According to the report from World Health Organization (WHO), there are more than 1 million people die from the traffic accident every year. Most reasons of the accident are caused by the bad driving behavior due to the fatigue driving and the poor emotion status of the driver. In order to reduce these kinds of traffic accident, a Driver Monitoring System is the core idea in this research. By using a camera inside the car, the facial image of the driver is captured and the 2D image will be converted into a 3D face model. Based on the 3D face model, the facial action units which are utilized to summarize the facial expressions can be detected. These facial action units, then, are employed to detect the driver fatigue, emotion status, and the expressions of pain. When the system detects the fatigue, the bad emotion status, the pain expression, or that the driver is not focus on the road, it can warn the driver to mitigate the rate of the traffic accident.