期刊
GISCIENCE & REMOTE SENSING
卷 59, 期 1, 页码 2247-2265出版社
TAYLOR & FRANCIS LTD
DOI: 10.1080/15481603.2022.2156123
关键词
Crop mapping; multi-source unsupervised domain adaptation; time-series remote sensing; transfer learning; deep learning
资金
- Zhejiang Provincial Natural Science Foundation of China
- Open Fund of Key Laboratory of Urban Land Resources Monitoring and Simulation
- Public Projects of Ningbo City
- Ningbo Science and Technology Innovation 2025 Major Special Project
- National Natural Science Foundation of China
- [LQ22D010007]
- [KF-2021-06-089]
- [2022S101]
- [2021Z107]
- [42201354]
- [42171326]
This study proposed a MUDA crop classification model (MUCCM) for unsupervised crop mapping and demonstrated its high performance in multiple target domains. The results indicated that the UDA models outperformed the direct transfer models significantly, and the MUCCM achieved the highest classification accuracy in each target domain.
Accurate crop mapping is critical for agricultural applications. Although studies have combined deep learning methods and time-series satellite images to crop classification with satisfactory results, most of them focused on supervised methods, which are usually applicable to a specific domain and lose their validity in new domains. Unsupervised domain adaptation (UDA) was proposed to solve this limitation by transferring knowledge from source domains with labeled samples to target domains with unlabeled samples. Particularly, multi-source UDA (MUDA) is a powerful extension that leverages knowledge from multiple source domains and can achieve better results in the target domain than single-source UDA (SUDA). However, few studies have explored the potential of MUDA for crop mapping. This study proposed a MUDA crop classification model (MUCCM) for unsupervised crop mapping. Specifically, 11 states in the U.S. were selected as the multi-source domains, and three provinces in Northeast China were selected as individual target domains. Ten spectral bands and five vegetation indexes were collected at a 10-day interval from time-series Sentinel-2 images to build the MUCCM. Subsequently, a SUDA model Domain Adversarial Neural Network (DANN) and two direct transfer methods, namely, the deep neural network and random forest, were constructed and compared with the MUCCM. The results indicated that the UDA models outperformed the direct transfer models significantly, and the MUCCM was superior to the DANN, achieving the highest classification accuracy (OA>85%) in each target domain. In addition, the MUCCM also performed best in in-season forecasting and crop mapping. This study is the first to apply a MUDA to crop classification and demonstrate a novel, effective solution for high-performance crop mapping in regions without labeled samples.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据