Abstract
Blind deconvolution is an important problem that arises in many fields of research. It is of particular relevance to imaging through turbulence where the point spread function can only be modelled statistically, and direct measurement may be difficult. We describe this problem by a noisy convolution, where f(x, y) represents the true image, h(x, y) the instantaneous atmospheric blurring, g(x, y) the noise free data and n(x, y) is the noise present on the detected image. We use to denote an estimate of these quantities and our objective is to recover both f(x, y) and h(x, y) from the observed data d(x, y).
© 1998 Optical Society of America
PDF ArticleMore Like This
David S. C. Biggs and Mark Andrews
STuA.3 Signal Recovery and Synthesis (SRS) 1998
Eric Thiébaut
SWB4 Signal Recovery and Synthesis (SRS) 2001
Timothy J. Schulz
RWA1 Signal Recovery and Synthesis (SRS) 1995