摘要
传统算法进行驾驶员疲劳驾驶识别,无法避免由于图像质量较差造成的眼睛瞳孔位置和直径无法确定的缺陷,降低了驾驶员疲劳驾驶识别的准确率。本文提出了一种基于眼睛与面部表情关联性算法的驾驶员疲劳驾驶识别方法。提取驾驶员眼睛与面部表情特征,计算驾驶员眼睛与面部表情特征关联性概率,克服了传统算法的弊端。实验表明,这种算法能够提高驾驶员疲劳驾驶识别的准确率,取得了令人满意得效果。
Using traditional algorithm pilot fatigue driving recognition,cannot avoid the image quality is poorer cause eye pupil the position and could not determine the diameter of the defect,reduce the pilot fatigue driving identification accuracy.Put forward based on the eyes and facial expressions relevance of pilot fatigue driving algorithm identification method.Extraction drivers＇ eyes and facial expressions characteristics,the calculation of drivers＇ eyes and facial expressions features relevance probability,overcome the disadvantages of the traditional algorithm.The experimental results show that this algorithm can improve the pilot fatigue driving accuracy of recognition,and satisfactory results have been obtained.
出处
《科技通报》
北大核心
2012年第12期158-160,共3页
Bulletin of Science and Technology
关键词
疲劳驾驶
面部表情
关联性概率
fatigue driving
facial expressions
relevance probability
