Fire-Net: A Deep Learning Framework for Active Forest Fire Detection

Table 4

Comparison of performance proposed Fire-Net algorithm with other fire detection methods.

	Index	Method	Dataset
Saeed, et al. [40]	OA: 99(%)	Deep learning based	Close rage dataset
Jang, Kang, Im, Lee, Yoon and Kim [17]	Precision: 93.08	RF and threshold based	Himawari-8 geostationary satellite data
Jiao, Zhang, Xin, Mu, Yi, Liu and Liu [18]	Precision: 83; FPR: 3.2	Deep learning (YOLOv3)	UAV dataset
Schroeder, Oliva, Giglio, Quayle, Lorenz and Morelli [14]	FPR: 0.2	Thresholding based	Landsat 8
Lin, Chen, Li, Yu, Jia, Zhang and Liang [23]	OA: 54; MD: 78	Contextual based	FengYun-2G S-VISSR data
Proposed fire-net	Precision: 93.49; FPR: 0.0001; OA: 99.98	Deep learning based	Landsat 8