https://github.com/kevalshah91/aerial-vehicle-mapper
Aerial Vehicle Mapper is an AI-powered tool that detects and maps vehicles from aerial images using the Detectron2 deep learning framework. It enables geospatial mapping of detected vehicles for applications in urban planning, surveillance, disaster management, and more.
https://github.com/kevalshah91/aerial-vehicle-mapper
aerial-imagery computer-vision detectron2 geospatial-mapping remote-sensing satellite-imagery urban-planning
Last synced: 7 months ago
JSON representation
Aerial Vehicle Mapper is an AI-powered tool that detects and maps vehicles from aerial images using the Detectron2 deep learning framework. It enables geospatial mapping of detected vehicles for applications in urban planning, surveillance, disaster management, and more.
- Host: GitHub
- URL: https://github.com/kevalshah91/aerial-vehicle-mapper
- Owner: Kevalshah91
- Created: 2025-01-24T15:49:21.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-03-30T04:56:37.000Z (7 months ago)
- Last Synced: 2025-03-30T05:24:55.590Z (7 months ago)
- Topics: aerial-imagery, computer-vision, detectron2, geospatial-mapping, remote-sensing, satellite-imagery, urban-planning
- Language: Jupyter Notebook
- Homepage:
- Size: 10.9 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Aerial Vehicle Mapper 🛁
Welcome to **Aerial Vehicle Mapper**! This repository provides a powerful tool for detecting and mapping vehicles from aerial images using the Detectron2 deep learning framework. 🚗🛰️
---
## Features ✨
- **Vehicle Detection**: Detects cars, trucks, buses, and motorbikes in aerial images.
- **High Accuracy**: Built on Detectron2's advanced object detection models.
- **Custom Thresholding**: Adjust detection confidence levels for greater flexibility.
- **Visualization**: Produces annotated images with bounding boxes for detected vehicles.
---
## Applications 🌍
- **Urban Planning**: Analyze traffic patterns and vehicle density in cities.
- **Surveillance**: Monitor vehicle movements for security purposes.
- **Disaster Management**: Identify vehicles in affected areas during emergencies.
- **Environmental Studies**: Study vehicular impact on ecosystems.
- **Infrastructure Development**: Aid in designing better road networks and parking facilities.
---
## Prerequisites 🛠️
Make sure you have the following:
- Python 3.8+
- NVIDIA GPU with CUDA support (for optimal performance)
- `pip` for package management
---
## Examples 🗂️
### Input Image
### Output Image
---
## How It Works ⚙️
```mermaid
graph TD;
A[Load Aerial Image] -->|Preprocess Image| B(Resize & Normalize);
B -->|Pass to Model| C[Detectron2 Model];
C -->|Run Inference| D{Object Detection};
D -->|Extract Vehicle Coordinates| E[Bounding Box Extraction];
E -->|Apply Confidence Threshold| F[Filter Detections];
F -->|Overlay Annotations| G[Visualize Results];
G -->|Save Annotated Image| H[Store Output];
H -->|Integrate with GIS| I[Geospatial Mapping];
I -->|Generate Reports| J[Final Output];
```
---
## Dependencies 🛆
- Detectron2
- PyTorch
- OpenCV
- NumPy
- PyYAML
**Happy Mapping! 🗰️**