This research aims to investigate the generation and application of synthetic images produced from 3D models to automate species identification, with a specific focus on ground beetles as a case study.

The process starts by creating 3D models of ground beetle specimens (Carabidae) housed in museum collections and the National Ecological Observation Network (NEON) biorepository at Arizona State University.

These 3D models can then be used to generate synthetic 2D images, aiding in the training of convolutional neural networks (CNNs) for the identification of ground beetle species from 2D images.

The final stage involves evaluating the effectiveness of these models in real-life imaging scenarios, such as field or museum settings, while also exploring potential factors that could hinder CNN performance, such as debris and specimen density.

Ultimately, this work seeks to bridge the gap between museum collections and improved ecosystem monitoring.