A Deep Learning Model to Recognize and Quantitatively Analyze Cold Seep Substrates and the Dominant Associated Species

doi:10.3389/fmars.2021.775433

	A Deep Learning Model to Recognize and Quantitatively Analyze Cold Seep Substrates and the Dominant Associated Species
	Wang, Haining 1,2,3; Fu, Xiaoxue 4; Zhao, Chengqian 5; Luan, Zhendong 2,3,6,7; Li, Chaolun1,2,3,6
	2021-11-25
发表期刊	FRONTIERS IN MARINE SCIENCE
卷号	8 页码:11
通讯作者	Li, Chaolun(lcl@qdio.ac.cn)
摘要	Characterizing habitats and species distribution is important to understand the structure and function of cold seep ecosystems. This paper develops a deep learning model for the fast and accurate recognition and classification of substrates and the dominant associated species in cold seeps. Considering the dense distribution of the dominant associated species and small objects caused by overlap in cold seeps, the feature pyramid network (FPN) embed into the faster region-convolutional neural network (R-CNN) was used to detect large-scale changes and small missing objects without increasing the number of calculations. We applied three classifiers (Faster R-CNN + FPN for mussel beds, lobster clusters and biological mixing, CNN for shell debris and exposed authigenic carbonates, and VGG16 for reduced sediments and muddy bottom) to improve the recognition accuracy of substrates. The model's results were manually verified using images obtained in the Formosa cold seep during a 2016 cruise. The recognition accuracy of the two dominant species, e.g., Gigantidas platifrons and Munidopsidae could be 70.85 and 56.16%, respectively. Seven subcategories of substrates were also classified with a mean accuracy of 74.87%. The developed model is a promising tool for the fast and accurate characterization of substrates and epifauna in cold seeps, which is crucial for large-scale quantitative analyses.
关键词	cold seep substrates epifauna Faster R-CNN FPN VGG16
DOI	10.3389/fmars.2021.775433
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[42030407] ; National Natural Science Foundation of China[42076091] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDA22050303] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDB42020401] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDA22050302] ; National Key R&D Program of the Ministry of Science and Technology[2018YFC0310802] ; Youth Innovation Promotion Association of the Chinese Academy of Sciences
WOS研究方向	Environmental Sciences & Ecology ; Marine & Freshwater Biology
WOS类目	Environmental Sciences ; Marine & Freshwater Biology
WOS记录号	WOS:000729362000001
出版者	FRONTIERS MEDIA SA
引用统计	被引频次：1[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.qdio.ac.cn/handle/337002/177447
专题	海洋生态与环境科学重点实验室海洋地质与环境重点实验室
通讯作者	Li, Chaolun
作者单位	1.Chinese Acad Sci, Inst Oceanol, Key Lab Marine Ecol & Environm Sci, Qingdao, Peoples R China 2.Univ Chinese Acad Sci, Beijing, Peoples R China 3.Chinese Acad Sci, Inst Oceanol, Deep Sea Res Ctr, Qingdao, Peoples R China 4.Qingdao Univ Sci & Technol, Artificial Intelligence Lab, Qingdao, Peoples R China 5.Fudan Univ, Sch Informat Sci & Technol, Shanghai, Peoples R China 6.Chinese Acad Sci, Ctr Ocean Mega Sci, Qingdao, Peoples R China 7.Chinese Acad Sci, Inst Oceanol, Key Lab Marine Geol & Environm, Qingdao, Peoples R China
第一作者单位	中国科学院海洋研究所
通讯作者单位	中国科学院海洋研究所
推荐引用方式 GB/T 7714	Wang, Haining,Fu, Xiaoxue,Zhao, Chengqian,et al. A Deep Learning Model to Recognize and Quantitatively Analyze Cold Seep Substrates and the Dominant Associated Species[J]. FRONTIERS IN MARINE SCIENCE,2021,8:11.
APA	Wang, Haining,Fu, Xiaoxue,Zhao, Chengqian,Luan, Zhendong,&Li, Chaolun.(2021).A Deep Learning Model to Recognize and Quantitatively Analyze Cold Seep Substrates and the Dominant Associated Species.FRONTIERS IN MARINE SCIENCE,8,11.
MLA	Wang, Haining,et al."A Deep Learning Model to Recognize and Quantitatively Analyze Cold Seep Substrates and the Dominant Associated Species".FRONTIERS IN MARINE SCIENCE 8(2021):11.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
fmars-08-775433.pdf（3853KB）	期刊论文	出版稿	限制开放	CC BY-NC-SA	浏览