Por favor, use este identificador para citar o enlazar a este item:
http://hdl.handle.net/10261/295989
COMPARTIR / EXPORTAR:
SHARE CORE BASE | |
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL | DATACITE | |
Título: | Image-based taxonomic classification of bulk insect biodiversity samples using deep learning and domain adaptation |
Autor: | Fujisawa, Tomochika; Noguerales, Víctor CSIC ORCID; Meramveliotakis, Emmanouil; Papadopoulou, Anna CSIC ORCID; Vogler, Alfried P. | Palabras clave: | Biodiversity assessments bulk sample coleoptera Convolutional neural networks domain adaptation image classification Machine learning |
Fecha de publicación: | 4-ene-2023 | Editor: | Royal Entomological Society (Great Britain) | Citación: | Systematic Entomology: 1-15 (2023) | Resumen: | Complex bulk samples of insects from biodiversity surveys present a challenge for taxonomic identification, which could be overcome by high-throughput imaging combined with machine learning for rapid classification of specimens. These procedures require that taxonomic labels from an existing source data set are used formodel training and prediction of an unknown target sample. However, such transfer learningmay be problematic for the study of newsamples not previously encountered in an image set, for example, from unexplored ecosystems, and require methods of domain adaptation that reduce the differences in the feature distribution of the source and target domains (training and test sets).We assessed the efficiency of domain adaptation for family-level classification of bulk samples of Coleoptera, as a critical first step in the characterization of biodiversity samples. Neural networkmodels trained with images from a global database of Coleoptera were applied to a biodiversity sample from understudied forests in Cyprus as the target. Within-dataset classification accuracy reached 98% and depended on the number and quality of training images, and on dataset complexity. The accuracy of between-datasets predictions (across disparate source–target pairs that do not share any species or genera) was at most 82% and depended greatly on the standardization of the imaging procedure. An algorithm for domain adaptation, domain adversarial training of neural networks (DANN), significantly improved the prediction performance of models trained by non-standardized, low-quality images. Our findings demonstrate that existing databases can be used to train models and successfully classify images from unexplored biota, but the imaging conditions and classification algorithms need careful consideration. | Versión del editor: | https://doi.org/10.1111/syen.12583 | URI: | http://hdl.handle.net/10261/295989 | DOI: | 10.1111/syen.12583 | ISSN: | 0307-6970 | E-ISSN: | 1365-3113 |
Aparece en las colecciones: | (IPNA) Artículos |
Ficheros en este ítem:
Fichero | Descripción | Tamaño | Formato | |
---|---|---|---|---|
Image_based-Fyjisawa_et_al-2023-systematic_Entomology.pdf | Artículo principal | 2,04 MB | Adobe PDF | Visualizar/Abrir |
CORE Recommender
SCOPUSTM
Citations
2
checked on 22-abr-2024
WEB OF SCIENCETM
Citations
1
checked on 27-feb-2024
Page view(s)
43
checked on 26-abr-2024
Download(s)
48
checked on 26-abr-2024
Google ScholarTM
Check
Altmetric
Altmetric
Este item está licenciado bajo una Licencia Creative Commons