eurosat

Classifying custom image datasets by creating Convolutional Neural Networks and Residual Networks from scratch with PyTorch

This is the ResNet50 implementation of the Eurosat dataset.

Satellite images classification

ANN to SNN conversion on land cover and land use classification problem for increased energy efficiency.

A fast and easy-to-use Remote sensing Image format COnverter for High-throughput Deep-Learning (rico-hdl).

Train Convolutional Neural Network to predict land cover type from multispectral Sentinel-2 satellite imagery

This repository will guide you how to use deep learning algorithms for land use land cover classification using satellite dataset!

End-to-end deep learning: satellite land-use classification (EuroSAT, EfficientNet-B0, 94.5% test accuracy) with Grad-CAM, FastAPI web app and drift monitoring

Interactive web app for land use classification from Sentinel-2 satellite imagery using deep learning.

Federated Learning in Satellite Constellations using Flower

Satellite land-cover classification and mapping: PyTorch transfer learning, Grad-CAM explanations, calibrated probabilities, FastAPI + web UI, Docker. Validated on real Sentinel-2 imagery against ESA WorldCover.

A reproducible cross‑framework study comparing CNN and CNN‑ViT hybrid architectures for EuroSAT satellite crop classification using aligned Keras and PyTorch implementations.

Satellite land cover classification using Transfer Learning with ResNet50 and PyTorch on the EuroSAT dataset.

Cross domain few-shot transfer learning from MiniImageNet to EuroSAT_RGB and CUB

WildSAT is an ecological AI system that lets users search satellite imagery using natural language queries like "dense tropical forest" or "wildfire-prone vegetation" via a CLIP-style contrastive pipeline, using a ResNet18 + ViT-B/16 hybrid encoder and DistilBERT as text encoder, visualized on an interactive Mapbox satellite map.

Domain adaptation for satellite image classification using a pretrained EuroSAT ResNet50 model adapted to Sentinel-2 + Dynamic World target data.

EuroSAT land-use classification with ResNet-18, two-phase fine-tuning, FastAPI, and Streamlit

Random Forest vs. a small PyTorch CNN on EuroSAT-style satellite tiles (73.22% vs 89.75%); machine learning course project

DL-LULC-Classifier is a deep learning project for Land Use and Land Cover (LULC) classification using Convolutional Neural Networks (CNNs). It features can support multiple models, easy integration with Django and HTMX as frontend. This tool is ideal for environmental monitoring and geospatial analysis.