Label Propagation and Contrastive Regularization for Semi-supervised Semantic Segmentation of Remote Sensing Images

• Published in: IEEE transactions on geoscience and remote sensing 2023-05, p.1-1
• Main Authors: Yang, Zhujun, Yan, Zhiyuan, Diao, Wenhui, Zhang, Qiang, Kang, Yuzhuo, Li, Junxi, Li, Xinming, Sun, Xian
• Format: Article
• Language: English
• Subjects: contrastive regularization; label propagation; Perturbation methods; Prototypes; Reliability; Remote sensing; remote sensing images; Semantic segmentation; semi-supervised learning; Sensors; Training
• ISSN: 0196-2892
• DOI: 10.1109/TGRS.2023.3277203

Remarkable progress based on deep neural networks has been achieved on the semantic segmentation in remote sensing images. However, pixel-level labeling is expensive for remote sensing images. Semi-supervised semantic segmentation becomes an alternative approach to reduce the cost of annotation, and it is crucial to utilize efficiently a large number of unlabeled data. Nevertheless inevitably, there is the unbalanced class distribution between labeled and unlabeled data of remote sensing scene. Existing semi-supervised methods train unlabeled images in isolation from labeled images and only learn reliable pixel pseudo-labels, leading to underutilization of unlabeled images. This article proposes a novel semi-supervised semantic segmentation approach based on label propagation and contrastive regularization for remote sensing images. Specifically, the unlabeled images are augmented by randomly copy-pasting the class regions from labeled images. A prototype feature constraint module is used to enforce the constraint on the pixel features of unlabeled images relying on the prototype features from labeled images, achieving feature alignment on the entire dataset. Furthermore, we present the region contrastive learning module that guides the model to learn feature consistency under different perturbations and compact feature representations over class regions on unlabeled images. Extensive experimental results on multiple remote sensing datasets demonstrate that our proposed approach achieves superior performance compared with state-of-the-art semi-supervised semantic segmentation methods.