Saved in:
| Hovedforfatter: | |
|---|---|
| Format: | Recurso digital |
| Sprog: | engelsk |
| Udgivet: |
Zenodo
2025
|
| Fag: | |
| Online adgang: | https://doi.org/10.5281/zenodo.16845599 |
| Tags: |
Tilføj Tag
Ingen Tags, Vær først til at tagge denne postø!
|
Indholdsfortegnelse:
- <p>Wildlife poaching and habitat destruction pose significant threats to biodiversity, leading to the rapid decline<br>of endangered species. This paper presents a sustainable, AI-driven Internet of Things (IoT) surveillance<br>framework for real-time wildlife monitoring and poaching prevention. The proposed system integrates thermal<br>imaging, acoustic sensors, and unmanned aerial vehicles (UAVs) with convolutional neural networks (CNN)<br>for automated detection and classification of potential threats. Data is transmitted via low-power wide-area<br>networks (LPWAN) to cloud servers for centralized analytics, alert dissemination, and decision-making. The<br>approach focuses on energy efficiency, cost-effectiveness, and scalability to remote forest reserves. Simulation<br>and prototype evaluations demonstrate a detection accuracy of over 94% for human intrusions and an average<br>latency of under 2 seconds for alert transmission, making it a viable solution for large-scale wildlife protection.</p>