HeliosIA is an AI-powered tool designed to analyze nighttime road usage by cars and pedestrians using drone-captured videos. By processing high-altitude, downward-looking footage, HeliosIA maps traffic patterns, visualizes pedestrian movement, and provides valuable insights into road utilization under low-light conditions.
- Features
- Prerequisites
- Installation
- Usage
- Processing Raw Videos
- Contributing
- License
- Acknowledgments
- Traffic Analysis: Detects and tracks vehicles and pedestrians in nighttime drone footage.
- Visualization: Maps and visualizes traffic patterns and pedestrian movements.
- Insights: Provides data-driven insights on road utilization during low-light conditions.
- AI-Powered: Utilizes advanced AI algorithms for object detection and tracking.
- Python 3.7 or higher
- Required Python packages (see
requirements.txt
) - Processed video data (see Processing Raw Videos)
-
Clone the Repository
git clone https://github.com/Septimus4/HeliosIA.git
-
Navigate to the Project Directory
cd HeliosIA
-
Install Required Packages
pip install -r requirements.txt
-
Prepare Processed Videos
Ensure your drone-captured videos are processed using HeliosPostProcessing before analysis.
-
Run the Analysis
python analyze.py --input path_to_processed_videos --output results/
-
View Results
Analysis results will be saved in the specified output directory. Use provided visualization tools to interpret the data.
To process raw drone videos for use with HeliosIA, utilize the HeliosPostProcessing tool. This pre-processing step includes stabilizing footage, enhancing low-light visibility, and formatting data for analysis.
Steps:
-
Clone HeliosPostProcessing
git clone https://github.com/Septimus4/HeliosPostProcessing.git
-
Process Videos
Follow the instructions in the HeliosPostProcessing repository to prepare your videos.
-
Output
Use the processed videos as input for HeliosIA.
Contributions are welcome! Please open an issue or submit a pull request for enhancements or bug fixes.
This project is licensed under the MIT License. See the LICENSE file for details.