Automated Detection of Salvin's Albatrosses: Improving Deep Learning Tools for Aerial Wildlife Surveys

TL;DR

This study evaluates the effectiveness of a general-purpose bird detection model for monitoring Salvin's albatross populations using drone imagery, demonstrating that fine-tuning and enhanced techniques significantly improve detection accuracy in remote environments.

Contribution

It introduces an assessment of a pre-trained avian detection model for species-specific monitoring and demonstrates the benefits of fine-tuning and augmentation techniques.

Findings

Zero-shot detection provides a strong baseline.

Fine-tuning improves detection accuracy.

Enhanced augmentation techniques lead to better results.

Abstract

Recent advancements in deep learning and aerial imaging have transformed wildlife monitoring, enabling researchers to survey wildlife populations at unprecedented scales. Unmanned Aerial Vehicles (UAVs) provide a cost-effective means of capturing high-resolution imagery, particularly for monitoring densely populated seabird colonies. In this study, we assess the performance of a general-purpose avian detection model, BirdDetector, in estimating the breeding population of Salvin's albatross (Thalassarche salvini) on the Bounty Islands, New Zealand. Using drone-derived imagery, we evaluate the model's effectiveness in both zero-shot and fine-tuned settings, incorporating enhanced inference techniques and stronger augmentation methods. Our findings indicate that while applying the model in a zero-shot setting offers a strong baseline, fine-tuning with annotations from the target domain and stronger image augmentation leads to marked improvements in detection accuracy. These results highlight the potential of leveraging pre-trained deep-learning models for species-specific monitoring in remote and challenging environments.