CAJ | 학술논문

智能交通系统中对驾驶员的行为监测具有重要的研究价值，针对驾驶员的头部姿态监测问题进行研究，实现基于计算机视觉的动态头部姿态估计算法。算法首先检测视频中的正面人脸，并作为基准，在此基础上对视频序列中人脸特征进行匹配与跟踪，计算不同视角下的头部姿态角度。考虑累计误差的影响，提出一种矫正累积误差的方法。实验表明算法可以较准确地估计视频中的驾驶员头部姿态角度。
지능교통계통중대가사원적행위감측구유중요적연구개치，침대가사원적두부자태감측문제진행연구，실현기우계산궤시각적동태두부자태고계산법。산법수선검측시빈중적정면인검，병작위기준，재차기출상대시빈서렬중인검특정진행필배여근종，계산불동시각하적두부자태각도。고필루계오차적영향，제출일충교정루적오차적방법。실험표명산법가이교준학지고계시빈중적가사원두부자태각도。
It is important to study the driver’s behavior consciousness in the intelligent traffic system .In this paper ,driver’s head pose estimation is researched ,and a vision‐based scheme of estimating the driver’ s head pose is realized .Firstly ,the driver’s front face is detected from the video image ,and the result is used as criterion .Then corresponding feature points between two adjacent views are matched .According to the points ,the driver’s head pose is estimated .It also brings forward a method to capture the key frames online through fusing the smoothness assumption of head’s motion and other prior motion information ,in order to eliminate the accumulating errors during continuous pose accumulation .The experiment shows the better performance on experimental data can be obtained .