Abstract: In view of some of the difficulties existing in fatigue driving detection,this paper proposes a novel method for drivers'eyes location and tracking.Firstly we uses Gamma and SQI (Self Quotient Image) algorithm to process images,which obtained by infrared light,then locate the brow area by AdaBoost algorithm,and on this basis,with the brow area 15 points ASM (Active Shape Models) model to accurately locate driver's eye area,finally the unscented kalman filter algorithm is adopted to drivers'eyes tracking.The experimental results show that the algorithm introduced can not only improve the robustness of the illumination,pose transformation,and driver's eye's strongly nonlinear movement,but also get ideal reault,even if the driver wearing glasses.