Comparative Performance of Random Forest and Support Vector Machine for Land use and Land Cover Mapping Using Sentinel-2 Imagery in Gabès, Tunisia (2024)
Domaine:
geospatialType de record:
paper
Créateur:
BenBenYahAli
Éditeur:
NatAme
Éditeur:
CCSD
Hôte:
International audience
Land Use and Land Cover (LULC) mapping is vital for managing natural resources in arid regions, which are highly vulnerable to human pressures and climate change. This study compares two machine learning classifiers—Random Forest (RF) and Support Vector Machine (SVM)—applied to Sentinel-2 imagery (March–September 2024) over the delegations of Gabès Medina, Gabès West, and Gabès South (Tunisia), using Google Earth Engine (GEE). Four land cover classes were analyzed: urban areas, moderate vegetation, sparse vegetation, and bare soil. The integration of spectral indices—Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-up Index (NDBI), Normalized Difference Water Index (NDWI), and Bare Soil Index (BSI)—contributed to a notable improvement in classification performance. RF achieved higher accuracy (91.2%, Kappa 0.883) than SVM (84.6%, Kappa 0.794), confirming its robustness in heterogeneous landscapes. These results highlight the potential of machine learning for detailed LULC characterization and offer a decision-support tool for sustainable land management.
Similaires
Land Use and Land Cover Classification in Google Earth Engine Using Sentinel-2 Based Random Forest: A Case Study of Katsina State, NigeriaDataSheet1_A performance evaluation of random forest, artificial neural network, and support vector machine learning algorithms to predict spatio-temporal land use-land cover dynamics: a case from lusaka and colombo.docxShort-term forecasting of land use and land cover changes in Senegal's great green wall using deep learning and Sentinel-2 imageryMachine Learning-Based Land Use and Land Cover Mapping Using Multi-Spectral Satellite Imagery: A Case Study in EgyptPresentation 1_Enhancing land cover classification in the heterogeneous landscape by integrating auxiliary data with Sentinel-2 imagery using the random forest algorithm.pdfComparative Evaluation of Supervised and Unsupervised Methods for Multitemporal Land-Use/Land-Cover Mapping in Lagos, Nigeria (2000–2024)