Friederike Barkmann

INTRODUCTION The number of image-based species records has increased considerably due to the growing involvement of citizen scientists in data collection and advancements in automated sensor technologies (Barta, 2023). Although such data offer many opportunities for a better understanding of species distributions and ecosystem dynamics in a rapidly changing world, processing and quality control are often time-consuming and expensive.

Abstract Deep learning models can accelerate the processing of image-based biodiversity data and provide educational value by giving direct feedback to citizen scientists. However, the training of such models requires large amounts of labelled data and not all species are equally suited for identification from images alone. Most butterfly and many moth species (Lepidoptera) which play an important role as biodiversity indicators are well-suited for such approaches.

1 Introduction In times of increasing anthropogenic pressure on ecosystems and a resulting decline in biodiversity (Johnson et al. 2017), reliable and comprehensive data on the state and trends of biodiversity is crucial. Such data is needed to raise awareness, inform policymakers, better understand ecological systems, and plan and evaluate management actions (Jones et al. 2011).