ViT and CNN Mosquito Larvae Classification Research
Title: A Mosquito is Worth 16x16 Larvae: Evaluation of Deep Learning Architectures for Mosquito Larvae Classification
Researchers: Aswin Surya, David Backer Peral, Austin VanLoon, Akhila Rajesh
Data Collection: Data collected between 5/31/2017 - 7/7/2022 from Africa, North America, and Latin America
Summary: This research presents an evalutaion of two vision transformers (ViT) and two convolutional neural networks (CNN) that aim to autonomously classify mosquito larvae images as either Aedes, Culex, or neither by running inference on readily available image data from GLOBE Mosquito Habitat Mapper database. By determining the classification of potentially dangerous mosquitos, we can aid in the prevention of the global transmission of mosquito borne vector diseases.