PARALLEL KALMAN FILTER BANK DESIGN FOR ADAPTIVE IMAGE RESTORATION.

Tzafestas, S; Skolarikos, M

dc.contributor.author	Tzafestas, S	en
dc.contributor.author	Skolarikos, M	en
dc.date.accessioned	2014-03-01T02:47:47Z
dc.date.available	2014-03-01T02:47:47Z
dc.date.issued	1986	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/33335
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-0022952420&partnerID=40&md5=a59e8d1bc416f0059ac00bc14a712e9c	en
dc.subject.other	SIGNAL FILTERING AND PREDICTION - Kalman Filtering	en
dc.subject.other	SYSTEMS SCIENCE AND CYBERNETICS - Adaptive Systems	en
dc.subject.other	ADAPTIVE IMAGE RESTORATION	en
dc.subject.other	PARALLEL KALMAN FILTER	en
dc.subject.other	IMAGE PROCESSING	en
dc.title	PARALLEL KALMAN FILTER BANK DESIGN FOR ADAPTIVE IMAGE RESTORATION.	en
heal.type	conferenceItem	en
heal.publicationDate	1986	en
heal.abstract	One of the basic problems in image reconstruction and restoration is to improve the visual quality of the degraded data of the image at hand. In many practical cases, the observed image is a degraded version of the ideal (original) image due to noise and blur. The problem which is solved here is that of finding an optimal estimate of the ideal image on the basis of the observed function that describes the degraded image and a specific optimality criterion. The image is modelled by a linear state space model involving space invariant additive Gaussian white noise. The adaptive image restoration is performed using a parallel bank of filters (partitioning approach) for estimating the state of the image model when the covariance function of the observed image has a separable exponential form depending on an unknown parameter 'a'. The method has so far been tested with simulated images.	en
heal.publisher	Plenum Press, USA & London, New York, NY, Engl	en
heal.journalName	[No source information available]	en
dc.identifier.spage	217	en
dc.identifier.epage	228	en