摘要: |
An increasing amount of research effort is being spent on the correlation between sleep deprivation and alertness. Sleep deprivation can cause poor human performance in highly monotonous activities. Driving an automobile is an example of such an activity. Since a largenumber of people today are sleep deprived, driver fatigue has drawn a lot of attention, due to the importance of being alert while driving. In this paper, we describe a system that locates and tracks the eyes of a driver, in order to perform detection of driver fatigue. By mounting a small camera inside the car, we can monitor the face of the driver and look for eye-movements which indicate that the driver is no longer in condition to drive. In such a case, a warning signal is issued. In this paper we focus on the detection of "microsleeps." These are short periods (approximately 2-3 seconds) during which the driver rapidly loses consciousness due to fatigue. By accurately detecting these microsleeps, we are able to make the driver aware that a dangerous fatigue-level has been reached. For detection of microsleeps, we have developed a method to determine whether the eyes are open or closed. The system presented in this paper is also able to automatically detect tracking errors, that may occur. The system is designed so that once started, it is able to run by itself, updating its parameters automatically if necessary. The primary criterion for this kind of systems is that they must be highly non-intrusive. The system should start when the ignition is turned on without having the driver initiate the system. Nor should the driver be responsible for providing any feedback to the system. The system must also operate regardless of the texture and the color of the face. It must also be able to handle changing conditions, such as changes in light, shadows reflections, etc. |