AANet: an attention-based alignment semantic segmentation network for high spatial resolution remote sensing images
Guangkuo Xue, Yikun Liu, Yuwen Huang, Mingsong Li, Gongping Yang · International Journal of Remote Sensing · 2022
In this paper, we present an efficient network to tackle three critical problems in high spatial resolution (HSR) remote sensing image segmentation: (i) feature misalignment, (ii) insufficient contextual information extraction and (iii) various class imbalance issues. In detail, we propose a novel Feature Alignment Block (FAB) to suppress misalignment issues with the guide of an anchor map. Further, to extract sufficient information, we design a Contextual Augmentation Block (CAB) to augment features of different semantic levels. Finally, we present an Annealing Online Hard Example Mining (AOHEM) strategy to handle the various class imbalance issues with a view to dynamically adjust the focus of the network. We apply the above proposed designs to FPN to form our Attention-based Alignment Network (AANet). Experimental results demonstrate that the proposed method achieves promising results on the challenging iSAID and Vaihingen datasets with a better trade-off between accuracy and complexity.