Table 3 This table presents a comparative analysis of different deep learning models for wildfire segmentation using SAR imagery. It evaluates the models based on key performance metrics, including Dice Loss, F1 Score, IoU, Pixel Accuracy and Custom Loss Function. The results highlight the effectiveness of FPANet (SAR) in achieving superior segmentation performance compared to other architectures.

From: Hybrid learning framework for synergistic fusion of SAR and optical UAV data in wildfire surveillance