Class-Aware Domain Adaptation for Semantic Segmentation of Remote Sensing Images | |
Xu QS(徐青松)1,2,3; Yuan X(袁鑫)4; 欧阳朝军1,2,5 | |
刊名 | IEEE Transactions on Geoscience and Remote Sensing |
2020 | |
页码 | DOI: 10.1109/TGRS.2020.3031926 |
关键词 | UDA semantic segmentation cross-scene and cross-spectrum remote sensing images lass-aware generative adversarial network (CaGAN) global domain alignment (GDA) class-aware domain alignment (CDA) |
ISSN号 | 1558-0644 |
DOI | 10.1109/TGRS.2020.3031926 |
产权排序 | 1 |
文献子类 | Early Access |
英文摘要 | Unsupervised domain adaptation (UDA) for the semantic segmentation of remote sensing images is challenging since the same class of objects may have different spectra while the different class of objects may have the same spectrum. To address this issue, we propose a class-aware generative adversarial network (CaGAN) for UDA semantic segmentation of multisource remote sensing images, which explicitly models the discrepancies of intraclass and the interclass between the source domain images with labels and the target domain images without labels. Specifically, first, to enhance the global domain alignment (GDA), we propose a transferable attention alignment (TAA) procedure to add more fine-grained features into the adversarial learning framework. Then, we propose a novel class-aware domain alignment (CDA) approach in semantic segmentation. CDA mainly includes two parts: the first one is adaptive category selection, which is to alleviate the class imbalance and select the reliable per-category centers in the source and target domains; the second one is adaptive category alignment, which is to model the intraclass compactness and interclass separability from source-only, target-only, and joint source and target images. Finally, the CDA plays as a penalty of GDA to train GaGAN in an alternating and iterative manner. Experiments on domain adaptation of space to space, spectrum to spectrum, both space-to-space and spectrum-to-spectrum data sets demonstrate that CaGAN outperforms the current state-of-the-art methods, which may serve as a starting point and baseline for the comprehensive applications of semantic segmentation in cross-space and cross-spectrum remote sensing images. |
URL标识 | 查看原文 |
语种 | 英语 |
内容类型 | 期刊论文 |
源URL | [http://ir.imde.ac.cn/handle/131551/50786] |
专题 | 成都山地灾害与环境研究所_山地灾害与地表过程重点实验室 |
通讯作者 | Xu QS(徐青松) |
作者单位 | 1.Key Laboratory of Mountain Hazards and Surface Process 2.Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, China 3.Bell Labs, Murray Hill, NJ 07974 USA 4.the School of Engineering Science, University of Chinese Academy of Sciences, Beijing 100049, China 5.the CAS Center for Excellence in Tibetan Plateau Earth Sciences, Chinese Academy of Sciences (CAS), Beijing 100101, China |
推荐引用方式 GB/T 7714 | Xu QS,Yuan X,欧阳朝军. Class-Aware Domain Adaptation for Semantic Segmentation of Remote Sensing Images[J]. IEEE Transactions on Geoscience and Remote Sensing,2020:DOI: 10.1109/TGRS.2020.3031926. |
APA | Xu QS,Yuan X,&欧阳朝军.(2020).Class-Aware Domain Adaptation for Semantic Segmentation of Remote Sensing Images.IEEE Transactions on Geoscience and Remote Sensing,DOI: 10.1109/TGRS.2020.3031926. |
MLA | Xu QS,et al."Class-Aware Domain Adaptation for Semantic Segmentation of Remote Sensing Images".IEEE Transactions on Geoscience and Remote Sensing (2020):DOI: 10.1109/TGRS.2020.3031926. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论