干旱区生态环境知识资源中心(Arid): Effectiveness of machine learning methods for water segmentation with ROI as the label: A case study of the Tuul River in Mongolia

Arid

DOI	10.1016/j.jag.2021.102497
	Effectiveness of machine learning methods for water segmentation with ROI as the label: A case study of the Tuul River in Mongolia
	Li, Kai; Wang, Juanle; Yao, Jinyi
通讯作者	Wang, JL (corresponding author), Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China.
来源期刊	INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
ISSN	1569-8432
EISSN	1872-826X
出版年	2021
卷号	103
英文摘要	The carrying capacity of water resources is key to the sustainable development of arid and semi-arid regions. There are important challenges related to the detection of discontinuous and crooked water bodies in the vast Mongolian Plateau, despite the availability of remote sensing technology which has the advantage of facilitating water observations over large areas and timelines. Given the high cost and low coverage of high-resolution images and the low resolution of images with high coverage, this study proposes a pixel-based convolutional neural network (CNN) method for the application of water extracted from the region of interest (ROI) to mediumresolution Landsat images. The pixel-based CNN method combines the texture and spectral features of the ground object by connecting the center pixels of the images to the surrounding pixels. ROI is used instead of full-label datasets, reduce the difficulty of building labels in low-to-medium-resolution images. Taking the Tuul River in Mongolia as a case, the pixel-based CNN method, the normalized difference water index threshold (NDWI) method, the modified normalized difference water index (MNDWI) threshold method, U-net model in deep learning, and the pixel-based deep neural network (DNN) method were used with medium-resolution Landsat 8 images with ROI labels. The pixel-based CNN method shows better water extraction results for the cloud, cloud shadows, and building areas, compared with other methods. The method proposed in this study had the highest verification accuracy (92.07%). It also has the advantages of fewer training parameters and shorter training time. The training accuracies of the pixel-based CNN, pixel-based DNN, and U-net were 99.90%, 96.98%, and 93.70%, respectively. All training models and calling methods were uploaded to GitHub (https://github.com/CaryLee 17/Pixel-based-CNN).
英文关键词	Water Segmentation Pixel-based CNN ROI U-net Mongolia
类型	Article
语种	英语
开放获取类型	hybrid
收录类别	SCI-E
WOS记录号	WOS:000696914600002
WOS关键词	SPECTRAL CHARACTERISTICS ; SATELLITE IMAGERY ; INDEX NDWI ; CLASSIFICATION ; BODY
WOS类目	Remote Sensing
WOS研究方向	Remote Sensing
来源机构	中国科学院地理科学与资源研究所
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/363615
作者单位	[Li, Kai; Wang, Juanle; Yao, Jinyi] Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China; [Li, Kai] China Univ Min & Technol Beijing, Coll Geosci & Surveying Engn, Beijing 100083, Peoples R China; [Yao, Jinyi] Shandong Univ Technol, Sch Civil & Architectural Engn, Zibo 255049, Peoples R China
推荐引用方式 GB/T 7714	Li, Kai,Wang, Juanle,Yao, Jinyi. Effectiveness of machine learning methods for water segmentation with ROI as the label: A case study of the Tuul River in Mongolia[J]. 中国科学院地理科学与资源研究所,2021,103.
APA	Li, Kai,Wang, Juanle,&Yao, Jinyi.(2021).Effectiveness of machine learning methods for water segmentation with ROI as the label: A case study of the Tuul River in Mongolia.INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION,103.
MLA	Li, Kai,et al."Effectiveness of machine learning methods for water segmentation with ROI as the label: A case study of the Tuul River in Mongolia".INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 103(2021).