SemiPSCN: Polarization Semantic Constraint Network for Semi-Supervised Segmentation in Large-Scale and Complex-Valued PolSAR Images

• Published in: IEEE transactions on geoscience and remote sensing Vol. 62; pp. 1 - 18
• Main Authors: Zeng, Xuan, Wang, Zhirui, Wang, Yuelei, Rong, Xuee, Guo, Pengyu, Gao, Xin, Sun, Xian
• Format: Journal Article
• Language: English
• Published: New York IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Annotations; Category representation; Consistency; Convolutional neural networks; Feature extraction; Image processing; Image segmentation; Localization; Modules; polarimetric synthetic aperture radar (PolSAR); pseudolabels; regional consistency; Representations; SAR (radar); Segmentation; Semi-supervised learning; semi-supervised segmentation; Semisupervised learning; Synthetic aperture radar; Task analysis; Terrain; Training
• ISSN: 0196-2892; 1558-0644
• DOI: 10.1109/TGRS.2023.3333431

Since polarimetric synthetic aperture radar (PolSAR) terrain segmentation is a dense prediction task, the disadvantage of inadequate labeled samples greatly limits its performance. In this article, we present a semi-supervised segmentation network called SemiPSCN to reduce the data reliance on label annotation, which integrates semi-supervised learning (SSL) paradigm and the characteristics of PolSAR data into a unified architecture. First, considering the unreliability of pseudolabels caused by noise interference in PolSAR data, a pseudolabel error localization (PEL) module is designed. By mapping the pixels that have mispredictions in pseudolabels, PEL can greatly enhance the confidence of pseudolabels. Then, SemiPSCN introduces a category representation constraint (CRC) module to explicitly boost the category consistency between labeled and unlabeled PolSAR data. Via explicit intracategory and intercategory constraints, CRC can guarantee the invariant representations on the same category region between labeled and unlabeled data. Furthermore, a region consistency constraint (RCC) module is designed to enhance the regional consistency in PolSAR data. RCC leverages the conception of graph to model the understanding of spatial relationships among terrain targets, thereby facilitating consistent spatial region expression in semi-supervised process. Finally, we build a challenging large-scale dataset called LSPolSAR-Seg and conduct abundant experiments on LSPolSAR-Seg. SemiPSCN exhibits superior performance when compared with other advanced approaches, especially improving mean intersection over union (mIoU) by 3.44%-12.77% under 20% split setting, which promotes the performance to a state-of-the-art level.