Applied Zoology and Nature Conservation
Tel.: +49 (0)3834 420-4353
Fax: +49 (0)3834 420-4252
Automated species identification of European bats
Wildlife monitoring using camera traps enables us to collect wildlife activity data at large time and space scales and to study the impact of human-induced environmental changes on biodiversity across regions, seasons and species. Automated monitoring systems, such as camera traps and acoustic recorders are increasingly used in bat population monitoring across Germany, producing a high volume of data, which must be analyzed by human experts. Although manual processing of these digital data is time-consuming and expensive, automated species identification from images for population monitoring still remains a challenge. The main focus of my current research is to automate the identification of insectivorous bat species present in Germany from camera trap images using deep neural networks and apply this method on large-scale, long-term datasets to explore species-specific winter phenology.