File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Model-based network architecture for image reconstruction in lensless imaging

TitleModel-based network architecture for image reconstruction in lensless imaging
Authors
Issue Date2020
PublisherSPIE - International Society for Optical Engineering.
Citation
SPIE Conference: SPIE/COS Photonics Asia: Holography, Diffractive Optics, and Applications X, Online Meeting, China, 11-16 October 2020. In Sheng, Y... (et al) (eds.), Proceedings of SPIE, v. 11551, paper 115510B How to Cite?
AbstractWe introduce a multi-branch model-based architecture for image reconstruction in lensless imaging. The structure consists of two learning branches, namely a physical model-based network, and a data-driven network. It uses intermediate outputs from the former as a prior for guiding the learning of the reconstruction neural network, which mimics the mapping between the reconstructed high-resolution images and raw images. We demonstrate that the proposed architecture offers a flexible combination of model-based methods and deep networks with superior reconstruction performance than methods using only an unrolled optimization network or pure deep neural networks for image reconstruction.
Persistent Identifierhttp://hdl.handle.net/10722/304058
ISBN
ISSN
2023 SCImago Journal Rankings: 0.152
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZeng, T-
dc.contributor.authorLam, EYM-
dc.date.accessioned2021-09-23T08:54:40Z-
dc.date.available2021-09-23T08:54:40Z-
dc.date.issued2020-
dc.identifier.citationSPIE Conference: SPIE/COS Photonics Asia: Holography, Diffractive Optics, and Applications X, Online Meeting, China, 11-16 October 2020. In Sheng, Y... (et al) (eds.), Proceedings of SPIE, v. 11551, paper 115510B-
dc.identifier.isbn9781510639171-
dc.identifier.issn0277-786X-
dc.identifier.urihttp://hdl.handle.net/10722/304058-
dc.description.abstractWe introduce a multi-branch model-based architecture for image reconstruction in lensless imaging. The structure consists of two learning branches, namely a physical model-based network, and a data-driven network. It uses intermediate outputs from the former as a prior for guiding the learning of the reconstruction neural network, which mimics the mapping between the reconstructed high-resolution images and raw images. We demonstrate that the proposed architecture offers a flexible combination of model-based methods and deep networks with superior reconstruction performance than methods using only an unrolled optimization network or pure deep neural networks for image reconstruction.-
dc.languageeng-
dc.publisherSPIE - International Society for Optical Engineering.-
dc.relation.ispartofProceedings of SPIE, v. 11551: Holography, Diffractive Optics, and Applications X-
dc.rights© SPIE - International Society for Optical Engineering.-
dc.titleModel-based network architecture for image reconstruction in lensless imaging-
dc.typeConference_Paper-
dc.identifier.emailLam, EYM: elam@eee.hku.hk-
dc.identifier.authorityLam, EYM=rp00131-
dc.identifier.doi10.1117/12.2575205-
dc.identifier.scopuseid_2-s2.0-85097158880-
dc.identifier.hkuros324998-
dc.identifier.volume11551-
dc.identifier.spage115510B-
dc.identifier.epage115510B-
dc.identifier.isiWOS:000651086200005-
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats