Post-Disaster Affected Area Segmentation with a Vision Transformer (ViT)-based EVAP Model using Sentinel-2 and Formosat-5 Imagery
This work addresses the challenge of accurate disaster mapping for emergency responders and agencies when ground truth data is limited, though it is incremental as it builds on existing EVAP products and ViT methods.
The authors tackled the problem of segmenting disaster-affected areas from remote sensing imagery by proposing a vision transformer-based framework that uses weakly supervised training with Sentinel-2 and Formosat-5 data, resulting in improved smoothness and reliability of segmentation results as demonstrated in case studies on the 2022 Poyang Lake drought and 2023 Rhodes wildfire.
We propose a vision transformer (ViT)-based deep learning framework to refine disaster-affected area segmentation from remote sensing imagery, aiming to support and enhance the Emergent Value Added Product (EVAP) developed by the Taiwan Space Agency (TASA). The process starts with a small set of manually annotated regions. We then apply principal component analysis (PCA)-based feature space analysis and construct a confidence index (CI) to expand these labels, producing a weakly supervised training set. These expanded labels are then used to train ViT-based encoder-decoder models with multi-band inputs from Sentinel-2 and Formosat-5 imagery. Our architecture supports multiple decoder variants and multi-stage loss strategies to improve performance under limited supervision. During the evaluation, model predictions are compared with higher-resolution EVAP output to assess spatial coherence and segmentation consistency. Case studies on the 2022 Poyang Lake drought and the 2023 Rhodes wildfire demonstrate that our framework improves the smoothness and reliability of segmentation results, offering a scalable approach for disaster mapping when accurate ground truth is unavailable.