Stack filters are a class of nonlinear filters with excellent properties fo
r signal restoration. Unfortunately, present algorithms for designing stack
filters can only be used for small window sizes because of either their co
mputational overhead or their serial nature.
This paper presents a new adaptive algorithm for determining a stack filter
that minimizes the mean absolute error criterion. The new algorithm retain
s the iterative nature of many current adaptive stack filtering algorithms,
but significantly reduces the number of iterations required to converge to
an optimal filter. This algorithm is faster than all currently available s
tack filter design algorithms, is simple to implement, and is shown in this
paper to always converge to an optimal stack filter.
Extensive comparisons between this new algorithm and all existing algorithm
s are provided. The comparisons are based both on the performance of the re
sulting filters and upon the time and space complexity of the algorithms, T
hey demonstrate that the new algorithm has three advantages: it is faster t
han all other available algorithms; it can be used on standard workstations
(SPARC 5 with 48MB) to design filters with windows containing 20 or more p
oints; and, its highly parallel structure allows very fast implementations
on parallel machines. This new algorithm allows cascades of stack filters t
o be designed; stack filters with windows containing 72 points have been de
signed in a matter of minutes under this new approach.