烟雾
红外线的
分割
融合
遥感
环境科学
计算机科学
计算机视觉
人工智能
材料科学
光学
气象学
地质学
地理
物理
语言学
哲学
作者
Dimitrios Fotiou,Vasileios Mygdalis,Ioannis Pitas
标识
DOI:10.1016/j.patrec.2026.04.024
摘要
Concurrent image segmentation of flames and smoke is challenging, as smoke frequently obscures fire in standard RGB imagery, necessitating the use of other spectral bands such as Infrared (IR). Existing multimodal models are either too computationally demanding for real-time deployment or too lightweight to capture small sparse fire patterns that may escalate into large wildfires. Moreover, they are typically trained and validated on simplistic datasets, such as Corsican or FLAME1, which lack the dense smoke occlusion present in real-world scenarios. We introduce RoboFireFuseNet (RFFNet), a real-time deep neural network that fuses RGB and IR data using attention mechanisms, a detail-preserving decoder, and class-balance training techniques. RFFNet establishes a benchmark on challenging, realistic wildfire datasets with dense smoke, creating a foundation for practical comparison in future wildfire segmentation research. Despite its lightweight design, it also achieves state-of-the-art results on a general urban benchmark, demonstrating versatility. Its combination of accuracy, real-time performance, and multimodal fusion makes RFFNet well-suited for proactive and accurate wildfire monitoring. Code is available: https://github.com/dimfot3/RoboFireFuseNet
科研通智能强力驱动
Strongly Powered by AbleSci AI