permits on-line variation of the coefficients to meet the changing specifications. Several examples are presented to illustrate the method.