International Journal of All Research Education & Scientific Methods

An ISO Certified Peer-Reviewed Journal

ISSN: 2455-6211

Latest News

Visitor Counter
6278102190

Drowsy Driver Detection Using Machine Learnin...

You Are Here :
> > > >
Drowsy Driver Detection Using Machine Learnin...

Drowsy Driver Detection Using Machine Learning and Open CV Techniques

Author Name : Kolagani Latha, Bandi Shashi Sree Ram Charan Sai, Eluri Surya Vamsi, Korpol Akhil Goud, Dr.G.Aparna

ABSTRACT The "Drowsy Driver Detection System" seeks to improve road safety by addressing the critical issue of driver fatigue, a leading cause of road accidents. By employing advanced computer vision and machine learning techniques, this system identifies signs of drowsiness in real-time, minimizing the risk of accidents caused by inattentive driving. Its core features include real-time video analysis through OpenCV, detection of vital facial landmarks such as eyes and mouth, and the monitoring of fatigue indicators like prolonged eye closure, yawning, and abnormal head posture. The methodology leverages a convolutional neural network (CNN) trained on the Kaggle driver drowsiness dataset to achieve high accuracy in detecting drowsy states. OpenCV processes live video feeds to extract and analyze facial features, ensuring robust performance under diverse conditions, including varying lighting and facial angles. The system focuses on key metrics such as eye closure duration and yawning frequency, triggering instant alerts when drowsiness is detected. The expected outcomes include reliable detection of driver fatigue with minimal false alarms, enabling timely interventions to prevent potential accidents. By addressing a significant safety concern, this project offers a proactive approach to fostering safer driving environments. The system's real-time capabilities highlight its practical application as a valuable addition to vehicular safety mechanisms.