干旱区生态环境知识资源中心(Arid): Deep learning for dune pattern mapping with the AW3D30 global surface model

Arid

DOI	10.1002/esp.4888
	Deep learning for dune pattern mapping with the AW3D30 global surface model
	Shumack, Samuel; Hesse, Paul; Farebrother, Will
通讯作者	Shumack, Samuel
来源期刊	EARTH SURFACE PROCESSES AND LANDFORMS
ISSN	0197-9337
EISSN	1096-9837
出版年	2020
卷号	45 期号:11 页码:2417-2431
英文摘要	In this paper we present a deep learning (U-Net)-based workflow for classifying linear dune landforms based on the discrete Laplacian convolution of a new global elevation dataset, the AW3D30 digital surface model. Crest vectors were then derived for landscape pattern analysis. The U-Net crest classification model was trained and evaluated on sample data from dunefields across the Australian continent. The resulting crest vectors and dune defect placement were then evaluated in typical semi-arid and arid dune landscapes in eastern central Australia where high-resolution (5 m horizontal) digital elevation models are available (for three out of our four study sites) as a reference dataset. The method was applied to quantify dune pattern metrics for the entire Simpson Desert dunefield, Australia. The U-Net does a very good job of segmenting dune crests, even where dunes are less clear in the Laplacian map (intersection over union score approximate to 0.68). When crest vectors and dune defects (network nodes) were derived, the defect predictions were typically correct (0.4 to 0.79 correctness) but incomplete (0.02 to 0.64 completeness). Much of the residual error was traced to the resolution of the input data. Through the application to the Simpson Desert, we nevertheless demonstrated that our method can effectively be used for regional-scale dune pattern analysis. Furthermore, we suggest that the combination of morphological filtering and a convolutional neural network could readily be adapted to target other geomorphic features, such as channel networks or geological lineaments. (c) 2020 John Wiley & Sons, Ltd.
英文关键词	linear dunes pattern analysis deep learning U-Net digital surface model
类型	Article
语种	英语
国家	Australia
收录类别	SCI-E
WOS记录号	WOS:000535495300001
WOS关键词	LINEAR SAND DUNE ; MORPHOMETRIC PARAMETERS ; SPATIAL-ANALYSIS ; SIMPSON DESERT ; FIELD ; BEDFORMS ; AUSTRALIA ; EVOLUTION ; VARIABILITY ; EXTRACTION
WOS类目	Geography, Physical ; Geosciences, Multidisciplinary
WOS研究方向	Physical Geography ; Geology
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/318682
作者单位	Macquarie Univ, Dept Environm Sci, Level 4,12 Wallys Walk, Sydney, NSW 2109, Australia
推荐引用方式 GB/T 7714	Shumack, Samuel,Hesse, Paul,Farebrother, Will. Deep learning for dune pattern mapping with the AW3D30 global surface model[J],2020,45(11):2417-2431.
APA	Shumack, Samuel,Hesse, Paul,&Farebrother, Will.(2020).Deep learning for dune pattern mapping with the AW3D30 global surface model.EARTH SURFACE PROCESSES AND LANDFORMS,45(11),2417-2431.
MLA	Shumack, Samuel,et al."Deep learning for dune pattern mapping with the AW3D30 global surface model".EARTH SURFACE PROCESSES AND LANDFORMS 45.11(2020):2417-2431.