CAJ | 학술논문

驾驶员精神状态监测对于交通事故控制具有重要作用。基于支持向量机（SVM ）建立信息融合的面部朝向检测模型，在驾驶员面部图像视觉特征（眼睛、嘴部及面部轮廓）检测基础上，通过分析头部运动过程中各特征的空间位置变化规律，利用SVM 对驾驶员面部朝向进行多信息融合估计。结果表明，特征检测方法可实现面部各视觉特征的可靠检测，分析得出的各特征运动变化规律可用于对面部朝向的有效估计，从而为驾驶员驾车精神状态监测提供技术支持。
가사원정신상태감측대우교통사고공제구유중요작용。기우지지향량궤（SVM ）건립신식융합적면부조향검측모형，재가사원면부도상시각특정（안정、취부급면부륜곽）검측기출상，통과분석두부운동과정중각특정적공간위치변화규률，이용SVM 대가사원면부조향진행다신식융합고계。결과표명，특정검측방법가실현면부각시각특정적가고검측，분석득출적각특정운동변화규률가용우대면부조향적유효고계，종이위가사원가차정신상태감측제공기술지지。
Driver′s spiritual state monitoring is important for reducing traffic accidents .Based on support vector machine (SVM ) , a face orientation recognition model with information fusion is presented .Firstly , based on the detected multiple visual features such as eyes , mouth and facial contours ,the features′ position variations during the head moving are analyzed . Then , the face orientation is estimated with these characteristics through SVM method . The experimental results show that the proposed method for features detection has good reliability and the face orientation can be inferred according to the relevant characteristics .It is clear that the proposed method has strong potential effectiveness to be applied to the driver′s spiritual state monitoring .