Cross-city matters: A multimodal remote sensing benchmark dataset for cross-city semantic segmentation using high-resolution domain adaptation networks

• Published in: Remote sensing of environment Vol. 299; p. 113856
• Main Authors: Hong, Danfeng, Zhang, Bing, Li, Hao, Li, Yuxuan, Yao, Jing, Li, Chenyu, Werner, Martin, Chanussot, Jocelyn, Zipf, Alexander, Zhu, Xiao Xiang
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 15.12.2023
• Subjects: artificial intelligence; China; class; Cross-city; data collection; Deep learning; Dice loss; domain; Domain adaptation; environment; Germany; High-resolution network; Land cover; Multimodal benchmark datasets; Remote sensing; Segmentation; topology; China; Germany; Deep learning; High-resolution network; Segmentation; Multimodal benchmark datasets; Domain adaptation; Dice loss; Land cover; Remote sensing; Cross-city
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2023.113856

Artificial intelligence (AI) approaches nowadays have gained remarkable success in single-modality-dominated remote sensing (RS) applications, especially with an emphasis on individual urban environments (e.g., single cities or regions). Yet these AI models tend to meet the performance bottleneck in the case studies across cities or regions, due to the lack of diverse RS information and cutting-edge solutions with high generalization ability. To this end, we build a new set of multimodal remote sensing benchmark datasets (including hyperspectral, multispectral, SAR) for the study purpose of the cross-city semantic segmentation task (called C2Seg dataset), which consists of two cross-city scenes, i.e., Berlin-Augsburg (in Germany) and Beijing-Wuhan (in China). Beyond the single city, we propose a high-resolution domain adaptation network, HighDAN for short, to promote the AI model's generalization ability from the multi-city environments. HighDAN is capable of retaining the spatially topological structure of the studied urban scene well in a parallel high-to-low resolution fusion fashion but also closing the gap derived from enormous differences of RS image representations between different cities by means of adversarial learning. In addition, the Dice loss is considered in HighDAN to alleviate the class imbalance issue caused by factors across cities. Extensive experiments conducted on the C2Seg dataset show the superiority of our HighDAN in terms of segmentation performance and generalization ability, compared to state-of-the-art competitors. The C2Seg dataset and the semantic segmentation toolbox (involving the proposed HighDAN) will be available publicly at https://github.com/danfenghong/RSE_Cross-city. •A new multimodal remote sensing benchmark for cross-city semantic segmentation.•Propose a high-resolution domain adaptation network for semantic segmentation.•Balance spatial topology, domain gaps, eases city class imbalance with Dice loss.