Domain Adaptive Land-Cover Classification via Local Consistency and Global Diversity

Unsupervised domain adaptive land-cover classification has recently gained more and more attention. Unsupervised domain adaptation (UDA) aims to learn a model from the annotated source data and the unlabeled target data that can perform well on the target domain. The existing UDA frameworks based on adversarial training and self-training methods have boosted this field a lot. However, these methods almost all originate from the computer vision field, and they ignore the very nature of high-resolution remote sensing (HRS) images. The core insight of this article is that a good land-cover classification result always has strong local consistency and good global diversity, which makes it possible to construct a metric representing the properties of good land-cover mapping, to improve the existing UDA algorithms. First, based on this finding, we prove that local consistency and global diversity can be measured by the Frobenius norm and nuclear norm, respectively. Second, we propose a novel local consistency and global diversity metric (LCGDM), which can be easily integrated into the existing UDA frameworks. Finally, the experiments conducted on the LoveDA dataset prove the validity of the proposed metric, which can not only improve the overall land-cover mapping but also the categorywise prediction.