thanks Rusty, that was an interesting read.
I guess that we cannot use active imaging for astro, but the method could be useful if anyone here does microscopy. The image noise reduction part seems to be based on regularisation concepts - if I read it right, we may already have related denoising built into some image processing packages (eg Pixinsight).
remarkable the ways that imaging theory is developing.
ref:
http://en.wikipedia.org/wiki/Total_variation_denoising