A Framework for Detecting and Tracking Elephants in Drone Videos

Elchik, Chaim Chai and Burger, André and Wich, Serge (2025) A Framework for Detecting and Tracking Elephants in Drone Videos. Drone Systems and Applications. ISSN 2564-4939 (In Press)

Text (Elephant_Paper_Final_Revisions_Implemented (1))
Elephant_Paper_Final_Revisions_Implemented_1_.pdf - Accepted Version
Available under License Creative Commons Attribution.
Download (2MB)

Abstract

The escalating global biodiversity crisis requires innovative and scalable solutions to monitor wildlife populations. Recent developments in remote sensing and deep learning offer promising avenues for improving the conservation of large mammals, including African elephants. This paper introduces a framework that utilizes drone video streams and integrates state-of-the-art object detection (YOLOv11) and tracking (BoT-SORT) methods, which are significantly enhanced by a custom post-track re-identification algorithm, to capture temporal dynamics and track individual elephants over time. The framework facilitates automated video analysis and elephant counting, generating key metrics such as individual elephant movement speed, group movement patterns, and elephant cluster statistics. By automating aspects of data processing and analyses, this approach provides valuable insights that contribute to more efficient and data-driven decision-making in wildlife research.