dc.contributor.author | Mo, Jeff | en_NZ |
dc.contributor.author | Frank, Eibe | en_NZ |
dc.contributor.author | Vetrova, Varvara | en_NZ |
dc.contributor.editor | Peng, Wei | en_NZ |
dc.contributor.editor | Alahakoon, Damminda | en_NZ |
dc.contributor.editor | Li, Xiaodong | en_NZ |
dc.coverage.spatial | Melbourne, Australia | en_NZ |
dc.date.accessioned | 2017-08-09T21:15:29Z | |
dc.date.available | 2017 | en_NZ |
dc.date.available | 2017-08-09T21:15:29Z | |
dc.date.issued | 2017 | en_NZ |
dc.identifier.citation | Mo, J., Frank, E., & Vetrova, V. (2017). Large-scale automatic species identification. In W. Peng, D. Alahakoon, & X. Li (Eds.), Proceedings of 30th Australasian Joint Conference on Advances in Artificial Intelligence (Vol. LNCS 10400, pp. 301–312). Springer, Cham: Springer. https://doi.org/10.1007/978-3-319-63004-5_24 | en |
dc.identifier.isbn | 978-3-319-63004-5 | en_NZ |
dc.identifier.isbn | 978-3-319-63003-8 | |
dc.identifier.uri | https://hdl.handle.net/10289/11266 | |
dc.description.abstract | The crowd-sourced Naturewatch GBIF dataset is used to obtain a species classification dataset containing approximately 1.2 million photos of nearly 20 thousand different species of biological organisms observed in their natural habitat. We present a general hierarchical species identification system based on deep convolutional neural networks trained on the NatureWatch dataset. The dataset contains images taken under a wide variety of conditions and is heavily imbalanced, with most species associated with only few images. We apply multi-view classification as a way to lend more influence to high frequency details, hierarchical fine-tuning to help with class imbalance and provide regularisation, and automatic specificity control for optimising classification depth. Our system achieves 55.8% accuracy when identifying individual species and around 90% accuracy at an average taxonomy depth of 5.1—equivalent to the taxonomic rank of “family”—when applying automatic specificity control. | |
dc.format.mimetype | application/pdf | |
dc.language.iso | en | |
dc.publisher | Springer | en_NZ |
dc.rights | © Springer International Publishing AG 2017.This is the author's accepted version. The final publication is available at Springer via https://doi.org/10.1007/978-3-319-63004-5_24 | |
dc.source | AI 2017: Advances in Artificial Intelligence | en_NZ |
dc.subject | computer science | en_NZ |
dc.subject | species identification | en_NZ |
dc.subject | convolutional neural networks | en_NZ |
dc.subject | Machine learning | |
dc.title | Large-scale automatic species identification | en_NZ |
dc.type | Conference Contribution | |
dc.identifier.doi | 10.1007/978-3-319-63004-5_24 | en_NZ |
dc.relation.isPartOf | Proceedings of 30th Australasian Joint Conference on Advances in Artificial Intelligence | en_NZ |
pubs.begin-page | 301 | |
pubs.declined | 2017-06-13T15:06:47.926+1200 | |
pubs.elements-id | 194358 | |
pubs.end-page | 312 | |
pubs.finish-date | 2017-08-20 | en_NZ |
pubs.place-of-publication | Springer, Cham | |
pubs.publication-status | Published | en_NZ |
pubs.start-date | 2017-08-19 | en_NZ |
pubs.volume | LNCS 10400 | en_NZ |