A Deep Neural Networks Approach for Augmenting Samples of Land Cover Classification

doi:10.3390/land9080271

	A Deep Neural Networks Approach for Augmenting Samples of Land Cover Classification
	Zhao, Chuanpeng 1,2; Huang, Yaohuan 1,2
刊名	LAND
	2020-08-01
卷号	9 期号:8 页码:17
关键词	sample augment deep neural network small size samples land cover object-based image analysis
DOI	10.3390/land9080271
通讯作者	Huang, Yaohuan(huangyh@lreis.ac.cn)
英文摘要	Land cover is one of key indicators for modeling ecological, environmental, and climatic processes, which changes frequently due to natural factors and anthropogenic activities. The changes demand various samples for updating land cover maps, although in reality the number of samples is always insufficient. Sample augment methods can fill this gap, but these methods still face difficulties, especially for high-resolution remote sensing data. The difficulties include the following: (1) excessive human involvement, which is mostly caused by human interpretation, even by active learning-based methods; (2) large variations of segmented land cover objects, which affects the generalization to unseen areas especially for proposed methods that are validated in small study areas. To solve these problems, we proposed a sample augment method incorporating the deep neural networks using a Gaofen-2 image. To avoid error accumulation, the neural network-based sample augment (NNSA) framework employs non-iterative procedure, and augments from 184 image objects with labels to 75,112 samples. The overall accuracy (OA) of NNSA is 20% higher than that of label propagation (LP) in reference to expert interpreted results; the LP has an OA of 61.16%. The accuracy decreases by approximately 10% in the coastal validation area, which has different characteristics from the inland samples. We also compared the iterative and non-iterative strategies without external information added. The results of the validation area containing original samples show that non-iterative methods have a higher OA and a lower sample imbalance. The NNSA method that augments sample size with higher accuracy can benefit the update of land cover information.
资助项目	National Key Research and Development Program of China[2016YFC0401404] ; National Key Research and Development Program of China[2017YFB0503005] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDA23100301]
WOS关键词	REMOTE-SENSING DATA ; IMAGE CLASSIFICATION ; DOMAIN-ADAPTATION ; TRAINING SET ; ALGORITHMS ; MACHINE ; FOREST
WOS研究方向	Environmental Sciences & Ecology
语种	英语
出版者	MDPI
WOS记录号	WOS:000578818000001
资助机构	National Key Research and Development Program of China ; Strategic Priority Research Program of the Chinese Academy of Sciences
内容类型	期刊论文
源URL	[http://ir.igsnrr.ac.cn/handle/311030/157147]
专题	中国科学院地理科学与资源研究所
通讯作者	Huang, Yaohuan
作者单位	1.Chinese Acad Sci, State Key Lab Resources & Environm Informat Syst, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China 2.Univ Chinese Acad Sci, Coll Resource & Environm, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zhao, Chuanpeng,Huang, Yaohuan. A Deep Neural Networks Approach for Augmenting Samples of Land Cover Classification[J]. LAND,2020,9(8):17.
APA	Zhao, Chuanpeng,&Huang, Yaohuan.(2020).A Deep Neural Networks Approach for Augmenting Samples of Land Cover Classification.LAND,9(8),17.
MLA	Zhao, Chuanpeng,et al."A Deep Neural Networks Approach for Augmenting Samples of Land Cover Classification".LAND 9.8(2020):17.