Species‐level image classification with convolutional neural network enables insect identification from habitus images

Abstract Changes in insect biomass, abundance, and diversity are challenging to track at sufficient spatial, temporal, and taxonomic resolution. Camera traps can capture habitus images of ground‐dwelling insects. However, currently sampling involves manually detecting and identifying specimens. Here...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Oskar L. P. Hansen, Jens‐Christian Svenning, Kent Olsen, Steen Dupont, Beulah H. Garner, Alexandros Iosifidis, Benjamin W. Price, Toke T. Høye
Formato: article
Lenguaje:EN
Publicado: Wiley 2020
Materias:
Acceso en línea:https://doaj.org/article/c01d4d02cbea445f9f5264d13dbac9e5
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:c01d4d02cbea445f9f5264d13dbac9e5
record_format dspace
spelling oai:doaj.org-article:c01d4d02cbea445f9f5264d13dbac9e52021-11-04T13:06:09ZSpecies‐level image classification with convolutional neural network enables insect identification from habitus images2045-775810.1002/ece3.5921https://doaj.org/article/c01d4d02cbea445f9f5264d13dbac9e52020-01-01T00:00:00Zhttps://doi.org/10.1002/ece3.5921https://doaj.org/toc/2045-7758Abstract Changes in insect biomass, abundance, and diversity are challenging to track at sufficient spatial, temporal, and taxonomic resolution. Camera traps can capture habitus images of ground‐dwelling insects. However, currently sampling involves manually detecting and identifying specimens. Here, we test whether a convolutional neural network (CNN) can classify habitus images of ground beetles to species level, and estimate how correct classification relates to body size, number of species inside genera, and species identity. We created an image database of 65,841 museum specimens comprising 361 carabid beetle species from the British Isles and fine‐tuned the parameters of a pretrained CNN from a training dataset. By summing up class confidence values within genus, tribe, and subfamily and setting a confidence threshold, we trade‐off between classification accuracy, precision, and recall and taxonomic resolution. The CNN classified 51.9% of 19,164 test images correctly to species level and 74.9% to genus level. Average classification recall on species level was 50.7%. Applying a threshold of 0.5 increased the average classification recall to 74.6% at the expense of taxonomic resolution. Higher top value from the output layer and larger sized species were more often classified correctly, as were images of species in genera with few species. Fine‐tuning enabled us to classify images with a high mean recall for the whole test dataset to species or higher taxonomic levels, however, with high variability. This indicates that some species are more difficult to identify because of properties such as their body size or the number of related species. Together, species‐level image classification of arthropods from museum collections and ecological monitoring can substantially increase the amount of occurrence data that can feasibly be collected. These tools thus provide new opportunities in understanding and predicting ecological responses to environmental change.Oskar L. P. HansenJens‐Christian SvenningKent OlsenSteen DupontBeulah H. GarnerAlexandros IosifidisBenjamin W. PriceToke T. HøyeWileyarticlearthropod samplingautomatic species identificationcamera trapentomological collectionimage classificationimage databaseEcologyQH540-549.5ENEcology and Evolution, Vol 10, Iss 2, Pp 737-747 (2020)
institution DOAJ
collection DOAJ
language EN
topic arthropod sampling
automatic species identification
camera trap
entomological collection
image classification
image database
Ecology
QH540-549.5
spellingShingle arthropod sampling
automatic species identification
camera trap
entomological collection
image classification
image database
Ecology
QH540-549.5
Oskar L. P. Hansen
Jens‐Christian Svenning
Kent Olsen
Steen Dupont
Beulah H. Garner
Alexandros Iosifidis
Benjamin W. Price
Toke T. Høye
Species‐level image classification with convolutional neural network enables insect identification from habitus images
description Abstract Changes in insect biomass, abundance, and diversity are challenging to track at sufficient spatial, temporal, and taxonomic resolution. Camera traps can capture habitus images of ground‐dwelling insects. However, currently sampling involves manually detecting and identifying specimens. Here, we test whether a convolutional neural network (CNN) can classify habitus images of ground beetles to species level, and estimate how correct classification relates to body size, number of species inside genera, and species identity. We created an image database of 65,841 museum specimens comprising 361 carabid beetle species from the British Isles and fine‐tuned the parameters of a pretrained CNN from a training dataset. By summing up class confidence values within genus, tribe, and subfamily and setting a confidence threshold, we trade‐off between classification accuracy, precision, and recall and taxonomic resolution. The CNN classified 51.9% of 19,164 test images correctly to species level and 74.9% to genus level. Average classification recall on species level was 50.7%. Applying a threshold of 0.5 increased the average classification recall to 74.6% at the expense of taxonomic resolution. Higher top value from the output layer and larger sized species were more often classified correctly, as were images of species in genera with few species. Fine‐tuning enabled us to classify images with a high mean recall for the whole test dataset to species or higher taxonomic levels, however, with high variability. This indicates that some species are more difficult to identify because of properties such as their body size or the number of related species. Together, species‐level image classification of arthropods from museum collections and ecological monitoring can substantially increase the amount of occurrence data that can feasibly be collected. These tools thus provide new opportunities in understanding and predicting ecological responses to environmental change.
format article
author Oskar L. P. Hansen
Jens‐Christian Svenning
Kent Olsen
Steen Dupont
Beulah H. Garner
Alexandros Iosifidis
Benjamin W. Price
Toke T. Høye
author_facet Oskar L. P. Hansen
Jens‐Christian Svenning
Kent Olsen
Steen Dupont
Beulah H. Garner
Alexandros Iosifidis
Benjamin W. Price
Toke T. Høye
author_sort Oskar L. P. Hansen
title Species‐level image classification with convolutional neural network enables insect identification from habitus images
title_short Species‐level image classification with convolutional neural network enables insect identification from habitus images
title_full Species‐level image classification with convolutional neural network enables insect identification from habitus images
title_fullStr Species‐level image classification with convolutional neural network enables insect identification from habitus images
title_full_unstemmed Species‐level image classification with convolutional neural network enables insect identification from habitus images
title_sort species‐level image classification with convolutional neural network enables insect identification from habitus images
publisher Wiley
publishDate 2020
url https://doaj.org/article/c01d4d02cbea445f9f5264d13dbac9e5
work_keys_str_mv AT oskarlphansen specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT jenschristiansvenning specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT kentolsen specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT steendupont specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT beulahhgarner specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT alexandrosiosifidis specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT benjaminwprice specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
AT tokethøye specieslevelimageclassificationwithconvolutionalneuralnetworkenablesinsectidentificationfromhabitusimages
_version_ 1718444926219321344