IJEP 45(9): 775-791 : Vol. 45 Issue. 9 (September 2025)
Debayan Bhattacharya1, Koj Sambyo1*, Chandrima Bandyopadhyay2 and Gupinath Bhandari3
1. National Institute of Technology, Department of Computer Science and Engineering, Jote – 791 113, Arunachal Pradesh, India
2. Jadavpur University, School of Water Resources Engineering, Kolkata – 700 032, West Bengal, India
3. Jadavpur University, Department of Civil Engineering, Kolkata – 700 032, West Bengal, India
Abstract
The study aims to develop a comprehensive scheme for landuse and land cover (LULC) classification in community development block of Ramnagar I, located in the south-western coastal plain of West Bengal, India. The study area was divided into 9 Gram Panchayats (GP) and classified GP-wise into 10 classes using Sentinel-2A MSI data for 2019 and 2023. Three machine learning algorithms, support vector machine (SVM), classification and regression trees (CART) and random forest (RF), were used to determine the most effective algorithm. The recursive feature elimination with random forest (RFE-RF) method was implemented to identify the optimal features from a pool of 30 features, including 12 spectral bands, 6 quality assurance (QA) bands, 3 true colour image (TCI) bands, 2 masking bands and 7 spectral indices. A total of 18 LULC products were generated, consisting of 9 maps of 9 GPs each for the years 2019 and 2023. The RF algorithm had the highest level of effectiveness, with a mean overall accuracy (OA) of 78.45% in 2019 and 80.30% in 2023. The results also showed that spectral bands and indices are crucial for LULC classification, with modified normalized difference water index (MNDWI) consistently identified as the most significant band. Despite some instances of misclassifications among LULC classes, this study provides valuable spatio-temporal LULC datasets for regional planners, policymakers and researchers to monitor and analyze temporal and spatial changes.
Keywords
Landuse and land cover, Machine learning, Google Earth Engine, Sentinel-2A, Feature selection
References
- Lambin, E. F. and P. Meyfroidt. 2010. Landuse transitions: Socio-ecological feedback versus socio-economic change. Land Use Policy. 27(2): 108–118. DOI: 10.1016/j.landusepol.2009.09.003.
- Gogoi, P. P., et al. 2019. Landuse and land cover change effect on surface temperature over eastern India. Sci. Rep., 9(1): 1–10. DOI: 10.1038/s4 1598-019-45213-z.
- Amini, S., et al. 2022. Urban landuse and land cover change analysis using random forest classification of Landsat time series. Remote Sensing. 14(11): 1–23. DOI: 10.3390/rs14112654.
- Shafizadeh-Moghadam, H., et al. 2021. Google Earth engine for large-scale landuse and land cover mapping: An object-based classification approach using spectral, textural and topographical factors. GISci. Remote Sensing. 58(6): 914–928. DOI: 10.1 080/15481603.2021.1947623.
- Jones, K. R., et al. 2018. One-third of global protected land is under intense human pressure. Sci., 360(6390): 788–791. DOI: 10.1126/science.aap 9565.
- Feng, S., et al. 2022. Landuse/land cover mapping based on GEE for the monitoring of changes in ecosystem types in the Upper Yellow river basin over the Tibetan plateau. Remote Sensing. 14(21): 1–25. DOI: 10.3390/rs14215361.
- González-González, A., N. Clerici and B. Quesada. 2022. A 30 m resolution landuse-land cover product for the Colombian Andes and Amazon using cloud-computing. Int. J. Appl. Earth Observation Geoinf., 107(1):102688. DOI: 10.1016/j.jag.2022. 102688.
- Chen, J., et al. 2015. Global land cover mapping at 30 m resolution: A POK-based operational appro-ach. ISPRS J. Photogrammetry Remote Sensing. 103: 7–27. DOI: 10.1016/j.isprsjprs.2014.09.002.
- Meng, B., et al. 2022. Mapping grassland classes using unmanned aerial vehicle and MODIS NDVI data for temperate grassland in inner Mongolia, China. Remote Sensing. 14(9): 2094. DOI: 10.3390/rs140 92094.
- Gorelick, N., et al. 2017. Google Earth engine: Planetary-scale geospatial analysis for everyone. Remote Sensing Env., 202: 18–27. DOI: 10.1016/j. rse.2017.06.031.
- Tassi, A. and M. Vizzari. 2020. Object-oriented LULC classification in Google Earth engine combining SNIC, GLCM and machine learning algorithms. Remote Sensing. 12(22): 1–17. DOI: 10.3390/rs12 223776.
- Abburu, S. and S.B. Golla. 2015. Satellite image classification methods and techniques: A review. Int. J. Computer Appl., 119(8): 20–25. DOI: 10.51 20/21088-3779.
- Ouchra, H., A. Belangour and A. Erraissi. 2023. Machine learning algorithms for satellite image classification using Google Earth engine and Landsat satellite data: Morocco case study. IEEE Access. 11(7): 71127–71142. DOI: 10.1109/ACCESS.20 23.3293828.
- Ouchra, H., A. Belangour and A. Erraissi. 2022. Spatial data mining technology for GIS: A review. 2022 International Conference on Data analytics for business and industry (ICDABI 2022). Proceedings, pp 655–659. DOI: 10.1109/ICDABI56818. 2022.10041574.
- Whiteside, T.G., G.S. Boggs and S.W. Maier. 2011. Comparing object-based and pixel-based classifications for mapping Savannas. Int. J. Appl. Earth Observation Geoinf., 13(6): 884–893. DOI: 10.10 16/j.jag.2011.06.008.
- Ouchra, H., A. Belangour and A. Erraissi. 2022. A comparative study on pixel-based classification and object-oriented classification of satellite image. Int. J. Eng. Trends Tech., 70(8): 206–215. DOI: 10.144 45/22315381/IJETT-V70I8P221.
- Zhao, Z., et al. 2024. Comparison of three machine learning algorithms using Google Earth engine for landuse land cover classification. Rangel Ecol. Manage., 92: 129–137. DOI: 10.1016/j.rama.2023. 10.007.
- Hu, Y., Y. Dong and Batunacun. 2018. An automatic approach for land-change detection and land updates based on integrated NDVI timing analysis and the CVAPS method with GEE support. ISPRS J. Photogrammetry Remote Sensing. 146(12): 347–359. DOI: 10.1016/j.isprsjprs.2018.10.008.
- Magidi, J., et al. 2021. Application of the random forest classifier to map irrigated areas using Google Earth engine. Remote Sensing. 13(5): 1–15. DOI: 10.3390/rs13050876.
- Dhingra, S. and D. Kumar. 2019. A review of remo-tely sensed satellite image classification. Int. J. Electrical Computer Eng., 9(3): 1720–1731. DOI: 10.11591/ijece.v9i3.pp.1720-1731.
- Usama, M., et al. 2019. Unsupervised machine learning for networking: Techniques, applications and research challenges. IEEE Access. 7: 65579–65615. DOI: 10.1109/ACCESS.2019.2916648.
- Chen, G. and J. Chen. 2015. A wrapper method for features election and its applications. Neuro-computing. 159: 219–226. DOI: 10.1016/j.neu com.2015.01.070.
- Amani, M., et al. 2019. Canadian wetland inventory using Google Earth engine: The first map and preliminary results. Remote Sensing. 11(7): 1–20. DOI: 10.3390/RS11070842.
- Pham-Duc, B., et al. 2023. Trends and applications of Google Earth engine in remote sensing and earth science research: A bibliometric analysis using scopus database. Earth Sci. Inf., 16(3): 2355–2371. DOI: 10.1007/s12145-023-01035-2.
- Kumar, L. and O. Mutanga. 2018. Google Earth engine applications since inception: Usage, trends and potential. Remote Sensing. 10(10): 1–15. DOI: 10.3390/rs10101509.
- Mondal, M., et al. 2012. Land people – A dynamic interaction of Purba Medinipur district, West Bengal. IOSR J. Pharmacy. 2(6): 56–61. DOI: 10.9790/3013-26405661.
- Das, R. and P. Dandapath. 2014. Existence and experience of Purba Medinipur coastal belt on its morpho-dynamic journey with the distinctive geology and geomorphology. Int. J. Sci. Res., 3(6): 1242–1251.
- Hossain, S.A., et al. 2022. Coastal vulnerability assessment of India’s Purba Medinipur-Balasore coas-tal stretch: A comparative study using empirical models. Int. J. Disaster Risk Red., 77(5): 103065. DOI: 10.1016/j.ijdrr.2022.103065.
- Sahoo, B. and P.K. Bhaskaran. 2016. Assessment on historical cyclone tracks in the Bay of Bengal, east coast of India. Int. J. Climatol., 36(1): 95–109. DOI: 10.1002/joc.4331.
- Bera, A. 2015. Impact of tourism on coastal ecology in the coastal region of Digha (West Bengal). Int. J. Sci. Res., 4(6): 2034–2039.
- Gascon, F., et al. 2017. Copernicus Sentinel-2A calibration and products validation status. Remote Sensing. 9(6): 584. DOI: 10.3390/rs9060584.
- Obregón, M.Á., et al. Validation of ESA Sentinel-2 L2A aerosol optical thickness and columnar water vapour during 2017-2018. Remote Sensing. 11(14): 1649. DOI: 10.3390/rs11141649.
- Sharma, V. 2022. A study on data scaling methods for machine learning. Int. J. Global Academic Sci. Res., 1(1): 31-42. DOI: 10.55938/ijgasr.v1i1.4.
- Yang, Y., et al. 2021. Testing accuracy of land cover classification algorithms in the Qilian mountains based on GEE cloud platform. Remote Sensing. 13(24): 5064. DOI: 10.3390/rs13245064.
- Talukdar, S., et al. 2020. Landuse land-cover classification by machine learning classifiers for satellite observations—A review. Remote Sensing. 12: 1135.
- Loukika, K.N., V.R. Keesara and V. Sridhar. 2021. Analysis of landuse and land cover using machine learning algorithms on Google Earth engine for Munneru river basin, India. Sustain., 13(24): 13758. DOI: 10.3390/su132413758.
- Hsu, C.W., C.C. Chang and C.J. Lin. 2008. A practical guide to support vector classification. BJU Int., 101(1): 1396–1400.
- Breiman, C. J., et al. 1984. Classification and regression trees (1st edn). Chapman and Hall/CRC. DOI: 10.1201/9781315139470.
- Jin, Z., et al. 2020. RFRSF: Employee turnover prediction based on random forests and survival analysis. In Web information systems engineering (WISE 2020) – 21st International Conference. Amsterdam, The Netherlands. Proceedings, 12343: 503–515. DOI: 10.1007/978-3-030-62008-0_35.
- Belgiu, M. and L. Drãgu. 2016. Random forest in remote sensing: A review of applications and future directions. ISPRS J. Photogrammetry Remote Sensing. 114: 24–31. DOI: 10.1016/j.isprsjprs.20 16.01.011.
- Kohavi, R. and G.H. John. 1997. Wrappers for feature subset selection. Artificial Intelligence. 97: 273–324.
- Rodin, A.S., et al. 2009. Use of wrapper algorithms coupled with a random forests classifier for variable selection in large-scale genomic association studies. J. Computational Biol.: J. Computational Molecular Cell Biol., 16(12): 1705–1718. DOI: 10.1089/cmb.2008.0037.
- Venkatesh, B. and J. Anuradha. 2019. A review of feature selection and its methods. Cybernetics Inf. Tech., 19(1): 3–26. DOI: 10.2478/CAIT-2019-0 001.
- Mostafa, E., et al. 2021. Monitoring and forecasting of urban expansion using machine learning-based techniques and remotely sensed data: A case study of Gharbia Governorate, Egypt. Remote Sensing. 13(22): 4498. DOI: 10.3390/rs13224498.
- Olofsson, P., et al. 2013. Making better use of accu-racy data in land change studies: Estimating accuracy and area and quantifying uncertainty using stratified estimation. Remote Sensing Env., 129: 122–131. DOI: 10.1016/j.rse.2012.10.031.
- Nasiri, V., et al. 2022. Landuse and land cover mapping using Sentinel-2 , Landsat-8 and Google Earth engine: A comparison of two composition methods. Remote Sensing. 14(9): 1977.
- Rwanga, S.S. and J.M. Ndambuki. 2017. Accuracy assessment of landuse/land cover classification using remote sensing and GIS. Int. J. Geosci., 8(4): 611–622. DOI: 10.4236/ijg.2017.84033.
- Bosin, A., N. Dessì and B. Pes. 2007. Capturing heuristics and intelligent methods for improving micro-array data classification. In Intelligent data engineering and automated learning (IDEAL 2007): 8th International Conference, Birmingham, United Kingdom. Proceedings, 4881: 790–799. DOI: 10.10 07/978-3-540-77226-2_79.
- Sánchez-Maroño, N., A. Alonso-Betanzos and R.M. Calvo-Estévez. 2009. A wrapper method for feature selection in multiple classes datasets. In Bio-inspired systems: Computational and ambient intelligence. 10th International Work-Conference on artificial neural networks (IWANN 2009), Salam-anca, Spain. Proceedings, 5517(1): 456–463. DOI: 10.1007/978-3-642-02478-8_57.
- Kursa, M.B. and W.R. Rudnicki. 2011. The all relevant feature selection using random forest. arXiv:1106.5112. Cornell University.
- Loi, D.T., T.Y. Chou and Y.M. Fang. 2017. Integration of GIS and remote sensing for evaluating forest canopy density index in Thai Nguyen province, Vietnam. Int. J. Env. Sci. Develop., 8(8): 539–542. DOI: 10.18178/ijesd.2017.8.8.1012.
- Kafy, A.A., et al. 2020. Prediction of future land surface temperature and its impact on climate change: A remote sensing based approach in Chattogram. 1st International Student Research Conference -2020. Dhaka University Research Society (DURS), Dhaka, Bangladesh. Proceedings, pp 1–11.
- Laonamsai, J., et al. 2023. Utilizing NDWI, MNDWI, SAVI, WRI and AWEI for estimating erosion and deposition in Ping river in Thailand. Hydrol., 10(3): 1–25. DOI: 10.3390/hydrology10030070.
- Du, Y., et al. 2016. Waterbodies’ mapping from Sentinel-2 imagery with modified normalized difference water index at 10 m spatial resolution produced by sharpening the SWIR band. Remote Sensing. 8(4): 354. DOI: 10.3390/rs8040354.
- Zhang, C., et al. 2022. Classification of aquaculture waters through remote sensing on the basis of a time-series water index. J. Coastal Res., 38(6): 1148-1162. DOI:10.2112/jcoastres-d-22-00009.1.