用于检测疲劳驾驶的自适应脸部定位方法

Adaptive Face Location for Detecting Fatigue Driving

针对疲劳驾驶视觉检测技术中驾驶员脸部定位快速性与准确性难以同时满足的情况,提出了一种将肤色和面部几何特征相结合的自适应脸部定位方法。该方法对YCbCr空间肤色相似度和传统Otsu阈值分割算法进行改进,降低了计算复杂度,采用新的区域连通和区域标记算法定位出脸部位置。最后,根据人脸几何特征对脸部区域做进一步确认。试验结果表明,该方法能对驾驶员图像脸部进行快速、准确定位。

According to the fact that rapidity and accuracy of face location of driver are not compatible with each other in the fatigue driving detection technology based on machine vision, a method of face location combined with skin color and geometric features of human face is proposed. Skin color similarity in YCbCr color space and traditional otsu arithmetic are improved,and computation cost is reduced. Face area is located by the novel region connecting and region labeling algorithms. Finally, geometric features of human face are used to verify the detected face in development. Experiment results show that this method is able to locate driver＇s face fast and accurately.