BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ZContent.net//ZapCalLib 1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTART;TZID=Atlantic/Canary:20131120T123000
DTEND;TZID=Atlantic/Canary:20131120T133000
UID:iactalks-553
X-WR-CALNAME: IAC Talks: Open Astronomy Seminars
X-ORIGINAL-URL: /iactalks/Talks/view/553
CREATED:2013-11-20T12:30:00+00:00
X-WR-CALDESC: IAC Talks upcomming talks
SUMMARY:Wavefront Reconstruction from Noisy Observations via Sparse Coding
DESCRIPTION:Wavefront Reconstruction from Noisy Observations via Sparse Cod
 ing\nProf. Vladimir Katkovnik\n\nMany images (and signals) admit sparse re
 presentations in the sense that  they are well approximated by linear comb
 inations of a small number of  functions taken from know sets. The topic o
 f sparse and redundant  representations, often termed as a sparse regressi
 on or sparse coding,  has attracted tremendous interest from the research 
 community in the  last ten years. This interest stems from the role that t
 he low  dimensional models play in many signal and image areas such as  co
 mpression, restoration, classification, and design of priors and  regulari
 zers, just to name a few.  In this talk we use the sparse approximations f
 or phase and magnitude of  a complex-valued wavefield. While our technique
 s are quite general here  they are illustrated for processing phase-shifti
 ng interferometry  measurements. It is assumed that the observations are P
 oissonian (photon  counting). In this way we are targeting at optimal spar
 se  reconstruction of both phase and magnitude taking into consideration a
 ll  details of the observation formation.  Contrary to the standard variat
 ional approaches we propose a vector  optimization with two objective func
 tions leading to decoupling of  inverse and denoising operations. This rec
 onstruction is framed as a  maximum likelihood constrained nonlinear optim
 ization problem. It is  demonstrated by simulation that proposed recursive
  algorithm is  efficient, demonstrates high accuracy and better imaging pe
 rformance in  comparison with the current state-of-the-art.
END:VEVENT
END:VCALENDAR