BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Noise estimation by PDE-constrained optimisation - Schönlieb Caro
 la-Bibiane\, Cambridge University
DTSTART:20130716T130000Z
DTEND:20130716T140000Z
UID:TALK45415@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:A key issue in image denoising is an adequate choice of the co
 rrect noise model. In a variational approach this amounts to the choice of
  the data fidelity and its weighting. Depending on this choice\, different
  results are obtained. \nIn this talk I will discuss a PDE-constrained opt
 imization approach for the determination of the noise distribution in tota
 l variation (TV) image denoising. An optimization problem for the determin
 ation of the weights correspondent to different types of noise distributio
 ns is stated and existence of an optimal solution is proved. A tailored re
 gularization approach for the approximation of the optimal parameter value
 s is proposed thereafter and its consistency studied. Additionally\, the d
 ifferentiability of the solution operator is proved and an optimality syst
 em characterizing the optimal solutions of each regularized problem is der
 ived. The optimal parameter values are numerically computed by using a qua
 si-Newton method\, together with semismooth Newton type algorithms for the
  solution of the TV-subproblems. The talk is furnished with numerical exam
 ples computed on simulated data.\n\nThis is joint work with Juan Carlos De
  Los Reyes\n
LOCATION:Microsoft Research Ltd\, 21 Station Road\, Cambridge\, CB1 2FB
END:VEVENT
END:VCALENDAR