André Minoro Fusioka

Active fire segmentation through satellite imagery is fundamental to prevention and damage prediction. Algorithms to address this problem usually rely on sensor-specific thresholds, empirically chosen based on a few image samples, and thus, are susceptible to many errors. Deep learning algorithms automatically extract information through various levels of abstraction, avoiding human intervention for feature engineering. However, training a deep network from scratch requires massive labeled data, and efforts in this direction have already been made for the Landsat-8 satellite. For Sentinel-2, another important satellite that has band wavelength similarities with Landsat-8, there is a limited number of such initiatives, with few studies even concerning hand-crafted (traditional) algorithms. In this context, we explored in this article the transfer of knowledge for active fire segmentation from Landsat-8 to Sentinel-2, avoiding the need of a vast amount of labeled data from Sentinel-2, and also reducing the computational resources for training. We also compiled a benchmark containing 12 584 image patches extracted from 26 Sentinel-2 images from around the globe, along with manually annotated fire pixels, to assess the algorithm's response compared to a human specialist. In a series of transfer learning experiments by using the U-Net architecture, we showed that even a single labeled image from Sentinel-2 for training was sufficient to allow achieving an F -score metric of 84.9% when Landsat-8 knowledge is transferred and further fine-tuned in the machine learning process, while the best hand-crafted algorithm designed for Sentinel-2 achieved an F -score of 75.8% in the segmentation task.

Active fire segmentation in satellite imagery is a critical remote sensing task, providing essential support for planning, decision-making, and policy development. Several techniques have been proposed for this problem over the years, generally based on specific equations and thresholds, which are sometimes empirically chosen. Some satellites, such as MODIS and Landsat-8, have consolidated algorithms for this task. However, for other important satellites such as Sentinel-2, this is still an open problem. In this letter, we explore the possibility of using transfer learning to train convolutional and transformer-based deep architectures (U-Net, DeepLabV3+, and SegFormer) for active fire segmentation. We pretrain these architectures based on Landsat-8 images and automatically labeled samples and fine-tune them to Sentinel-2 images. The experiments show that the proposed method achieves F1 -scores of up to 88.4% for Sentinel-2 images, outperforming three threshold-based algorithms by at least 19% while maintaining a low demand for manually labeled samples. We also address detection over seam-line regions that present a particular challenge for existing methods. The source code and trained models are available at https://github.com/Minoro/l8tos2-transf-seamlines.

Active fire segmentation in satellite imagery is a critical remote sensing task, providing essential support for planning, decision-making, and policy development. Several techniques have been proposed for this problem over the years, generally based on specific equations and thresholds, which are sometimes empirically chosen. Some satellites, such as MODIS and Landsat-8, have consolidated algorithms for this task. However, for other important satellites such as Sentinel-2, this is still an open problem. In this letter, we explore the possibility of using transfer learning to train convolutional and transformer-based deep architectures (U-Net, DeepLabV3+, and SegFormer) for active fire segmentation. We pretrain these architectures based on Landsat-8 images and automatically labeled samples and fine-tune them to Sentinel-2 images. The experiments show that the proposed method achieves F1 -scores of up to 88.4% for Sentinel-2 images, outperforming three threshold-based algorithms by at least 19% while maintaining a low demand for manually labeled samples. We also address detection over seam-line regions that present a particular challenge for existing methods. The source code and trained models are available at https://github.com/Minoro/l8tos2-transf-seamlines.