Inversion of large-support ill-conditioned linear operators using a Markov model with a line process (1994) [14 citations — 9 self]
Download:
by Mila Nikolova, Ali Mohammad-djafari, Jdrdme Idler
in Proceedings of IEEE ICASSP
http://www.irccyn.ec-nantes.fr/~idier/./pub/Nikolova94C.pdf
Add To MetaCart
Abstract:
We propose a method for the reconstruction of an im-age, only partially observed through a linear integral operator. As such an inverse problem is ill-posed, prior information must be introduced. We consider the case of a compound Markov random field with a non-interacting line process. In order to maximise the posterior likelihood function, wc propose an extension of the Graduated Non Convexity principle pioneered by Blake & Zisserman which allows its use for ill-posed linear inverse problems. We discuss the role of the observation scale and some aspects of the implemented algorithm. Finally, we present an application of the method to a diffraction tomography imaging problem.
