Jl. Marroquin et al., ROBUST QUADRATURE FILTERS, Journal of the Optical Society of America. A, Optics, image science,and vision., 14(4), 1997, pp. 779-791
We deal with the relation between two well-known topics in signal proc
essing and computational vision: quadrature filters (QF's) and Bayesia
n estimation with Markov random fields (MRF's) as prior models. We pre
sent a new class of complex-valued MRF models such that the optimal es
timators obtained with them correspond to the output of QF's tuned at
particular frequencies. It is shown that the machinery that has proven
to be effective in classical (real-valued) MRF modeling may be genera
lized to the complex case in a straightforward way. To illustrate the
power of this technique, we present complex MRF models that implement
robust QF's that exhibit good performance in situations in which ordin
ary linear, shift-invariant filters fail. These include robust filters
that are relatively insensitive to edge Effects and missing data and
that can reliably estimate the local phase in singularity neighborhood
s; we also present models for the specification of piecewise-smooth QF
's. Examples of applications to fringe pattern analysis, phase-based s
tereo reconstruction, and texture segmentation are presented as well.
(C) 1997 Optical Society of America.