Identification of soybean in Argentina using Sentinel-2 composite images

Linsheng Huang, Yue Chen, Yuhao Pan, Zihang Lou, Shijun Zheng, Xiaoyang Zhang, Le Yu, Shengwei Liu, Dailiang Peng

Abstract


Soybean is one of the most important oil crops, and Argentina is the third-largest soybean producer in the world, accounting for 17% of the global soybean yield. Timely and accurate information on soybean spatial distribution is critical for ensuring global food security. Sentinel-2 multispectral data and machine learning classification models are used to investigate the potential of soybean identification in the early stage of the growing season in Argentina, with the help of Google Earth Engine (GEE). The earliest time window and optimal feature set for soybean identification are explored. Results are as follows: 1) the random forest (RF) classification model demonstrated the highest level of classification accuracy compared to the backpropagation neural network (BPNN), support vector machine (SVM), and naive Bayes (NB) models; 2) Soybean can be accurately identified as early as the end of February (filling stage), which is approximately one month before harvest; 3) The optimal feature-subset can reduce the amount of input data by 80% while maintaining high classification accuracy. The overall accuracy (OA) of the RF classification model is 85.87%, and the relative error between the estimated soybean planting area and the agricultural statistics is 3.45%. This study provided a high-precision method for early-season identification of soybeans over large scales. The results can provide a data support for early futures trading and agricultural insurance, as well as a reference for policy-making to ensure global soybean food security.
Keywords: soybean, machine learning, time window, feature selection, Sentinel-2, Google Earth Engine
DOI: 10.25165/j.ijabe.20241705.7634

Citation: Huang L S, Chen Y, Pan Y H, Lou Z H, Zheng S J, Zhang X Y, et al. Identification of soybean in Argentina using Sentinel-2 composite images. Int J Agric & Biol Eng, 2024; 17(5): 266-274.

Keywords


soybean, machine learning, time window, feature selection, Sentinel-2, Google Earth Engine

Full Text:

PDF

References


Li X Y, Yu L, Peng D L, Gong P. A large-scale, long time-series (1984‒2020) of soybean mapping with phenological features: Heilongjiang Province as a test case. International Journal of Remote Sensing, 2021; 42(19): 7332–7356.

King L A, Adusei B, Stehman S V, Potapov P V, Song X P, Krylov A, et al. A multi-resolution approach to national-scale cultivated area estimation of soybean. Remote Sensing of Environment, 2017; 195: 13–29.

Food and Agriculture Organization of the United Nations (FAO). Available: http://www.fao.org/faostat/en/#data/QC. Accessed on [2024-09-04].

Masuda T, Goldsmith P D. World soybean production: area harvested, yield, and long-term projections. International Food & Agribusiness Management Review, 2009; 12(4): 143–162.

Hilbert J A, Luciana S, Manosalva J, Patricio G, Ariana C. Greenhouse gas emission from the cultivation of soybean in Argentina. 2021. Available: https://www.researchgate.net/publication/350767239_Greenhouse_gas_emission_from_the_cultivation_of_soybean_in_Argentina. Accessed on [2024-09-04].

Zhang H Y, Kang J Z, Xu X, Zhang L P. Accessing the temporal and spectral features in crop type mapping using multi-temporal Sentinel-2 imagery: A case study of Yi’an County, Heilongjiang province, China. Computers and Electronics in Agriculture, 2020; 176: 105618.

She B, Yang Y Y, Zhao Z G, Huang L S, Liang D, Zhang D Y. Identification and mapping of soybean and maize crops based on Sentinel-2 data. Int J Agric & Biol Eng, 2020; 13(6): 171–182.

Wang L M, Liu J, Yang L B, Yang F G, Fu C H. Impact of short infrared wave band on identification accuracy of corn and soybean area. Transactions of the CSAE, 2016; 32(19): 169–178. (in Chinese)

Wang L M, Liu J, Yang L B, Yang F G, Fu C H. Application of random forest method in maize-soybean accurate identification. Acta Agronomica Sinica, 2018; 44(4): 569–580. (in Chinese)

Song X P, Potapov P V, Krylov A, King L A, Di Bella C M, Hudson A, et al. National-scale soybean mapping and area estimation in the United States using medium resolution satellite imagery and field survey. Remote Sensing of Environment, 2017; 190: 383–395.

Bocco M, Ovando G, Sayago S, Willington E, Heredia S. Estimating soybean ground cover from satellite images using neural-networks models. International Journal of Remote Sensing, 2012; 33(6): 1717–1728.

Willington E, Clemente J P, Bocco M. Determination of agricultural land use: incidence of atmospheric corrections and the implementation in multi-sensor and multi-temporal images. Revista de Teledetección, 2015; 44: 81–89.

Hu Q, Ma Y X, Xu B D, Song Q, Tang H J, Wu W B. Estimating sub-pixel soybean fraction from time-series MODIS data using an optimized geographically weighted regression model. Remote Sensing, 2018; 10(4): 491.

Wardlow B D, Egbert S L, Kastens J H. Analysis of time-series MODIS 250 m vegetation index data for crop classification in the U. S. Central Great Plains. Remote Sensing of Environment, 2007; 108(3): 290–310.

Zhong L H, Yu L, Li X C, Hu L N, Gong P. Rapid corn and soybean mapping in US Corn Belt and neighboring areas. Scientific Reports, 2016; 6: 36240.

Hu Q, Wu W B, Song Q, Lu M, Chen D, Yu Q Y, et al. How do temporal and spectral features matter in crop classification in Heilongjiang Province, China? Journal of Integrative Agriculture, 2017; 16(2): 324–336.

Ajadi O A, Barr J, Liang S Z, Ferreira R, Kumpatla S P, Patel R, et al. Large-scale crop type and crop area mapping across Brazil using synthetic aperture radar and optical imagery. International Journal of Applied Earth Observation and Geoinformation, 2021; 97: 102294.

Arvor D, Jonathan M, Penello Meirelles M S, Dubreuil V, Durieux L. Classification of MODIS EVI time series for crop mapping in the state of Mato Grosso, Brazil. International Journal of Remote Sensing, 2011; 32: 7847–7871.

Brown J C, Kastens J H, Coutinho A C, Victoria D D C, Bishop C R. Classifying multiyear agricultural land use data from Mato Grosso using time-series MODIS vegetation index data. Remote Sensing of Environment, 2013; 130: 39–50.

Zhong L H, Hu L N, Yu L, Gong P, Biging G. Automated mapping of soybean and corn using phenology. ISPRS Journal of Photogrammetry and Remote Sensing, 2016; 119: 151–164.

Kuchler P C, Bégué A, Simões M, Gaetano R, Arvor D, Ferraz R P D. Assessing the optimal preprocessing steps of MODIS time series to map cropping systems in Mato Grosso, Brazil. International Journal of Applied Earth Observation and Geoinformation, 2020; 92: 102150.

Chen Y L, Lu D S, Moran E, Batistella M, Dutra L V, Sanches I D, da Silva R F B, Huang J F, et al. Mapping croplands, cropping patterns, and crop types using MODIS time series data. Internatinal Journal of Applied Earth Observation and Geoinformation, 2018; 69: 133–147.

Picoli M C A, Camara G, Sanches I, Simões R, Carvalho A, Maciel A, et al. Big earth observation time series analysis for monitoring Brazilian agriculture. ISPRS Journal of Photogrammetry and Remote Sensing, 2018; 145(Part B): 328–339.

Formaggio A R, Vieira M A, Rennó C D. Object based image analysis (OBIA) and data mining (DM) in Landsat time series for mapping soybean in intensive agricultural regions. In: 2012 IEEE International Geoscience and Remote Sensing Symposium, Munich: IEEE, 2012; 2257-2260. doi: 10.1109/IGARSS.2012.6351047.

da Silva Junior C A, Leonel-Junior A H S, Rossi F S, Filho W L F C, de Barros Santiago D, de Oliveira-Júnior J F, et al. Mapping soybean planting area in midwest Brazil with remotely sensed images and phenology-based algorithm using the Google Earth Engine platform. Computers and Electronics in Agriculture, 2020; 169: 105194.

Paludo A, Becker W R, Richetti J, De Albuquerque Silva L C, Johann J A. Mapping summer soybean and corn with remote sensing on Google Earth Engine cloud computing in Parana state - Brazil. International Journal of Digital Earth, 2020; 13(12): 1624–1636.

Boryan C G, Yang Z W. Implementation of a new automatic stratification method using geospatial cropland data layers in NASS area frame construction. In: 2014 IEEE Geoscience & Remote Sensing Symposium, Quebec City: IEEE, 2014; pp.2110–2113. doi: 10.1109/IGARSS.2014.6946882.

Fisette T, Davidson A, Daneshfar B, Rollin P, Aly Z, Campbell L. Annual space-based crop inventory for Canada: 2009–2014. In: 2014 IEEE Geoscience & Remote Sensing Symposium, Quebec City: IEEE, 2014; 5095-5098. doi: 10.1109/IGARSS.2014.6947643.

Defourny P, Bontemps S, Bellemans N, Cara C, Dedieu G, Guzzonato E, et al. Near real-time agriculture monitoring at national scale at parcel resolution: Performance assessment of the Sen2-Agri automated system in various cropping systems around the world. Remote Sensing of Environment, 2019; 221: 551–568.

Liu X, X Yu L, Zhong L H, Hao P Y, Wu B, Wang H S, et al. Spatial-temporal patterns of features selected using Random Forests: a case study of corn and soybeans mapping in the US. International Journal of Remote Sensing, 2018; 40(1): 269–283.

Wang S, Di Tommaso S, Deines J M, Lobell D B. Mapping twenty years of corn and soybean across the US Midwest using the Landsat archive. Scientific Data, 2020; 7(1): 307.

De Abelleyra D, Veron S, Banchero S, Mosciaro M J, Propato T, Ferraina A, et al. First large extent and high resolution cropland and crop type map of Argentina. In: 2020 IEEE Latin American GRSS & ISPRS Remote Sensing Conference (LAGIRS), Santiago: IEEE, 2020; 392-396. doi: 10.1109/LAGIRS48042.2020.9165610.

Junior C A D S, Lima M, Johann J A, Rossi F S, de Oliveira-Júnior J F, Junior A S H L, et al. SojaMaps: project of monitoring of soybean areas in Brazil using big data in cloud computing. In: XIX Simpósio Brasileiro de Sensoriamento Remoto, 2019.

Immitzer M, Vuolo F, Atzberger C. First experience with Sentinel-2 data for crop and tree species classifications in Central Europe. Remote Sensing, 2016; 8(3): 166.

Veloso A, Mermoz S, Bouvet A, Le Toan T, Planells M, Dejoux J F, et al. Understanding the temporal behavior of crops using Sentinel-1 and Sentinel-2-like data for agricultural applications. Remote Sensing of Environment, 2017; 199: 415–426.

You N S, Dong J W, Huang J X, Du G M, Zhang G L, He Y L, et al. The 10-m crop type maps in Northeast China during 2017-2019. Scientific Data, 2021; 8: 41.

Luo C, Liu H J, Lu L P, Liu Z R, Kong F C, Zhang X L. Monthly composites from Sentinel-1 and Sentinel-2 images for regional major crop mapping with Google Earth Engine. Journal of Integrative Agriculture, 2021; 207(7): 1944–1957.

Vuolo F, Neuwirth M, Immitzer M, Atzberger C, Ng W T. How much does multi-temporal Sentinel-2 data improve crop type classification? International Journal of Applied Earth Observation and Geoinformation, 2018; 72: 122–130.

Luo C, Qi B S, Liu H J, Guo D, Lu L P, Fu Q, et al. Using time series Sentinel-1 images for object-oriented crop classification in Google Earth Engine. Remote Sensing, 2021; 13(4): 561.

Luo J S, Ma X W, Chu Q F, Xie M, Cao Y J. Characterizing the up-to-date land-use and land-cover change in Xiong’an New Area from 2017 to 2020 using the multi-temporal Sentinel-2 images on Google Earth Engine. ISPRS International Journal of Geo-Information, 2021; 10(7): 464.

Song Q, Hu Q, Zhou Q B, Hovis C, Xiang M T, Tang H J, et al. In-season crop mapping with GF-1/WFV data by combining object-based image analysis and Random Forest. Remote Sensing, 2017; 9(11): 1184.

Amani M, Kakooei M, Moghimi A, Ghorbanian A, Ranjgar B, Mahdavi S, et al. Application of Google Earth Engine cloud computing platform, Sentinel Imagery, and neural networks for crop mapping in Canada. Remote Sensing, 2020; 12(21): 3561.

Kumari M, Pandey V, Choudhary K K, Murthy C S. Object-based machine learning approach for soybean mapping using temporal sentinel-1/sentinel-2 data. Geocarto International, 2021; 37(23): 6848–6866.

Oldoni L V, Prudente V H R, Diniz J M F S, Wiederkehr N C, Sanches I D, Gama F F. Polarimetric Sar data from Sentinel-1a applied to early crop classification. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2020; XLIII-B3-2020: 1039–1046. doi: 10.5194/isprs-archives-XLIII-B3-2020-1039-2020.

Yin L K, You N S, Zhang G L, Huang J X, Dong J W. Optimizing feature selection of individual crop types for improved crop mapping. Remote Sensing, 2020; 12(1): 162.

Buenos Aires Grain Exchange. 2019. Available: http://www.bolsadecereales.com/ver-acerca-del-panorama-agricola-semanal-78#. Accessed on [2024-09-04].

Ministry of Agriculture, Livestock and Fisheries Presidency of the Nation, Argentina. Available: http://datosestimaciones.magyp.gob.ar/. Accessed on [2024-09-04].

Emelyanova I, Barron O, Alaibakhsh M. A comparative evaluation of arid inflow dependent vegetation maps derived from LANDSAT top-of-atmosphere and surface reflectances. International Journal of Remote Sensing, 2018; 39(20): 6607–6630.

Song C H, Woodcock C E, Seto K C, Lenney M P, Macomber S A. Classification and change detection using Landsat TM data: When and how to correct atmospheric effects. Remote Sensing of Environment, 2001; 75(2): 230–244.

Chen Y S, Hou J L, Huang C L, Zhang Y, Li X H. Mapping maize area in heterogeneous agricultural landscape with multi-temporal Sentinel-1 and Sentinel-2 images based on Random Forest. Remote Sensing, 2021; 13(15): 2988.

Hao P Y, Tang H J, Chen Z X, Liu Z J. Early-season crop mapping using improved artificial immune network (IAIN) and Sentinel data. PeerJ, 2018; 6: e5431.

You N S, Dong J W. Examining earliest identifiable timing of crops using all available Sentinel 1/2 imagery and Google Earth Engine. ISPRS Journal of Photogrammetry and Remote Sensing, 2020; 161: 109–123.

GlobeLand30: Global Geo-information Public Product, 2020. Available: https://www.webmap.cn/commres.do?method=globeIndex. Accessed on [2024-09-04].

National Institute of Agricultural Technology of Argentina. Available: http://www.geointa.inta.gob.ar/. Accessed on [2022-04-23].

Rumelhart D E, Hinton G E. Learning representations by back-propagating errors. Nature, 1986; 323: 533–536.

Hermes L, Frieauff D, Puzicha J, Buhmann J M. Support vector machines for land usage classification in Landsat TM imagery. In: IEEE 1999 International Geoscience and Remote Sensing Symposium (IGARSS’99), 1999; 1: 348–350.

Teluguntla P, Thenkabail P, Oliphant A, Xiong J, Gumma M K, Congalton R G, et al. A 30-m landsat-derived cropland extent product of Australia and China using Random Forest machine learning algorithm on Google Earth Engine cloud computing platform. ISPRS Journal of Photogrammetry and Remote Sensing, 2018; 144: 325–340.

Tatsumi K, Yamashiki Y, Torres M A C, Taipe C L R. Crop classification of upland fields using Random forest of time-series Landsat 7 ETM+ data. Computers and Electronics in Agriculture, 2015; 115: 171–179.

Rodriguez-Galiano V F, Ghimire B, Rogan J, Chica-Olmo M, Rigol-Sanchez J P. An assessment of the effectiveness of a Random Forest classifier for land-cover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 2012; 67: 93–104.

Pelletier C, Valero S, Inglada J, Champion N, Dedieu G. Assessing the robustness of Random Forests to map land cover with high-resolution satellite image time series over large areas. Remote Sensing of Environment, 2016; 187: 156–168.

Müller H, Rufin P, Griffiths P, Siqueira A J B, Hostert P. Mining dense Landsat time series for separating cropland and pasture in a heterogeneous Brazilian savanna landscape. Remote Sensing of Environment, 2015; 156: 490–499.

Dong J, Fu Y Y, Wang J J, Tian H F, Fu S, Niu Z, et al. Early-season mapping of winter wheat in China based on Landsat and Sentinel images. Earth System Science Data, 2020; 12(4): 3081–3095.

Fan D D, Li Q Z, Wang H Y, Zhang Y, Du X, Shen Y. Improvement in recognition accuracy of minority crops by resampling of imbalanced training datasets of remote sensing. National Remote Sensing Bulletin, 2019; 23(4): 730–742. (in Chinese)

Zhong L H, Peng G, Biging G S. Efficient corn and soybean mapping with temporal extendability: A multi-year experiment using Landsat imagery. Remote Sensing of Environment, 2014; 140: 1–13.

Cai Y P, Guan K Y, Peng J, Wang S W, Seifert C, Wardlow B, et al. A high-performance and in-season classification system of field-level crop types using time-series Landsat data and a machine learning approach. Remote Sensing of Environment, 2018; 210: 35–47.

Löw F, Michel U, Dech S, Conrad C. Impact of feature selection on the accuracy and spatial uncertainty of per-field crop classification using Support Vector Machines. ISPRS Journal of Photogrammetry and Remote Sensing, 2013; 85: 102–119.

Husak G J, Marshall M T, Michaelsen J, Pedreros D, Funk C, Galu G. Crop area estimation using high and medium resolution satellite imagery in areas with complex topography. Journal of Geophysical Research, 2008; 113(D14): D14112.




Copyright (c) 2024 International Journal of Agricultural and Biological Engineering

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

2023-2026 Copyright IJABE Editing and Publishing Office