Tree counting with high spatial-resolution satellite imagery based on deep neural networks

doi:10.1016/j.ecolind.2021.107591

	Tree counting with high spatial-resolution satellite imagery based on deep neural networks
	Yao, Ling 1,3,5; Liu, Tang 1,2; Qin, Jun 1,4,5; Lu, Ning 1,3,5; Zhou, Chenghu 1,3,5
刊名	ECOLOGICAL INDICATORS
	2021-06-01
卷号	125 页码:12
关键词	Forestry Deep learning algorithms Remote sensing Density estimation Tree counting
ISSN号	1470-160X
DOI	10.1016/j.ecolind.2021.107591
通讯作者	Yao, Ling(yaoling@lreis.ac.cn)
英文摘要	Forest inventory at single-tree level is of great importance to modern forest management. The inventory contains two critical parameters about trees, including their numbers and spatial locations. Traditional methods to catalogue single trees are laborious, while deep neural networks enable to discover the multi-scale features hidden in images and thus make it possible to count trees with remote sensing imagery. In this study, four different tree counting networks, which were constructed by remodeling four different classical deep convolutional neural networks, were evaluated to determine their abilities to grasp the relationship between remote sensing images and tree locations for automatic tree counting end-to-end. To this end, a tree counting dataset was constructed with remote sensing images of 0.8-m spatial resolution in distinct regions. This dataset consisted of 24 GF-II images and the corresponding manually annotated locations of trees based on these images. Thereafter, a large number of experiments were conducted to examine the performance of these networks in regards to tree counting. The results demonstrated that all networks could achieve the competitive performance (above 0.91) in terms of the determination coefficient (R-2) between the ground truth and the estimated values. The average accuracy of the Encoder-Decoder Network (one of the four networks) was greater than 91.58% and its R-2 was equal to 0.97, achieving the best performance. It has been found that the deep learning is an efficient and effective means for tree counting task.
资助项目	National Natural Science Foundation of China[41771380] ; Key Special Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou)[GML2019ZD0301] ; National Postdoctoral Program for Innovative Talents, China[BX20200100] ; National Data Sharing Infrastructure of Earth System Science
WOS研究方向	Biodiversity & Conservation ; Environmental Sciences & Ecology
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000637798700004
资助机构	National Natural Science Foundation of China ; Key Special Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou) ; National Postdoctoral Program for Innovative Talents, China ; National Data Sharing Infrastructure of Earth System Science
内容类型	期刊论文
源URL	[http://ir.igsnrr.ac.cn/handle/311030/161762]
专题	中国科学院地理科学与资源研究所
通讯作者	Yao, Ling
作者单位	1.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing, Peoples R China 2.China Univ Geosci, Beijing, Peoples R China 3.Jiangsu Ctr Collaborat Innovat Geog Informat Reso, Nanjing, Peoples R China 4.Univ Chinese Acad Sci, Beijing, Peoples R China 5.Southern Marine Sci & Engn Guangdong Lab, Guangzhou, Peoples R China
推荐引用方式 GB/T 7714	Yao, Ling,Liu, Tang,Qin, Jun,et al. Tree counting with high spatial-resolution satellite imagery based on deep neural networks[J]. ECOLOGICAL INDICATORS,2021,125:12.
APA	Yao, Ling,Liu, Tang,Qin, Jun,Lu, Ning,&Zhou, Chenghu.(2021).Tree counting with high spatial-resolution satellite imagery based on deep neural networks.ECOLOGICAL INDICATORS,125,12.
MLA	Yao, Ling,et al."Tree counting with high spatial-resolution satellite imagery based on deep neural networks".ECOLOGICAL INDICATORS 125(2021):12.