Published Paper Details:

Abstract: Forest fires are becoming more frequent and intense worldwide due to climate change, human activity, and shifting environmental conditions. Rising temperatures, prolonged droughts, and land-use changes have significantly increased wildfire risks, causing severe ecological and economic losses. Early detection and reliable risk mapping are essential to minimize such impacts and enable timely response. However, conventional methods like ground observation or single-sensor monitoring often face limitations due to restricted coverage, delayed detection, and interference from clouds or smoke. This study proposes an artificial intelligence (AI)-based multi-sensor image fusion framework for early forest fire detection and risk assessment. The approach integrates thermal, optical, and radar satellite data from publicly available open-access sources to enhance spatial and temporal accuracy. Fusion is applied at pixel, feature, and decision levels, improving detection precision and minimizing false alarms. The results demonstrate that combining multi-sensor data with AI significantly improves early fire identification and mapping capabilities. The developed framework provides scalable, near-real-time monitoring suitable for various forest ecosystems. It supports faster emergency response, efficient resource deployment, and improved forest management practices, contributing to the development of smarter, data-driven wildfire mitigation systems