Research On Fatigue Driving Arithmetic Based On Computer Vision

With the development of society, the cars increase quickly. Traffic accidents caused by fatigue driving increase more and more too, bringing serious threat to people's life and property. In order to protect the traffic security and to prevent traffic accidents,it is very necessary and meaningful to research effective ways to detect driver fatigue status and fatigue alarm.The thesis analyses and researches the status about driver fatigue of both at home and abroad systemic, especially the arithmetic of driver fatigue detection based on computer vision, the main job is as follows:A new face detection method is studied which is based on a skin model and image processing technique in this thesis. And a new color space YCgCr of the skin model is introduced to extract the skin color using the method of an elliptical clustering model with the constraint conditions of Cg and Cr. Potent factor of face judgment is introduced to further removing non-face regions in the skin-color segmentation results, so the searching regions for face detection reduce and the face is located at last. The thesis deals with the face tracking in succession, based on an algorithm for real-time automatic face tracking which combined Mean shift algorithm and Kalman filter. At last on the basis of the confirming face region, the driver's eyes are detected with a variety of image processing methods, using coarse-to-fine and gradually positioning strategy. After an accurate positioning of the eye region, we calculate the eye frame curves and the distance between canthus line connecting to get the open degree of the eye, and by calculating the corresponding PERCLOS, the average eye opening degree, and maximum eye closure duration, we can detect the driver's fatigue state.