International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 11 | Nov 2024
p-ISSN: 2395-0072
www.irjet.net
DYNAMIC DRIVER FATIGUE MONITORING THROUGH VISUAL ANALYSIS G. Jithin1, B. Jothsna Sai Sri2, R. Akshaya3, Dr. G Ganapathi Rao4 123Computer Science and Engineering (Data Science) Institute of Aeronautical Engineering Hyderabad, India
4Assistant Professor Computer Science and Engineering (Data Science) Institute of Aeronautical Engineering
Hyderabad, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract –Road accidents are a common issue, with India
techniques, these systems can alert drivers promptly, helping to prevent accidents.
recording the highest rate of fatalities, often due to driver drowsiness. Fatigue and sleepiness are significant factors contributing to these accidents. This paper outlines a system to detect driver drowsiness by analyzing eye aspect ratio (EAR) and mouth aspect ratio (MAR). If these values exceed set thresholds, an alert is triggered, notifying both the driver and passengers, thereby enhancing road safety and reducing accident risks. Based on EAR, drowsiness for the eye is detected and based on MAR , yawning is detected. Both yawning and sleepiness cause fatigue in drivers. So, to overcome this in this location of nearby hotels is displayed when a driver is under drowsiness.
It monitors the driver's behavior for signs of drowsiness and provides timely alerts, encouraging them to take a break before losing focus or falling asleep. This system reduces the risk of accidents, especially on long drives or for commercial drivers, ensuring safer journeys and minimizing financial and legal impacts related to collisions. Road accidents frequently result from driver drowsiness, reduced focus, and fatigue, leading to slower reaction times and impaired control. Detecting drowsiness is critical for preventing these accidents by continuously monitoring the driver's alertness and issuing timely warnings. This system helps reduce crashes, protect lives, and ensure safer driving, particularly on long trips or during nighttime driving.
Key Words: Drowsiness Detection, Hybrid Features, Transfer Learning, EAR, MAR, Hotel locations.
Haar Cascade Classifier is a popular machine learning-based approach for face detection. It uses Haar-like features, which are patterns of pixel intensities, to identify objects like faces in images. The classifier is trained on numerous positive images (faces) and negative images (nonfaces) to distinguish between them. During detection, it scans the image at multiple scales and positions, looking for areas where the features match those of a face. The method is efficient and works well for real-time face detection, commonly used in applications like security systems and camera-based face recognition.
1.INTRODUCTION Driver drowsiness detection systems are designed to monitor and analyze a driver's physical or behavioural signals to assess alertness levels and identify signs of fatigue. By detecting cues like slow blinking, yawning, or erratic head movements using cameras or sensors, the system can alert the driver to take action before an accident occurs. Some advanced systems also monitor physiological signals, such as heart rate, or use machine learning algorithms to improve detection accuracy.
Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) are key indicators used in driver drowsiness detection. EAR measures the ratio of distances between vertical eye landmarks and the horizontal distances between the eye corners. When a driver becomes drowsy, the eyes tend to blink more frequently or remain partially closed, causing a significant drop in EAR values. By continuously monitoring this ratio, the system can detect prolonged eye closures, signalling potential drowsiness.
In today's fast-paced world, people often overlook the importance of proper sleep due to their busy schedules, which can have a significant impact on road safety. Driving while drowsy is one of the leading causes of accidents, alongside drunk driving and general negligence. The consequences of these accidents are far-reaching, affecting not only the drivers but also others on the road. The growing number of vehicles has heightened the need for effective measures to ensure safety, with driver fatigue being a major factor in traffic accidents.
Also, MAR tracks the ratio of mouth opening based on facial landmarks. As drivers get drowsy, they tend to yawn frequently, which increases the MAR value. By analysing these yawning patterns in combination with eye closure detection, the system can more accurately identify signs of fatigue, providing an early warning to help prevent accidents. Combining EAR and MAR enhances the overall accuracy of drowsiness detection systems.
To address this, driver drowsiness detection systems have emerged as crucial technologies that monitor the driver's behavior in real-time. These systems rely on visual analysis, focusing on cues such as eye closures, yawning, and head movements to detect signs of fatigue. By using advanced computer vision and machine learning
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 415