Comparative Analysis of Machine Learning Techniques on EuroSAT RGB for Land Use Land Cover Classification

In a rapidly urbanizing world, significant environmental changes such as deforestation, climate change, and disasters are occurring swiftly, leading to critical land use and land cover (LULC) shifts. Although survey data is limited, satellite imagery provides extensive coverage for monitoring these changes. This paper evaluates popular machine learning algorithms - namely, random forest, support vector machines, multilayer perceptron, decision tree, and k-nearest neighbor - for LULC classification using the EuroSat RGB dataset from Sentinel-2 satellite imagery. Our findings highlight random forest as the most accurate and robust machine learning algorithm for this task. We have compared our results with those of bag-of-visual-words (BoVW) support vector machines, convolutional neural network. ResNet-50. and GoogleNeta