MultiTempLSTM: prediction and compression of multitemporal hyperspectral images using LSTM networks

Basit Öğe Kaydı
dc.authoridKaraca, Ali Can / 0000-0002-6835-7634
dc.authoridGULLU, MEHMET KEMAL / 0000-0003-2310-2985
dc.authorwosidKaraca, Ali Can/B-6629-2016
dc.authorwosidGULLU, MEHMET KEMAL/F-7390-2018
dc.contributor.authorKaraca, Ali Can
dc.contributor.authorGüllü, Mehmet Kemal
dc.date.accessioned2022-02-15T16:57:57Z
dc.date.available2022-02-15T16:57:57Z
dc.date.issued2021
dc.departmentBakırçay Üniversitesien_US
dc.description.abstractSince multitemporal hyperspectral imaging has an excellent ability to observe the Earth's surface over time, it has been used for various remote sensing applications. On the other hand, multitemporal hyperspectral images (HSIs) contain HSI sequences acquired multiple times over the same scene, resulting in large amounts of data. Conventional HSI compression methods cannot benefit from temporal correlation, which can be very high, depending on the acquisition cycle. We propose a prediction and compression framework that directly considers temporal correlation for the compression of HSIs. The main objective of the proposed method is to predict each spectral signature in the target HSI from the corresponding spectral signature of the reference HSI using a long short-term memory network model that supports clustering. Then, the residual image between the predicted HSI and the target HSI is quantized and entropy encoded for the compression purpose. The experiments are conducted on a ground-based multitemporal dataset named Noguiero, which contains nine HSIs, in terms of prediction and compression performances. Experiments show that the proposed method not only provides the best quality metrics from the perspective of prediction but also has convincing compression performances compared to the other methods. (C) 2021 Society of Photo-Optical Instrumentation Engineers (SPIE)en_US
dc.description.sponsorshipScientific and Technological Research Council of Turkey (TUBITAK)Turkiye Bilimsel ve Teknolojik Arastirma Kurumu (TUBITAK) [119E405]en_US
dc.description.sponsorshipThis study has been supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under Project No. 119E405.en_US
dc.identifier.doi10.1117/1.JRS.15.042409
dc.identifier.issn1931-3195
dc.identifier.issue4en_US
dc.identifier.scopus2-s2.0-85122693084en_US
dc.identifier.scopusqualityQ2en_US
dc.identifier.urihttps://doi.org/10.1117/1.JRS.15.042409
dc.identifier.urihttps://hdl.handle.net/20.500.14034/313
dc.identifier.volume15en_US
dc.identifier.wosWOS:000693616700001en_US
dc.identifier.wosqualityQ4en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.language.isoenen_US
dc.publisherSpie-Soc Photo-Optical Instrumentation Engineersen_US
dc.relation.journalJournal Of Applied Remote Sensingen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectmultitemporal imagesen_US
dc.subjecthyperspectral image compressionen_US
dc.subjectremote sensingen_US
dc.subjectlong short-term memory networksen_US
dc.subjectEfficient Lossless Compressionen_US
dc.subjectNeural-Networksen_US
dc.titleMultiTempLSTM: prediction and compression of multitemporal hyperspectral images using LSTM networksen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Elektrik-Elektronik Mühendisliği Bölümü Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu