| Project Overview | This B.Tech Aerospace / Aeronautical Engineering project is based on the recent research direction 'Onboard Satellite Image Classification for Earth Observation: A Comparative Study of Pre-Trained Vision Transformer Models'. The project focuses on applying artificial intelligence, machine learning, deep learning, computer vision, reinforcement learning, surrogate modelling, or RAG-style intelligent assistance to the Satellite and Space Applications area. Students can use the linked 2023-onward research paper/source as the academic base, then convert it into an implementation-focused final-year project with a simplified dataset, simulation model, Python workflow, dashboard, or prototype demonstration. |
|---|---|
| Research Paper Title | Onboard Satellite Image Classification for Earth Observation: A Comparative Study of Pre-Trained Vision Transformer Models |
| Research Paper / PDF Link | Open Paper / PDF |
| Year | 2024 |
| Project Area | Satellite and Space Applications |
| Project Type | Deep Learning Satellite Classification |
| Required Tools / Software | Python, PyTorch/TensorFlow, OpenCV, Rasterio, GeoPandas, Sentinel/Landsat datasets, Streamlit |
| Main Features / Working Principle | Use pretrained vision transformer concepts to classify Earth observation satellite images |
| Expected Output | A satellite image classification dashboard with class prediction |
| Possible Add-ons | Add edge/on-board model compression comparison |
| Get Help | Get Help on WhatsApp
Message: Hi FE, I need help with "Onboard Satellite Image Classification for Earth Observation: A Comparative Study of Pre-Trained Vision Transformer Models" in "Aerospace / Aeronautical Engineering" |
This B.Tech aerospace project resource helps students connect a recent AI-based research direction with a practical implementation plan, tools, expected output, and possible extensions.