Show simple item record

dc.contributor.authorAlam, Fahim
dc.contributor.authorZhou, Jun
dc.contributor.authorLiew, Wee Chung
dc.contributor.authorJo, Jun
dc.contributor.authorGao, Yongsheng
dc.date.accessioned2020-01-31T06:07:41Z
dc.date.available2020-01-31T06:07:41Z
dc.date.issued2018
dc.identifier.isbn9781728115818
dc.identifier.issn2158-6268
dc.identifier.doi10.1109/WHISPERS.2018.8747227
dc.identifier.urihttp://hdl.handle.net/10072/391054
dc.description.abstractConvolutional neural networks (CNNs) have demonstrated significant performance in various visual recognition problems in recent years. Recent research has shown that training multilayer neural networks can extensively improve the performance of hyperspectral image (HSI) classification. In this paper, we apply a triplet constraint property on a 3D CNN. This method directly learns a mapping from images to a Euclidean space in which distances directly correspond to a measure of spectral-spatial similarity. Once this embedding has been established, classification can be implemented with such embeddings as feature vectors. Moreover, we also augment the size of the training samples in different band groups. This produces different yet useful estimation of spectral-spatial characteristics of HSI data and contributes considerably in accurate classification. This method is evaluated on a new dataset and compared with several state-of-the-art models, which shows the promising potential of our method.
dc.description.peerreviewedYes
dc.languageEnglish
dc.publisherIEEE
dc.relation.ispartofconferencename9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS 2018)
dc.relation.ispartofconferencetitle2018 9TH WORKSHOP ON HYPERSPECTRAL IMAGE AND SIGNAL PROCESSING: EVOLUTION IN REMOTE SENSING (WHISPERS)
dc.relation.ispartofdatefrom2018-09-23
dc.relation.ispartofdateto2018-09-26
dc.relation.ispartoflocationAmsterdam, Netherlands
dc.relation.ispartofpagefrom1
dc.relation.ispartofpagefrom5 pages
dc.relation.ispartofpageto5
dc.relation.ispartofpageto5 pages
dc.relation.ispartofvolume2018-September
dc.subject.fieldofresearchArtificial Intelligence and Image Processing
dc.subject.fieldofresearchcode0801
dc.subject.keywordsScience & Technology
dc.subject.keywordsEngineering, Electrical & Electronic
dc.subject.keywordsRemote Sensing
dc.subject.keywordsTelecommunications
dc.titleTriplet constrained deep feature extraction for hyperspectral image classification
dc.typeConference output
dc.type.descriptionE1 - Conferences
dcterms.bibliographicCitationAlam, F; Zhou, J; Liew, WC; Jo, J; Gao, Y, Triplet constrained deep feature extraction for hyperspectral image classification, 2018 9th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), 2018, pp. 1-5
dc.date.updated2020-01-31T03:40:19Z
dc.description.versionAccepted Manuscript (AM)
gro.hasfulltextFull Text
gro.griffith.authorZhou, Jun
gro.griffith.authorAlam, Fahim
gro.griffith.authorLiew, Alan Wee-Chung
gro.griffith.authorJo, Jun
gro.griffith.authorGao, Yongsheng


Files in this item

This item appears in the following Collection(s)

  • Conference outputs
    Contains papers delivered by Griffith authors at national and international conferences.

Show simple item record