• Aasish
• Abhinav
• Adwaith
• Daya Naik
• Nivin K S

Land use and land cover (LULC) classification is important to monitor environmental changes and efficiently manage natural resources. This study uses a deep learning approach with the ResNet-50 architecture, pre-trained on the ImageNet database, to classify satellite images of the EuroSAT dataset into ten distinct land cover categories. The classification is based entirely on RGB bands, with the performance of the model enhanced further through data augmentation methods. With the trained model, high accuracy is obtained, with the ability to distinguish between land cover classes like forests, urban settlements, and agricultural lands. These outcomes show the tremendous potential of transfer learning in remote sensing and satellite image analysis.

Land Cover Classification, ResNet-50 Architecture, EuroSAT Dataset, Deep Learning Techniques, Remote Sensing Applications.

"Satellite-Based Land Cover Mapping Through Fine-Tuned ResNet-50 Architecture", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 7, page no.729-732, July-2025, Available :http://www.jetir.org/papers/JETIRGX06136.pdf

"Satellite-Based Land Cover Mapping Through Fine-Tuned ResNet-50 Architecture", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 7, page no. pp729-732, July-2025, Available at : http://www.jetir.org/papers/JETIRGX06136.pdf