Deep learning in environmental remote sensing: Achievements and challenges

被引:1135
作者
Yuan, Qiangqiang [1 ,4 ,6 ]
Shen, Huanfeng [2 ,4 ,5 ]
Li, Tongwen [2 ]
Li, Zhiwei [2 ]
Li, Shuwen [1 ]
Jiang, Yun [2 ]
Xu, Hongzhang [1 ]
Tan, Weiwei [3 ]
Yang, Qianqian [1 ]
Wang, Jiwen [1 ]
Gao, Jianhao [1 ]
Zhang, Liangpei [3 ,4 ]
机构
[1] Wuhan Univ, Sch Geodesy & Geomat, Wuhan, Peoples R China
[2] Wuhan Univ, Sch Resource & Environm Sci, Wuhan, Peoples R China
[3] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan, Peoples R China
[4] Collaborat Innovat Ctr Geospatial Technol, Wuhan, Peoples R China
[5] Wuhan Univ, Key Lab Geog Informat Syst, Minist Educ, Wuhan, Peoples R China
[6] Wuhan Univ, Key Lab Geospace Environm & Geodesy, Minist Educ, Wuhan, Peoples R China
基金
中国国家自然科学基金;
关键词
Environmental remote sensing; Deep learning; Parameter retrieval; Neural network; ARTIFICIAL NEURAL-NETWORK; LAND-SURFACE TEMPERATURE; SNOW WATER EQUIVALENT; RETRIEVING SOIL-MOISTURE; GLOBAL SOLAR-RADIATION; AEROSOL OPTICAL DEPTH; CROP YIELD PREDICTION; LEAF-AREA INDEX; PASSIVE MICROWAVE DATA; NDVI TIME-SERIES;
D O I
10.1016/j.rse.2020.111716
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Various forms of machine learning (ML) methods have historically played a valuable role in environmental remote sensing research. With an increasing amount of "big data" from earth observation and rapid advances in ML, increasing opportunities for novel methods have emerged to aid in earth environmental monitoring. Over the last decade, a typical and state-of-the-art ML framework named deep learning (DL), which is developed from the traditional neural network (NN), has outperformed traditional models with considerable improvement in performance. Substantial progress in developing a DL methodology for a variety of earth science applications has been observed. Therefore, this review will concentrate on the use of the traditional NN and DL methods to advance the environmental remote sensing process. First, the potential of DL in environmental remote sensing, including land cover mapping, environmental parameter retrieval, data fusion and downscaling, and information reconstruction and prediction, will be analyzed. A typical network structure will then be introduced. Afterward, the applications of DL environmental monitoring in the atmosphere, vegetation, hydrology, air and land surface temperature, evapotranspiration, solar radiation, and ocean color are specifically reviewed. Finally, challenges and future perspectives will be comprehensively analyzed and discussed.
引用
收藏
页数:24
相关论文
共 299 条
[1]   A new neural network approach including first guess for retrieval of atmospheric water vapor, cloud liquid water path, surface temperature, and emissivities over land from satellite microwave observations [J].
Aires, F ;
Prigent, C ;
Rossow, WB ;
Rothstein, M .
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 2001, 106 (D14) :14887-14907
[2]   Using Remote Sensing Satellite Data and Artificial Neural Network for prediction of Potato yield in Bangladesh [J].
Akhand, Kawsar ;
Nizamuddin, Mohammad ;
Roytman, Leonid ;
Kogan, Felix .
REMOTE SENSING AND MODELING OF ECOSYSTEMS FOR SUSTAINABILITY XIII, 2016, 9975
[3]   Global downscaling of remotely sensed soil moisture using neural networks [J].
Alemohammad, Seyed Hamed ;
Kolassa, Jana ;
Prigent, Catherine ;
Aires, Filipe ;
Gentine, Pierre .
HYDROLOGY AND EARTH SYSTEM SCIENCES, 2018, 22 (10) :5341-5356
[4]  
Alipour A., 2014, J. Climatol, V2014, P839205, DOI [10.1155/2014/839205, DOI 10.1155/2014/839205]
[5]  
Allen R.G., 1998, Paper No. 56
[6]  
[Anonymous], J GEOPHYS RES ATMOS
[7]  
[Anonymous], ESTIMATING CROP PRIM
[8]  
[Anonymous], COMPUT INTELL NEUROS
[9]  
[Anonymous], 2014, PROC C EMPIRICAL MET
[10]  
[Anonymous], GEOSC REM SENS S 200