Blind iterative restoration of images with spatially-varying blur

TL;DRAbstract

Removing non-uniform blur and noise from optical images is a very difficult problem to resolve. In this paper we describe a strategy that can be used for solving such problems. We describe how to restore images blurred by an unknown spatially-varying point spread function (PSF) by using a combination of methods including sectioning and phase diversity blind deconvolution. The PSFs on the individual sections are not known in advance. We treat the sections as a sequence of frames whose PSFs are correlated and approximately spatially-invariant, and apply iterative blind deconvolution schemes based on phase diversity to approximate these PSFs. A technique by Nagy and O’Leary is then used to restore the image globally. Test results on star cluster data are presented. 1.

Chat with Paper

AI Agents for this Paper