Show simple item record

Associative Learning of Standard Regularizing Operators in Early Vision

dc.date.accessioned2008-04-22T12:18:46Z
dc.date.accessioned2018-11-24T10:30:59Z
dc.date.available2008-04-22T12:18:46Z
dc.date.available2018-11-24T10:30:59Z
dc.date.issued1984-12
dc.identifier.urihttp://hdl.handle.net/1721.1/41218
dc.identifier.urihttp://repository.aust.edu.ng/xmlui/handle/1721.1/41218
dc.description.abstractStandard regularization methods can be used to solve satisfactorily several problems in early vision, including edge detection, surface reconstruction, the computation of motion and the recovery of color. In this paper, we suggest (a) that quadratic variational principles corresponding to standard regularization methods are equivalent to a linear regularizing operator acting on the data and (b) that this operator can be synthesized through associative learning. The synthesis of the regularizing operator involves the computation of the pseudoinverse of the data. The pseudoinverse can be computed by iterative methods, that can be implemented in analog networks. Possible implications for biological visual systems are also discussed.en
dc.language.isoen_USen
dc.publisherMIT Artificial Intelligence Laboratoryen
dc.titleAssociative Learning of Standard Regularizing Operators in Early Visionen
dc.typeWorking Paperen


Files in this item

FilesSizeFormatView
AI_WP_264.pdf341.9Kbapplication/pdfView/Open

This item appears in the following Collection(s)

Show simple item record