Driver Fatigue Detection Based On Eye Track: A Survey

Most of the road accident is caused by sleepiness of the drivers during the night time. Monitoring the driver and detecting the sleepiness by sensor is expensive. Some of the driver fatigue system is built fully with wire that may disturb the driver. To avoid this issue of expensiveness and other disturbance, here an image processing technique is used to find the driver’s sleepiness detection. It is used for preventing the accidents caused by driver’s sleepiness and the detection is based on tracking the eyes of the driver. The IR camera is fixed in front of the driver to obtain the PERCLOS. Through that, the driver state has been captured by the mounted camera and the further process is composed of three different stages, first stage is detection of face, eye detection and normalizing it, second stage performs driver position detection and characterization for light filtering. Final stage is for calculating the PERCLOS. This system does not need any calibration process and includes techniques in order to efficiently overcome the typical problems of the image processing algorithms such as: changes of lighting conditions, user appearance and fast head movements. Key Words: Face and eye detection, PERCLOS, drowsiness parameter, calibration process.