Distracted Driver Detection
A computer vision model that detects distracted drivers and classifies their behaviour into one of ten distraction categories.
Project Details
About the Project
Road safety has always been a concern, and one of its leading causes is distracted driving.
To help address this issue, we developed a computer vision project that detects distracted drivers and classifies their behaviour into one of ten distraction categories.
How It Works
Using machine learning techniques, we trained models to recognise different types of driver distraction from images.
The categories include safe driving, texting, talking on the phone, operating the radio, drinking, reaching behind, doing hair or makeup, and talking to a passenger.
Project Outcome
The project achieved strong accuracy and precision, allowing us to explore the potential of AI in contributing to safer roads and smarter transportation systems.
Overall, we felt satisfied with the project and what we managed to achieve.
Future Improvements
Beyond image classification, we were also interested in extending the project to detect distractions from videos. This would make the system more practical for real-world applications.
However, time constraints meant that we were unable to fully explore this feature. Given more time, we hope to build upon our current work by improving its real-time video detection capabilities.
Biggest Challenge
One of our biggest challenges was sourcing suitable datasets, as high-quality training data is essential for building reliable machine learning models.
Key Takeaway
We learnt that developing machine learning solutions involves considerable trial and error, from experimenting with different models to fine-tuning their performance.