EU Moths Dataset

This webpage contains datasets and supplementary information for the following papers:

Deep Learning Pipeline for Automated Visual Moth Monitoring: Insect Localization and Species Classification. Dimitri Korsch, Paul Bodesheim, and Joachim Denzler. INFORMATIK 2021, Computer Science for Biodiversity Workshop (CS4Biodiversity). Pages 443-460. 2021.

Exploiting Web Images for Moth Species Classification. Julia Böhlke, Dimitri Korsch, Paul Bodesheim, and Joachim Denzler. Computer Science for Biodiversity Workshop (CS4Biodiversity), INFORMATIK 2021. Pages 481-498. 2021.

Lightweight Filtering of Noisy Web Data: Augmenting Fine-grained Datasets with Selected Internet Images. Julia Böhlke, Dimitri Korsch, Paul Bodesheim, and Joachim Denzler. International Conference on Computer Vision Theory and Applications (VISAPP). 2021.

Description

This dataset consists of 200 moth species common in Central Europe. Each of the species is represented by roughly 11 images. In total, there are 2205 images.

Class labels and validation splits

For the experiments in our paper “Moth Detection and Classification”, we split the dataset in four splits (552 images in the first split and 551 in the other three). We used the first split for the evaluation and the other three for training of the classifier.

Bounding box annotations

Additionally to the class labels, we manually annotated* bounding boxes around the single insects.

Ownership

This dataset was acquired by the Zoologisches Forschungsmuseum Alexander König, and in the context of the AMMOD project, we have the permission to provide this dataset publicly. The photographer is Dr. Josef Bücker.

Downloads

There are two versions of the dataset:

The download list for the images can be found here:

To avoid automatic downloads and unreasonable load on our server, you need to enter a password (“eu_moths_download“)

* Special thanks to Christoph Theiß