The accident rate in the world is rising every year. According to the national committee for the prevention of traffic accidents, in Morocco The human factor is responsible for more than 80% of the road accident, it occurs due to many reasons like speeding, drinking, and drug use while driving, the use of mobile, tiredness or the lack of sleep. These problems can be solved by increasing intelligence and technology with intelligent transportation systems. The driver behavior is an essential component of the driver-vehicle-environment system, so an improper driving behavior is the leading cause of the accidents and thus, the identification of inappropriate behavior is important to achieve intelligent driving and to improve road safety. We chose to work on real-time and non-intrusive techniques to identify the driver behavior in current time and without attachment which is more practical. This work gathers the cues and parameters that we will use for detecting driver behavior, especially drivers with fatigue. The facial expression of a person in fatigue or in the onset of fatigue can be detected by the face pose that gives us the frequency of head tilts, the eyelid movement which is one of the visual behaviors that reflect a person's level of fatigue, the eye gaze, the facial expression like lagging facial muscles, and the mouth movement that allows to compute the occurrence frequency of yawning over time.