In this work, me develop the concept of partitioning the observation space
to build a general class of filters referred to as partition-based weighted
sum (PWS) filters. In the general framework, each observation vector is ma
pped to one of M partitions comprising the observation space, and each part
ition has an associated filtering function. Here, we focus on partitioning
the observation space utilizing vector quantization and restrict the filter
ing function within each partition to be linear. In this formulation, a wei
ghted sum of the observation samples forms the estimate, where the weights
are allowed to be unique within each partition. The partitions are selected
and weights tuned by training on a representative set of data. It is shown
that the proposed data adaptive processing allows for greater detail prese
rvation when encountering nonstationarities in the data and yields superior
results compared to several previously defined filters. Optimization of th
e PWS filters is addressed and experimental results are provided illustrati
ng the performance of PWS filters in the restoration of images corrupted by
Gaussian noise.