Vis enkel innførsel

dc.contributor.authorChoi, Changkyu
dc.contributor.authorKampffmeyer, Michael
dc.contributor.authorHandegard, Nils Olav
dc.contributor.authorSalberg, Arnt Børre
dc.contributor.authorBrautaset, Olav
dc.contributor.authorEikvil, Line
dc.contributor.authorJenssen, Robert
dc.date.accessioned2021-09-07T09:01:26Z
dc.date.available2021-09-07T09:01:26Z
dc.date.created2021-08-19T15:36:00Z
dc.date.issued2021
dc.identifier.citationICES Journal of Marine Science. 2021, .en_US
dc.identifier.issn1054-3139
dc.identifier.urihttps://hdl.handle.net/11250/2773937
dc.description.abstractAcoustic target classification in multi-frequency echosounder data is a major interest for the marine ecosystem and fishery management since it can potentially estimate the abundance or biomass of the species. A key problem of current methods is the heavy dependence on the manual categorization of data samples. As a solution, we propose a novel semi-supervised deep learning method leveraging a few annotated data samples together with vast amounts of unannotated data samples, all in a single model. Specifically, two inter-connected objectives, namely, a clustering objective and a classification objective, optimize one shared convolutional neural network in an alternating manner. The clustering objective exploits the underlying structure of all data, both annotated and unannotated; the classification objective enforces a certain consistency to given classes using the few annotated data samples. We evaluate our classification method using echosounder data from the sandeel case study in the North Sea. In the semi-supervised setting with only a tenth of the training data annotated, our method achieves 67.6% accuracy, outperforming a conventional semi-supervised method by 7.0 percentage points. When applying the proposed method in a fully supervised setup, we achieve 74.7% accuracy, surpassing the standard supervised deep learning method by 4.7 percentage points.
dc.language.isoengen_US
dc.relation.urihttps://academic.oup.com/icesjms/advance-article/doi/10.1093/icesjms/fsab140/6348794
dc.rightsNavngivelse-Ikkekommersiell-DelPåSammeVilkår 4.0 Internasjonal*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/4.0/deed.no*
dc.titleSemi-supervised target classification in multi-frequency echosounder dataen_US
dc.typeJournal articleen_US
dc.typePeer revieweden_US
dc.description.versionpublishedVersion
cristin.ispublishedtrue
cristin.fulltextoriginal
cristin.qualitycode1
dc.identifier.doi10.1093/icesjms/fsab140
dc.identifier.cristin1927395
dc.source.journalICES Journal of Marine Scienceen_US
dc.source.pagenumber13en_US
dc.relation.projectNorges forskningsråd: 309512
dc.relation.projectNorges forskningsråd: 270966


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel

Navngivelse-Ikkekommersiell-DelPåSammeVilkår 4.0 Internasjonal
Med mindre annet er angitt, så er denne innførselen lisensiert som Navngivelse-Ikkekommersiell-DelPåSammeVilkår 4.0 Internasjonal