In this thesis, we present our latest research on the development an non-intrusive system for monitoring driver vigilance from driver's facial video images. Our system simultaneously extracts in real time from the video images of the driver's face multiple visual cues that typically characterize human fatigue, including eyelid movement, head movement, and gaze movement.; The research has led to the development a prototype system that can monitor eye and head activities and compute various fatigue parameters in real time. The main components of the system consists of a hardware system for real time image acquisition and for controlling the illuminator and the alarm system, and various computer vision algorithms and their software implementations for real time eye tracking, eyelid movement computation, face pose discrimination, and near-real time gaze estimation. Each part of our fatigue monitor system was tested under different illumination conditions. The system was found very robust, reliable and accurate.
展开▼
