Classifying custom image datasets by creating Convolutional Neural Networks and Residual Networks from scratch with PyTorch
-
Updated
Feb 12, 2023 - Jupyter Notebook
#
eurosat-satellite
Here are 14 public repositories matching this topic...
A reproducible cross‑framework study comparing CNN and CNN‑ViT hybrid architectures for EuroSAT satellite crop classification using aligned Keras and PyTorch implementations.
computer-vision deep-learning agriculture tensorflow cv keras cnn pytorch remote-sensing image-classification satellite-imagery satellite-data cnn-keras satellite-images cnn-pytorch eurosat land-cover-classification vision-transformer eurosat-satellite cnn-vit-ensemble
-
Updated
Mar 8, 2026 - Jupyter Notebook
Satellite land cover classification using Transfer Learning with ResNet50 and PyTorch on the EuroSAT dataset.
python machine-learning computer-vision deep-learning satellite pytorch image-classification satellite-imagery transfer-learning resnet-50 adam-optimizer eurosat eurosat-satellite
-
Updated
Aug 11, 2026 - Jupyter Notebook
-
Updated
Jul 26, 2026 - Jupyter Notebook
The Computer Vision and Deep Learning Project is designed to predict dominant land use from Zenodo satellite images from EuroSAT dataset.
-
Updated
Apr 11, 2024 - Python
Using a custom CNN trained on the EuroSAT dataset to classify satellite images into 10 classes. It provides a Streamlit app for interactive classification and a Jupyter notebook for experimentation and visualization. Users can test, visualize, and customize predictions with their own images or modify the model architecture.
-
Updated
Aug 17, 2025 - Jupyter Notebook
Trains a CNN on EuroSAT satellite imagery for land-cover classification (93.6% test accuracy), then applies it to real GeoTIFF scenes across two time points to automatically flag Forest → non-forest transitions as candidate deforestation events.
computer-vision pytorch satellite-imagery cnn-model deforestation-model cnn-architecture land-cover-classification eurosat-satellite deforestation-detection
-
Updated
Jul 16, 2026 - Jupyter Notebook
Aether-OS is a production-ready satellite imagery intelligence platform designed to process, analyze, and prioritize multispectral EuroSAT Sentinel-2 imagery through an end-to-end intelligent pipeline. The system combines machine learning, adaptive calibration, and automated verification to deliver highly reliable satellite image classification.
-
Updated
Apr 26, 2026 - Python
Deep Learning based Satellite Land Use Classification and Temporal Change Detection using ResNet18, EuroSAT and Streamlit.
machine-learning computer-vision deep-learning cnn pytorch remote-sensing satellite-imagery change-detection resnet18 streamlit eurosat-satellite
-
Updated
Jul 13, 2026 - Python
Hybrid CNN-ViT for 13-band EuroSAT satellite classification | 5-model ablation | Grad-CAM XAI | MC Dropout | 96.98% accuracy | Google Scholar publication
deep-learning pytorch transformer remote-sensing google-scholar ablation-study vision-transformer eurosat-satellite statelite-imagery
-
Updated
Mar 11, 2026 - Jupyter Notebook
Satellite image classification using ResNet50 transfer learning on the EuroSAT dataset, achieving 98.02% validation accuracy.
machine-learning computer-vision deep-learning tensorflow image-classification satellite-imagery resnet-50 eurosat-satellite
-
Updated
Jun 14, 2026 - Jupyter Notebook
Custom CNN for 13-band multispectral land cover classification using EuroSAT Sentinel-2 satellite dataset
machine-learning computer-vision deep-learning tensorflow cnn remote-sensing satellite-imagery sentinel-2 eurosat-satellite
-
Updated
Sep 16, 2026 - Jupyter Notebook
Advanced Land Use Detection and Classification Using EuroSAT Dataset and Machine Learning
machine-learning deep-learning rgb resnet convolutional-neural-networks resnet-50 cnn-classification resnet50 multispectral eurosat land-use-classification convnext eurosat-satellite land-use-detection
-
Updated
Apr 25, 2026 - Jupyter Notebook
Self-supervised learning for Sentinel-2 imagery
-
Updated
Aug 1, 2026 - Python
Add this topic to your repo
To associate your repository with the eurosat-satellite topic, visit your repo's landing page and select "manage topics."