Driver’s fatigue detection system based on facial features
With increasing number of vehicles on roads the risk of getting involved in an accident is increasing as well. In Malaysia alone, the number of traffic accidents in 2007 almost doubled as compared to the number of traffic accidents that occurred in 1997. This high accident rate has led to road...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/1477/1/24p%20IJAZ%20KHAN.pdf http://eprints.uthm.edu.my/1477/2/IJAZ%20KHAN%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/1477/3/IJAZ%20KHAN%20WATERMARK.pdf http://eprints.uthm.edu.my/1477/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | With increasing number of vehicles on roads the risk of getting involved in an
accident is increasing as well. In Malaysia alone, the number of traffic accidents in
2007 almost doubled as compared to the number of traffic accidents that occurred in
1997. This high accident rate has led to road accidents being the 5th leading cause of
death in Malaysia and caused 9.3 billion ringgit of losses to the country in the year
2003. According to NHTSA (National Highway Traffic System Administration)
reports one of the major reasons of road side accidents is fatigue while driving.
Therefore, to prevent road side accidents that occurs due to fatigued drivers, it is
essential to have an assistive system inside vehicle that monitors the vigilance level
of driver and alert the driver in case of fatigue detection. This thesis presents a
fatigue detection system based on yawning and eyes status that continuously analyse
the face and facial features of the driver. Vision based approach is adopted to detect
fatigue because other developed approaches are either intrusive (physical approach)
that makes the driver uncomfortable or less sensitive (vehicle based approach). This
system has improved the accuracy of fatigue detection by contributing in 3 steps of
fatigue detection process. First step is face detection for which combination of Viola
Jones and skin color pixels detection is used. Second is accurate detection of eyes
and mouth in detected face area. The system uses knowledge based division and
Viola Jones technique for second step. The third step is the introduction of dynamic
threshold value, to check weather driver is in yawning or sleeping state. The
accuracy of the system to detect fatigue level of driver is 98 % and average
processing time per frame is 0.0948 seconds. The simulation results show that this
system is able to detect fatigue even if driver is wearing spectacles or having beard.
The algorithm is developed in MATLAB software. |
|---|