CityU Institutional Repository >
Student Final Year Projects >
Electronic Engineering - Undergraduate Final Year Projects >
Please use this identifier to cite or link to this item:
|Title: ||Driver Fatigue Detection System using Open CV|
|Authors: ||Wong, Mei Ching|
|Department: ||Department of Electronic Engineering|
|Issue Date: ||2011|
|Supervisor: ||Supervisor: Dr. Po, L M; Assessor: Mr. Chau, W N|
|Abstract: ||At the present day, many traffic accidents were occurring everyday. Through the car have
seat belt or SRS to protect drivers, the accidents still occur with drivers in
unconsciousness. The most reason of the traffic accidents is the drivers were distracted by
drug, drunk or talking to the phone/others people. Those things are weaken drives focus
and judgment to increase accident probability.
In this project, the aim is thought a surveillance system to keep track the drivers face and
behavior who are misconduct when they are driving. System provided an environment to
analyze driver’s face-emotion and head moving. When drives have a frequency tired
motion such as system detect driver head is not focus or close his eyes have a long time,
system will alarm derives when exceeded a threshold time. In the driver monitoring
technology, it can be able to detect driver "tried level" and provide a warning system to
inform derives. In this project, a statistical model will be use of find out the facial feature
points, through its analysis the program can find the user face location and tracking it.
This program can using in a real time detecting, also it can record the detecting result for
after analysis and inference the driver behavior.
This system can supervise 24-hours without any assist. System can issue an alarm to
drivers when the accident is going to happen. Using this method to increase driving
safely and is great for reduce the traffic accident.|
|Appears in Collections:||Electronic Engineering - Undergraduate Final Year Projects|
Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.