Taxon and trait recognition from digitized herbarium specimens using deep convolutional neural networks

Handle URI:
http://hdl.handle.net/10754/627471
Title:
Taxon and trait recognition from digitized herbarium specimens using deep convolutional neural networks
Authors:
Younis, Sohaib ( 0000-0001-9171-783X ) ; Weiland, Claus ( 0000-0003-0351-6523 ) ; Hoehndorf, Robert ( 0000-0001-8149-5890 ) ; Dressler, Stefan; Hickler, Thomas ( 0000-0002-4668-7552 ) ; Seeger, Bernhard; Schmidt, Marco ( 0000-0001-6087-6117 )
Abstract:
Herbaria worldwide are housing a treasure of hundreds of millions of herbarium specimens, which are increasingly being digitized and thereby more accessible to the scientific community. At the same time, deep-learning algorithms are rapidly improving pattern recognition from images and these techniques are more and more being applied to biological objects. In this study, we are using digital images of herbarium specimens in order to identify taxa and traits of these collection objects by applying convolutional neural networks (CNN). Images of the 1000 species most frequently documented by herbarium specimens on GBIF have been downloaded and combined with morphological trait data, preprocessed and divided into training and test datasets for species and trait recognition. Good performance in both domains suggests substantial potential of this approach for supporting taxonomy and natural history collection management. Trait recognition is also promising for applications in functional ecology.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computer Science Program; Computational Bioscience Research Center (CBRC)
Citation:
Younis S, Weiland C, Hoehndorf R, Dressler S, Hickler T, et al. (2018) Taxon and trait recognition from digitized herbarium specimens using deep convolutional neural networks. Botany Letters: 1–7. Available: http://dx.doi.org/10.1080/23818107.2018.1446357.
Publisher:
Informa UK Limited
Journal:
Botany Letters
Issue Date:
13-Mar-2018
DOI:
10.1080/23818107.2018.1446357
Type:
Article
ISSN:
2381-8107; 2381-8115
Sponsors:
SY, MS and SD received funding from the DFG Project Mobilization of trait data from digital image files by deep learning approaches (grant 316452578). Parts of RH’s & CW’s work were funded by the National Bioscience Database Center (NBDC) and the Database Center for Life Science (DBCLS) Biohackathon 2017 grants. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the TITAN Xp GPU to CW used for this research.
Additional Links:
https://www.tandfonline.com/doi/full/10.1080/23818107.2018.1446357
Appears in Collections:
Articles; Computer Science Program; Computational Bioscience Research Center (CBRC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorYounis, Sohaiben
dc.contributor.authorWeiland, Clausen
dc.contributor.authorHoehndorf, Roberten
dc.contributor.authorDressler, Stefanen
dc.contributor.authorHickler, Thomasen
dc.contributor.authorSeeger, Bernharden
dc.contributor.authorSchmidt, Marcoen
dc.date.accessioned2018-04-15T07:13:36Z-
dc.date.available2018-04-15T07:13:36Z-
dc.date.issued2018-03-13en
dc.identifier.citationYounis S, Weiland C, Hoehndorf R, Dressler S, Hickler T, et al. (2018) Taxon and trait recognition from digitized herbarium specimens using deep convolutional neural networks. Botany Letters: 1–7. Available: http://dx.doi.org/10.1080/23818107.2018.1446357.en
dc.identifier.issn2381-8107en
dc.identifier.issn2381-8115en
dc.identifier.doi10.1080/23818107.2018.1446357en
dc.identifier.urihttp://hdl.handle.net/10754/627471-
dc.description.abstractHerbaria worldwide are housing a treasure of hundreds of millions of herbarium specimens, which are increasingly being digitized and thereby more accessible to the scientific community. At the same time, deep-learning algorithms are rapidly improving pattern recognition from images and these techniques are more and more being applied to biological objects. In this study, we are using digital images of herbarium specimens in order to identify taxa and traits of these collection objects by applying convolutional neural networks (CNN). Images of the 1000 species most frequently documented by herbarium specimens on GBIF have been downloaded and combined with morphological trait data, preprocessed and divided into training and test datasets for species and trait recognition. Good performance in both domains suggests substantial potential of this approach for supporting taxonomy and natural history collection management. Trait recognition is also promising for applications in functional ecology.en
dc.description.sponsorshipSY, MS and SD received funding from the DFG Project Mobilization of trait data from digital image files by deep learning approaches (grant 316452578). Parts of RH’s & CW’s work were funded by the National Bioscience Database Center (NBDC) and the Database Center for Life Science (DBCLS) Biohackathon 2017 grants. We gratefully acknowledge the support of NVIDIA Corporation with the donation of the TITAN Xp GPU to CW used for this research.en
dc.publisherInforma UK Limiteden
dc.relation.urlhttps://www.tandfonline.com/doi/full/10.1080/23818107.2018.1446357en
dc.subjectHerbarium specimensen
dc.subjectspecies recognitionen
dc.subjectconvolutional neural networksen
dc.subjectmorphological traitsen
dc.subjecttrait recognitionen
dc.subjectdigitizationen
dc.titleTaxon and trait recognition from digitized herbarium specimens using deep convolutional neural networksen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputer Science Programen
dc.contributor.departmentComputational Bioscience Research Center (CBRC)en
dc.identifier.journalBotany Lettersen
dc.contributor.institutionDepartment of Mathematics and Computer Science, University of Marburg, Marburg, Germanyen
dc.contributor.institutionData and Modelling Centre, Senckenberg Biodiversity and Climate Research Centre (SBiK-F), Frankfurt am Main, Germanyen
dc.contributor.institutionDepartment of Botany and Molecular Evolution, Senckenberg Research Institute and Natural History Museum Frankfurt, Frankfurt am Main, Germanyen
dc.contributor.institutionDepartment of Physical Geography, Goethe University, Frankfurt am Main, Germanyen
dc.contributor.institutionPalmengarten der Stadt Frankfurt am Main, Frankfurt am Main, Germanyen
kaust.authorHoehndorf, Roberten
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.