Akademik Veri Yönetim Sistemi
Ever E., Kizilkaya B., Khan M. T., Yazici A., Ansari S., Amangeldi A.
Yükseköğretim Kurumları Destekli Proje, BAP Diğer, 2025 - 2027
Current forest
fire detection frameworks based on WSNs, multimedia sensors, and UAVs each offer
unique strengths, yet face important limitations when used independently.
WSN-based systems are known for their energy efficiency, scalability, and
cost-effectiveness. However, they mostly rely on scalar data (e.g.,
temperature, humidity), which limits their precision and reliability in
detecting early signs of fire [62].
Multimedia
sensor-based approaches improve detection accuracy by incorporating visual data
and leveraging advanced deep learning models. While they offer high accuracy,
they demand substantial computational power and energy. This makes them less
suitable for real-time and continuous monitoring, particularly on edge devices
with limited resources [63].
UAV-based
systems add mobility, wide-area coverage, and rapid deployment capabilities,
allowing real-time fire detection and faster response [64]. However, UAVs are
constrained by limited battery life, short flight durations, and high energy
consumption. These factors require careful mission planning and energy-aware
decision-making to ensure efficient and timely operations.
Given these
individual constraints, we aim to develop an effective forest fire detection
system that integrates the energy efficiency and scalability of WSNs, the high
accuracy of multimedia sensors, and the flexibility of UAVs. A unified approach
that combines these technologies can help achieve real-time, reliable, and
energy-aware fire detection, offering a practical and scalable solution for
real-world deployment. In order to validate such an integrated system and assess
its performance under realistic conditions, we also aim to develop a complete
testbed implementation, which will enable us to rigorously evaluate system
components, data flow, energy consumption, and communication reliability. It
will also provide a controlled environment to refine system coordination and
robustness before large-scale deployment, ultimately bridging the gap between
simulation-based research and real-world application.