Classification of Land Cover in Sentinel-2 Imagery Using Machine Learning Models

Authors

Ehsan Ali Al-Zubaidi, Department of Environmental Planning, Faculty of Physical, Planning, University of Kufa, Najaf, IraqFollow
Mohammed Ridha Hammoodi, Computer Science Department, Faculty of Computer Science and Mathematics, University of Kufa, Najaf, Iraq. Email: mridha.hamoodi@uokufa.edu.iq
Ahmed Naser Alzurfi, Department of Electrical Engineering, Faculty of Engineering, University of Kufa, Najaf, Iraq

Abstract

Remote sensing data of medium resolution are commonly used to classify land cover, and machine learning (ML) models have taken on a central aspect in the necessary data analysis. Ordinarily, land cover is coded on a pixel basis on the basis of Digital Number (DN) values, which in turn are computed across several spectral bands. This paper is concerned with land cover mapping in Mosul, Iraq, based on satellite images captured by Sentinel-2. Two platforms featuring unsupervised classification algorithms were used, Google Earth Engine and ArcMap, making it possible to use K-means and X-means in Google Earth Engine and ISO Cluster in ArcMap. The comparative analysis of these algorithms allowed development of four significant land cover classes: built-up areas, vegetation, water bodies, and bare land. The results of the classification for the Mosul study area also indicated that using X-means gave an accuracy of 76.34, while using ISO Cluster gave 73.61, and K-means gave 80.14. The highest classification accuracy was achieved by the use of K-means, suggesting that it is the best algorithm to use in this region

Recommended Citation

Al-Zubaidi, Ehsan Ali; Hammoodi, Mohammed Ridha; and Alzurfi, Ahmed Naser (2026) "Classification of Land Cover in Sentinel-2 Imagery Using Machine Learning Models," Al-Bahir: Vol. 8: Iss. 2, Article 4.
Available at: https://doi.org/10.55810/2313-0083.1122

References

[1] Li A, Fan M, Qin G, Xu Y, Wang H. Comparative analysis of machine learning algorithms in automatic identification and extraction of water boundaries. Appl Sci 2021;11(21). https:// doi.org/10.3390/app112110062.

[2] Basheer S, Wang X, Farooque AA, Nawaz RA, Liu K, Adekanmbi T, et al. Comparison of land use land cover classifiers using different satellite imagery and machine learning techniques. Remote Sens 2022;14(19). https://doi. org/10.3390/rs14194978

. [3] Swetanisha S, Panda AR, Behera DK. Land use/land cover classification using machine learning models. Int J Electr Comput Eng 2022;12(2). https://doi.org/10.11591/ijece.v12i2. pp2040-2046.

[4] Athiyah U, Rustad S, Soeleman MA, Muljono, Akrom M. Effectiveness of K-means clustering-based outlier handling techniques in colonic condition endoscopy image classification using CNN. In: 2024 IEEE international conference on communication, networks and satellite (COMNETSAT); 2024. p. 276—82. https://doi.org/10.1109/comnetsat63286. 2024.10862507. Mataram, Indonesia.

[5] El-Omairi MA, El Garouani A. A review on advancements in lithological mapping utilizing machine learning algorithms and remote sensing data. Heliyon Sep 2023;9(9): e20168. https://doi.org/10.1016/j.heliyon.2023.e20168.

[6] Mantilla L. Classification of satellite images using Rp fuzzy c means for unsupervised classification algorithm. In: 2019 IEEE Colombian conference on applications in computational intelligence (ColCACI); 2019. p. 1—5. https://doi.org/ 10.1109/colcaci.2019.8781988. Barranquilla, Colombia.

[7] Tibshirani R, Walther G, Hastie T. K-Means clustering and related algorithms. Technical report. Stanford University; 2004. https://doi.org/10.17760/d20291355.

[8] Li XC, Liu LL, Huang LK. Comparison of several remote sensing image classification methods based on Envi. Int Arch Photogram Rem Sens Spatial Inf Sci 2020;XLII-3/W10: 605—11. https://doi.org/10.5194/isprs-archives-xlii-3-w10- 605-2020.

[9] Dibs H, Hasab HA, Al-Rifaie JK, Al-Ansari N. An optimal approach for land-use/land-cover mapping by integration and fusion of multispectral landsat oli images: case study in Baghdad, Iraq. Water, Air, Soil Pollut 2020;231(9). https:// doi.org/10.1007/s11270-020-04846-x.

[10] Masini N, Lasaponara R. Recent and past archaeological looting by satellite remote sensing: approach and application in Syria. In: Remote sensing for archaeology and cultural landscapes. Springer Remote Sensing/ Photogrammetry; 2020. p. 123—37. https://doi.org/10.1007/ 978-3-030-10979-0_8. ch. Chapter 8.

[11] Amal Muhammad Saleh FWA. Analyzing land use/land cover change using remote sensing and GIS in Mosul District, Iraq. Annals Romanian Soc Cell Biol 2021;25(3): 4342—59. https://doi.org/10.21275/v5i2.nov161379. 182 AL-BAHIR (JOURNAL FOR ENGINEERING AND PURE SCIENCES) 2026;8:169—183

[12] Jain T, Khan MJ, Pandey S, Mishra S, Yadav S. Satellite image classification and analysis using machine learning with ISRO LISS IV. https://doi.org/10.26706/ijceae.2.2. 20210404; 2022.

[13] Anul Haq M. Planetscope nanosatellites image classification using machine learning. Comput Syst Sci Eng 2022;42(3): 1031—46. https://doi.org/10.32604/csse.2022.023221.

[14] Ouchra H, Belangour A, Erraissi A. Comparison of machine learning methods for satellite image classification: a case study of Casablanca using landsat imagery and google earth engine. J Environ Earth Sci 2023;5(2):118—34. https://doi.org/ 10.30564/jees.v5i2.5928.

[15] Han S, Lee J. Parallelized inter-image k-means clustering algorithm for unsupervised classification of series of satellite images. Remote Sens 2023;16(1). https://doi.org/10.3390/ rs16010102.

[16] Hassan EK, Saeed HM, Al-Ghrairi AHT. Classification and measurement of land cover of wildfires in Australia using remote sensing. Iraqi J Sci 2022:420—30. https://doi.org/10. 24996/ijs.2022.63.1.38.

[17] Mohammed MG. Land use land cover changes detection of Erbil City using GIS and remote sensing. Polytechnic Journal 2023;13(1). https://doi.org/10.59341/2707-7799.1729.

[18] Khazaal ST, Hussan WH, AL-Shammari MH. Land use/ Land cover assessment for Karbala City by using geographic information systems (GIS) technique. https:// doi.org/10.65115/7zq3sp06; 2024.

[19] Chomani K, Pshdari S. Evaluation of different classification algorithms for land use land cover mapping. Kurdistan Journal of Applied Research 2024;9(2):13—22. https://doi. org/10.55277/researchhub.2vexo15g.1.

[20] Memon AV, Shah NV, Patel YS, Parangi T. Enhancing land use/land cover analysis with sentinel-2 bands: comparative evaluation of classification algorithms and dimensionality reduction for improved accuracy assessment. Nat Environ Pollut Technol 2025;24(2). https://doi.org/10.46488/nept. 2025.v24i02.b4264.

[21] Saad SAK, Al-Zubaidi EA. Classification of satellite Sentinel-2 imagery using unsupervised methods: a case study of Erbil. In: AIP Conf Proc, 3318. AIP Publishing LLC; 2025. p. 030028. https://doi.org/10.1063/5.0286607. 1.

[22] Sharma M, Kumar CJ, Deka A. Land cover classification: a comparative analysis of clustering techniques using Sentinel-2 data. Int J Sustain Agric Manag Inform 2021;7(4): 321—42. https://doi.org/10.1504/ijsami.2021.122008.

[23] Rahman A, Abdullah HM, Tanzir MT, Hossain MJ, Khan BM, Miah MG, et al. Performance of different machine learning algorithms on satellite image classification in rural and urban setup. Remote Sens Appl: Society and Environment 2020;20. https://doi.org/10.1016/j.rsase.2020. 100410.

[24] Zhang P, Hu X, Ban Y, Nascetti A, Gong M. Assessing Sentinel-2, Sentinel-1, and ALOS-2 PALSAR-2 data for large-scale wildfire-burned area mapping: insights from the 2017—2019 Canada Wildfires. Remote Sens 2024;16(3). https://doi.org/10.3390/rs16030556.

[25] Marlina D. Land cover classification in sentinel-2 image Kuningan District with NDVI and algorithm random forest. https://doi.org/10.30998/string.v7i1.12948; 2022.

[26] Al-Zubaidi EA, Al-Sulttani AH, Rabee F. Sand dunes spectral index determination using machine learning model: case study of Baiji Sand Dunes Field Northern Iraq. Iraqi Geological Journal 2022;55(1F):102—21. https://doi.org/ 10.46717/igj.55.1f.9ms-2022-06-24.

[27] MacQueen J. Some methods for classification and analysis of multivariate observations. In: Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, Volume 1: Statistics, 5. University of California press; 1967. p. 281—98. https://doi.org/10.1002/0471271357. ch9.

[28] Pelleg D, Moore A. X-means: extending K-means with efficient estimation of the number of clusters. In: ICML’00; 2000. p. 727—34. https://doi.org/10.7717/peerj-cs.2516/fig-1. Citeseer.

[29] Congalton RG. Accuracy assessment and validation of remotely sensed and other spatial information. Int J Wildland Fire 2001;10(4). https://doi.org/10.1071/wf01031.

[30] Congalton RG, Green K. Assessing the accuracy of remotely sensed data: principles and practices. CRC press; 2019. https://doi.org/10.1201/9781420055139-12.

[31] Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas 1960;20(1):37—46. https://doi.org/10. 1177/00131644600200010.

[32] Foody GM. Status of land cover classification accuracy assessment. Rem Sens Environ 2002;80(1):185—201. https:// doi.org/10.1016/s0034-4257(01)00295-4.

[33] Sokolova M, Lapalme G. A systematic analysis of performance measures for classification tasks. Inf Process Manag 2009;45(4):427—37. https://doi.org/10.1016/j.ipm.2009.03. 002.