Assessment of machine learning on sugarcane classification using Landsat-8 and Sentinel-2 satellite imagery

Main Article Content

Teerapat Butkhot
Pipat Reungsang


Agriculture and agricultural product development are important aspects of a country's economic development. Sugarcane is one of the key industrial crops in Thailand, Brazil, China, and India. Therefore, monitoring sugarcane growth and harvest is important for evaluating yield, optimizing logistic operations, and forecasting crop productivity. To monitor sugarcane growth more effectively and efficiently, this study aimed to classify the sugarcane cultivation regions in Chuenchom District, Maha Sarakham Province, Thailand, using Landsat-8 and Sentinel-2 satellite images. To this end, three algorithms were used for classification: support vector machine (SVM), random forest (RF), and maximum likelihood (ML). A combination of parameter sets using four bands (red, green, blue, and NIR) and two vegetation indices: normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) was set up for the classification. The overall accuracy and kappa coefficient values were computed to validate the classification results with visual interpretation of high-resolution images. Results from the study showed that RF outperformed the SVM and ML classification techniques with overall accuracy and kappa coefficient values of 75.93 and 0.616, respectively, for Landsat-8 images and 78.60 and 0.656, respectively, for Sentinel-2 images. Specifically, RF classification with red, green, blue, and NIR provided the highest accuracy for the Landsat-8 images, while RF classification with red, green, blue, and NDVI proved to be the most accurate for the Sentinel-2 images. In summary, both Landsat-8 and Sentinel-2 satellite images have great potential for sugarcane mapping using remote sensing.


Download data is not yet available.

Article Details

Research Articles


[1] OECD/FAO. OECD-FAO agricultural outlook 2019-2028. Rome: OECD Publishing, Paris/Food and Agriculture Organization of the United Nations; 2019.
[2] Sawaengsak W, Prasara AJ, Gheewala SH. Assessing the socio-economic sustainability of sugarcane harvesting in Thailand. Sugar Tech. 2021;23(2):263-277.
[3] Oré G, Alcântara M, Góes J, Oliveira L, Yepes J, Teruel B, et al. (2020). Crop growth monitoring with drone-borne DInSAR. Remote Sens. 2020;12:615.
[4] Luna I, Lobo A. Mapping crop planting quality in sugarcane from UAV imagery: a pilot study in Nicaragua. Remote Sens. 2016;8:500.
[5] Som-ard J, Hossain MD, Ninsawat S, Veerachitt V. Pre-harvest sugarcane yield estimation using UAV-based RGB images and ground observation. Sugar Tech. 2018;20(6):645-657.
[6] Chea C, Saengprachatanarug K, Posom J, Wongphati M, Taira E. Sugar yield parameters and fiber prediction in sugarcane fields using a multispectral camera mounted on a small unmanned aerial system (UAS). Sugar Tech. 2020;22(4):605-621.
[7] Avci ZDU, Sunar F. Process-based image analysis for agricultural mapping: a case study in Turkgeldi region, Turkey. Adv Space Res. 2015;56:1635-1644.
[8] Foody GM. Status of land cover classification accuracy assessment. Remote Sens Environ. 2002;80:185-201.
[9] Gao F, Anderson MC, Zhang XY, Yang ZW, Alfieri JG, Kustas WP, et al. Toward mapping crop progress at field scales through fusion of Landsat and MODIS imagery. Remote Sens Environ. 2017;188:9-25.
[10] Korhonen L, Packalen P, Rautiainen M. Remote sensing of environment comparison of Sentinel-2 and Landsat-8 in the estimation of boreal forest canopy cover and leaf area index. Remote Sens Environ. 2017;195:259-274.
[11] Dos Santos Luciano AC, Picoli MCA, Rocha JV, Franco HCJ, Sanches GM, Leal MRLV, et al. Generalized space-time classifiers for monitoring sugarcane areas in Brazil. Remote Sens Environ. 2018;215:438-451.
[12] Kavats O, Khramov D, Sergieieva K, Vasyliev V. Monitoring of sugarcane harvest in Brazil based on optical and SAR data. Remote Sens. 2020;12:4080.
[13] Jiang H, Li D, Jing W, Xu J, Huang J, Yang J, et al. Early season mapping of sugarcane by applying machine learning algorithms to Sentinel-1A/2 time series data: a case study in Zhanjiang City, China. Remote Sens. 2019;11:861.
[14] Whyte A, Ferentinos KP, Petropoulos GP. A new synergistic approach for monitoring wetlands using Sentinels -1 and 2 data with object-based machine learning algorithms. Environ Model Softw. 2018;104:40-54.
[15] Rodriguez-galiano VF, Ghimire B, Rogan J, Chica-olmo M, Rigol-sanchez JP. An assessment of the effectiveness of a random forest classifier for land-cover classification. ISPRS J Photogramm Remote Sens. 2012;67:93-104.
[16] Li M, Ma L, Blaschke T, Cheng L, Tiede D. A systematic comparison of different object-based classification techniques using high spatial resolution imagery in agricultural environments. Int J Appl Earth Obs Geoinf. 2016;49:87-98.
[17] Hammer RG, Sentelhas PC, Mariano JCQ. Sugarcane yield prediction through data mining and crop simulation models. Sugar Tech. 2020;22(2):216-225.
[18] Maldaner LF, Corredo LP, Canata TF, Molin JP. Predicting the sugarcane yield in real-time by harvester engine parameters and machine learning approaches. Comput Electron Agric. 2021;181:105945.
[19] Luciano ACS, Picoli MCA, Duft DG, Rocha JV, Leal MRLV, Maire G. Empirical model for forecasting sugarcane yield on a local scale in Brazil using Landsat imagery and random forest algorithm. Comput Electron Agric. 2021;184:106063.
[20] Wanga M, Liub Z, Baig MHA, Wang Y, Lib Y, Chen Y. Mapping sugarcane in complex landscapes by integrating multi-temporal Sentinel-2 images and machine learning algorithms. Land use policy. 2019;88: 104190.
[21] Agricultural Meteorology to know for Maha Sarakham [Internet]. 2014 [cited 2020 Mar 12]. Available from: /Agromettoknow/NE/.
[22] Landuse Northeast_GISTDA_25k [Internet]. 2017 [cited 2020 Mar 15]. Available from:
[23] Lawrence RL, Ripple WJ. Comparisons among vegetation indices and bandwise regression in a highly disturbed, heterogeneous landscape. Remote Sens Environ. 1998;64:91-102.
[24] Chen D, Huang J, Jackson TJ. Vegetation water content estimation for corn and soybeans using spectral indices derived from MODIS near –and short wave infrared bands. Remote Sens Environ. 2005;98:225-236.
[25] Gu Y, Brown WJ, Verdin JP, Wardlow B. A five-year analysis of MODIS NDVI and NDWI for grassland drought assessment over the central Great Plains of the United States. Geophys Res Lett. 2007;34:L06407.
[26] Bégué A, Lebourgeois V, Bappel E, Todoroff P, Pellegrino A, Baillarin F, et al. Spatio-temporal variability of sugarcane fields and recommendations for yield forecast using NDVI. Remote Sens. 2010;31:5391-5407.
[27] Fernandes JL, Ebecken NFF, Esquerdo JCDN. Sugarcane yield prediction in Brazil using NDVI time series and neural networks ensemble. Remote Sens. 2017;38:4631-4644.
[28] Huete A, Didan K, Miura T, Rodriguez EP, Gao X, Ferreira LG. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sens Environ. 2002;83:195-213.
[29] Matsushita B, Yang W, Chen J, Onda Y, Qiu G. Sensitivity of the enhanced vegetation index (EVI) and normalized difference vegetation index (NDVI) to topographic effects: a case study in high-density cypress forest. Sensors. 2007;7:2636-2651.
[30] Foody GM, Mathur A. Toward intelligent training of supervised image classifications: directing training data acquisition for SVM classification. Remote Sens Environ. 2004;93:107-117.
[31] Pal M. Random forest classifier for remote sensing classification. Int J Remote Sens. 2005;26:217-222.
[32] Sonobe R, Yamaya Y, Tani H, Wang X, Kobayashi N, Mochizuki K. Assessing the suitability of data from Sentinel-1A and 2A for crop classification. GIScience Remote Sens. 2017;54:918-938.
[33] Hawryło P, Bednarz B, Wężyk P, Szostak M. Estimating defoliation of Scots pine stands using machine learning methods and vegetation indices of Sentinel-2. Eur J Remote Sens. 2018;51:194-204.
[34] Waldner F, Canto GS, Defourny P. Automated annual cropland mapping using knowledge-based temporal features. ISPRS J Photogramm Remote Sens. 2015;110:1-13.
[35] Mountrakis G, Im J, Ogole C. Support vector machines in remote sensing: a review. ISPRS J Photogramm Remote Sens. 2011;66:247-259.
[36] Vapnik V. The nature of statistical learning theory. New York: Springer; 1995.
[37] Maxwell AE, Warner TA, Fang F, Maxwell AE, Warner TA. Implementation of machine-learning classification in remote sensing: an applied review. Int J Remote Sens. 2018;39:2784-817.
[38] Kavzoglu T, Colkesen I. A kernel function analysis for support vector machines for land cover classification. Int J Appl Earth Obs Geoinf. 2009;11(5):352-359.
[39] Breiman L. Random forests. Mach. Learn. 2001;45(1):5-32.
[40] Yin H, Dirk P, Li A, Li Z, Hostert P. Land use and land cover change in Inner Mongolia - understanding the effects of China’s revegetation programs. Remote Sens Environ. 2018;204:918-930.
[41] Chan JCW, Paelinckx D. Evaluation of random forest and adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery. Remote Sens Environ. 2008;112(6):2999-3011.
[42] Liaw A, Wiener M. Classification and regression by random forest. R news. 2002;2:18-22.
[43] Ishwaran H, Kogalur UB, Blackstone EH, Lauer MS. Random survival forests. Ann Appl Stat. 2008;2:841-860.
[44] Ishwaran H, Kogalur UB. Random survival forests for R. R News. 2007;7(2):25-31.
[45] Richards JA. Remote sensing digital image analysis: an introduction. Berlin: Springer; 1999.
[46] Congalton RG. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens Environ. 1991;37:35-46.
[47] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-174.
[48] Saini R, Ghosh SK. Crop classification on single date Sentinel-2 imagery using random forest and support vector machine. Int arch photogramm remote sens spat inf sci. 2018;XLII-5:683-688.