Semi-Supervised Semantic Segmentation for Remote Sensing Images via Multiscale Uncertainty Consistency and Cross-Teacher–Student Attention

Semi-supervised learning offers an appealing solution for remote sensing (RS) image segmentation to relieve the burden of labor-intensive pixel-level labeling. However, RS images pose unique challenges, including rich multiscale features and high interclass similarity. To address these problems, this article proposes a novel semi-supervised multiscale uncertainty and cross-teacher-student attention (MUCA) model for RS image semantic segmentation tasks. Specifically, MUCA constrains the consistency among feature maps at different layers of the network by introducing a multiscale uncertainty consistency (MSUC) regularization. It improves the multiscale learning capability of semi-supervised algorithms on unlabeled data. In addition, MUCA utilizes a cross-teacher–student attention (CTSA) mechanism to guide the student network, guiding the student network to construct more discriminative feature representations through complementary features from the teacher network. This design effectively integrates weak augmentation (WA) and strong augmentation (SA) to further boost segmentation performance. To verify the effectiveness of our model, we conduct extensive experiments on ISPRS-Potsdam and LoveDA datasets. The experimental results show the superiority of our method over state-of-the-art semi-supervised methods. Notably, our model excels in distinguishing highly similar objects, showcasing its potential for advancing semi-supervised RS image segmentation tasks.