BatNet is an open-source, deep learning-based tool for rapid and accurate bat species identification from camera trap images, as described in Krivek et al., 2023. The baseline model is trained to identify the following 13 bat species or species complexes: Barbastella barbastellus, Eptesicus serotinus, Myotis bechsteinii, Myotis dasycneme, Myotis daubentonii, Myotis emarginatus, Myotis myotis/M. blythii, Myotis brandtii/M. alcathoe/M. mystacinus, Myotis nattereri, Nyctalus noctula, Pipistrellus pipistrellus/P. pygmaeus, Plecotus auritus/P. austriacus, Rhinolophus hipposideros/R. ferrumequinum.
Download: BatNet
Download: Test images
Download: Manual species identification guide
View: BatNet User's manual
Citation: Krivek G., Gillert A., Harder M., Fritze M., Frankowski K., Timm L., Meyer-Olbersleben L., Freiherr von Lukas U., Kerth G., van Schaik J. (2023) BatNet: a deep learning-based tool for automated bat species identification from camera trap images. Remote Sensing in Ecology and Conservation. https://zslpublications.onlinelibrary.wiley.com/doi/full/10.1002/rse2.339
Corresponding author: Gabriella Krivek, [email protected]
LICENSE for BatNet: CC BY-NC-SA 4.0
Data and R scripts used for creating figures and tables in publication are available here, under a CC BY-NC-ND 4.0 license: BatNet scripts
The R Markdown files should be opened via the R project file ("BatNet_scripts.R") in RStudio for setting up the correct working directory.