Multi-Species Object Detection in Drone Imagery for Population Monitoring of Endangered Animals
This work addresses the need for efficient population monitoring of endangered animals, though it is incremental as it builds on existing YOLOv8 methods.
The researchers tackled the problem of accurately counting endangered animal species from drone imagery by fine-tuning object detection models, achieving 95% accuracy on a safari animal dataset compared to a baseline of 0.7%.
Animal populations worldwide are rapidly declining, and a technology that can accurately count endangered species could be vital for monitoring population changes over several years. This research focused on fine-tuning object detection models for drone images to create accurate counts of animal species. Hundreds of images taken using a drone and large, openly available drone-image datasets were used to fine-tune machine learning models with the baseline YOLOv8 architecture. We trained 30 different models, with the largest having 43.7 million parameters and 365 layers, and used hyperparameter tuning and data augmentation techniques to improve accuracy. While the state-of-the-art YOLOv8 baseline had only 0.7% accuracy on a dataset of safari animals, our models had 95% accuracy on the same dataset. Finally, we deployed the models on the Jetson Orin Nano for demonstration of low-power real-time species detection for easy inference on drones.