Gaussian Convolution Filter

Filter to introduce spatial autocorrelation while keeping original distribution.

Simple script to apply a gaussian convolution filter to a matrix (e.g. of white noise) to introduce spatial correlation while (generally) preserving the original distribution .
Based on the approach of spatial moving averages according to Oksanen and Sarjakoski 2005: Error propagation analysis of DEM-based drainage basin delineation. International Journal of Remote Sensing 26 (14), 3085-3102
DOI:10.1080/01431160500057947

Note: Distribution is not preserved _exactly_. Approach is valid for equidistant matrices of values that are the result of a gaussian process. Please read the help/inline comments!

Plz give feedback or report bugs in the comments section!