DriveAware distraction detection model
Convolutional Neural Network (CNN) design
Computer Science
DriveAware: Next Generation of Free Distracted Driving Detection
Computer Science
DriveAware: Next Generation of Free Distracted Driving Detection
A team of computer science students developed an Android app to combat distracted driving. This app was created for CSC 491: Senior Project I and CSC 492: Senior Project II.
Overview
DriveAware is an Android app that uses a smartphone’s camera to track eye movements and detect distractions—eliminating the need for specialized hardware. When distraction or drowsiness is detected, the app issues auditory and visual alerts.
Student Team
Caden Effrece '25
Computer Science
School of Computing & Engineering
Ryan Sliger '25
Computer Science
School of Computing & Engineering
Klaus Schroeder '25
Computer Science
School of Computing & Engineering
DriveAware: Next Generation of Free Distracted Driving Detection
Introduction
Distracted driving is a major issue in today's technology-driven world, yet most solutions are costly or require complex setups. To address this, we developed DriveAware, an Android app that uses a smartphone’s camera to track eye movements and detect distractions—eliminating the need for specialized hardware. When distraction or drowsiness is detected, the app issues auditory and visual alerts. Using MediaPipe and machine learning, DriveAware operates seamlessly in the background with an intuitive interface.
Material and Procedures
System Architecture
- Core Tech: Google MediaPipe (eye/head tracking) + K-NN ML model (80% accuracy).
- Database: Firebase (user auth, drive reports, sharing).
- Algorithm: Processes camera images every 2s to classify focus.
Evaluation
- Participants: 7-10 drivers in simulated sessions.
- Metrics: Precision, recall, response times to alerts.
Requirements and Design
The DriveAware distraction detection model is a Convolutional Neural Network (CNN) designed to classify whether a driver is focused or distracted based on various images of the user’s eyes. The input consists of 64x64 grayscale images of the driver's eyes.
Steps for Distraction Algorithm:
- First flatten the image
- Then passes the processed image through multiple layers to recognize patterns in the eyes
- Finally outputs a score between 0 and 1, predicting whether the driver is distracted or not
- Reinforcement learning is used to fine-tune key settings, such as:
- Number of layers
- Learning speed
- Dropout amount
App Features
- Reduces Distracted Driving
- Enhances Road Safety
- User-Friendly and Free
- Personalized Driving
Summary
DriveAware is an Android app that uses AI and smartphone cameras to detect distracted driving in real-time.
- AI Tracking: Monitors eye/head movement via MediaPipe + KNN (80% accuracy)
- Instant Alerts: Auditory/visual cues refocus drivers
- No Cost: Works with existing phones
- Proven: Reduced distractions in tests
Conclusions
DriveAware offers a free, accessible solution using existing smartphones.
Future Plans
- iOS Compatibility – Expanding the app to work on iPhones to reach more users.
- Improved AI Accuracy – Enhancing the machine learning model with a larger dataset to improve distraction detection accuracy beyond 80%.
- Integration with Car Systems – Connecting with Bluetooth or vehicle infotainment systems to provide seamless, hands-free functionality.
- Offline Mode – Allowing the app to function without an internet connection for more reliable performance in all driving conditions.
Supporting Visual Elements
Graph depicting the ML-algorithms tested
App interface for DriveAware
For Further Discussion
This serves as an overview of the project and does not include the complete work. To further discuss this project, please email Caden Effrece.
Course Overview
CSC 491: Senior Project I is the first part of a two-semester, capstone experience for computer science students. Students analyze and develop a solution to a major project that requires integration and application of knowledge and skills acquired in earlier coursework. Students develop professional experience by working on a team and communicating progress and results to a variety of audiences. Students explore the ethical and legal responsibilities of a computing professional.
CSC 492: Senior Project II is the second part of a two-semester, capstone experience for computer science students. Students implement and evaluate a solution to a major project that requires integration and application of knowledge and skills acquired in earlier coursework. Students continue to develop professional skills in teamwork and communications, and knowledge of their responsibilities as computing professionals.
Explore Our Areas of Interest
We've sorted each of our undergraduate, graduate and doctoral programs into unique Areas of Interest. Explore these categories to discover which programs and delivery methods best align with your educational and career goals.