Smoke Source Localization Using Optical Flow Imagery in Video-Based Detection

Forest fires pose significant global threats, causing widespread environmental damage and endangering lives. Effective early detection systems are crucial for mitigating these impacts, and video surveillance has become a key tool due to its ability to monitor large areas in real-time. However, accurately localizing the smoke source remains a challenge, especially in dynamic outdoor environments where traditional methods often fail. This paper presents a novel video-based framework for smoke source localization, combining optical flow analysis with advanced segmentation techniques. Our approach identifies the smoke source by analyzing the divergence of optical flow within detected smoke regions, providing a robust solution even in complex scenarios. Experimental results demonstrate that the proposed method effectively overcomes the challenges posed by smoke's irregular movement, offering significant improvements in both accuracy and computational efficiency. The Farneback optical flow algorithm was found to offer the best balance between performance and resource usage, making it a suitable choice for practical applications.