Published November 25, 2019
| Version v1
Dataset
Open
Image data used for publication "Species-level image classification with convolutional neural network enable insect identification from habitus images "
Creators
- 1. Natural History Museum Aarhus & Aarhus University
- 2. Aarhus University
- 3. Natural History Museum Aarhus
- 4. Natural History Museum
Description
Image-crops of specimens from insect drawers
In 2017 we scanned 208 insect drawers containing the collection of british carabids from the Natural History Museum London and extracted crops from the scanned images. This database contain 63.364 specimens that we used to train, validate and test a convolutional neural network.
Each folder is named as the gbif id number. E.g. Carabus problematicus is 4470555: https://www.gbif.org/species/4470555
Files
Files
(2.6 GB)
Name | Size | Download all |
---|---|---|
md5:6655ce1edc0612cf26f0520229284f0b
|
2.6 GB | Download |
Additional details
Related works
- Is published in
- Journal article: 10.1002/ece3.5921 (DOI)