Xiong Xingliang, Zhang Yan, Chen Mengmeng, Chen Longcong
Lab of Biomedical Engineering, Chongqing Medical University, Chongqing 400016, China.
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2013 Apr;30(2):239-44.
Objective evaluation of driver drowsiness is necessary toward suppression of fatigued driving and prevention of traffic accident. We have developed a new method in which we utilized pupillary diameter variability (PDV) under spontaneous pupillary fluctuation conditions. The method consists of three main steps. Firstly, we use a 90s long infrared video of pupillogram infrared-sensitive CCD camera. Secondly, we employed edge detection algorithm based on curvature characteristics of pupil boundary to extract a set of points of visible pupil boundary, and then we adopted these points to fit a circle to obtain the diameter of the pupil in current frame of video. Finally, the values of PDV in 90s long video is calculated. In an experimental pilot study, the values of PDV of two groups were measured. One group rated themselves as alert (12 men), the other group as sleepy (13 men). The results showed that significant differences could be found between the two groups, and the values were 0.06 +/- 0.005 and 0.141 +/- 0.042, respectively. Taking into account of the knowledge that spontaneous pupillary fluctuation is innervated by autonomic nervous system which activity is known to change in parallel with drowsiness and cannot be influenced by subjective motive of people. From the results of the experiments, we concluded that PDV could be used to evaluate driver fatigue objectively.
客观评估驾驶员困倦对于抑制疲劳驾驶和预防交通事故是必要的。我们开发了一种新方法，该方法利用自发瞳孔波动条件下的瞳孔直径变异性（PDV）。该方法包括三个主要步骤。首先，我们使用对红外敏感的CCD摄像机拍摄90秒长的瞳孔图红外视频。其次，我们采用基于瞳孔边界曲率特征的边缘检测算法来提取可见瞳孔边界的一组点，然后用这些点拟合一个圆以获得视频当前帧中瞳孔的直径。最后，计算90秒长视频中的PDV值。在一项实验性初步研究中，测量了两组的PDV值。一组自我评定为清醒（12名男性），另一组为困倦（13名男性）。结果表明，两组之间存在显著差异，其值分别为0.06±0.005和0.141±0.042。考虑到自发瞳孔波动受自主神经系统支配，而自主神经系统的活动已知会随着困倦程度而平行变化，且不受人的主观动机影响。从实验结果来看，我们得出结论，PDV可用于客观评估驾驶员疲劳。
