Land Cover Classification in Foggy Conditions: Toward Robust Models

Robust semantic labeling of high-resolution remote sensing images (RSIs) in foggy conditions is crucial for automatic monitoring of land covers. This remains a challenging task owing to the low interclass differentiation yet high intraclass variance and geometric size diversity. Although conventional convolutional neural networks (CNNs) have demonstrated state-of-the-art (SOTA) performance in semantic segmentation, most networks are primarily concerned with standard accuracy, while the influence on robustness is rarely explored. This letter proposes a reliable framework which is evaluated across various severity levels of fog corruptions. Utilizing HRNet as the backbone to maintain high-resolution representations, we develop a multimodal fusion module (MMF) to exploit the complementary information of lidar and multispectral data. Based on the evaluation experiment on fog corrupted datasets, our model demonstrates promising performance with an average mean Intersection over Union (mIoU) on the clean along with the corrupted datasets exceeding 80% and 56%, respectively.