| Issue |
BIO Web Conf.
Volume 156, 2025
The 6th International Conference on Fisheries, Aquatic, and Environmental Sciences (ICFAES 2024)
|
|
|---|---|---|
| Article Number | 01004 | |
| Number of page(s) | 14 | |
| Section | Aquatic (Oceanography, Marine Science, Protecting Marine Life, Endangered Species Conservation, Water Protection, Flood Risk Management, Urban Adaptation on Global Climate Change) | |
| DOI | https://doi.org/10.1051/bioconf/202515601004 | |
| Published online | 30 January 2025 | |
The use of drones and Artificial Intelligence for dugong sighting detection in a limited resource scenario
1 YAPEKA, Kota Bogor 16153, Jawa Barat, Indonesia
2 Center for Transdisciplinary and Sustainability Science (CTSS), IPB University, Indonesia
3 International Union for Conservation of Nature and Natural Resource (IUCN) SSC Sirenia Specialist Group
* Corresponding author: akbar@yapeka.or.id
The use of commercially available drones and artificial intelligence (AI) has grown in popularity in the last decade. Nonetheless, its usage to detect cryptic and high-mobility marine mammals remains constrained by resource-intensive nature, vast coverage areas, hardware limitations, and environmental variables. This study aims to recount our experience conducting a combination of drone and AI-assisted detection (WISDAM) in a scenario with limited resources to detect dugongs. The operation was conducted in September 2023, April 2024, and May 2024 in North Sulawesi, Indonesia. Prior to flight path design, CMS questionnaires and satellite data were utilized in order to comprehend the spatial and temporal context of the dugongs and their preferred habitat. A DJI Air 2s drone was used in 28 flights, covering 12.09 km2, yielding 8,509 photos. In total, 47 photos comprise dugongs, including seven with multiple individuals. 56 sightings were successfully identified manually by multiple analysts to minimize bias, and seven photos (12.5%) were considered dubious. AI detection is rather limited compared with manual detection’s numbers of positively identified dugong photos. Out of 153 AI detections, only 27 (17.6%) were True Positives. Therefore, more flights are needed to enhance the sample size for machine learning.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.