An efficient least-squares approach for the design of two-dimensional linear-phase nonrecursive filters

A method is described which can be used to design two-dimensional nonrecursive linear-phase filters. The approach is based on formulating the absolute mean-square error between the amplitude responses of the practical and ideal digital filters as a quadratic function. The coefficients of the filters are obtained by solving a set of linear equations. This method leads to a lower mean-square error and is computationally more efficient than the eigenfilter method. The method is extended to the design of filters with time-domain constraints.<<ETX>>