Classification of Eye Movement and Its Application in Driving Based on a Refined Pre-Processing and Machine Learning Algorithm
Xiansheng Li, Zhi-Zhen Fan, Yuanyuan Ren, Xuelian Zheng, Ran Yang
Abstract
The eyes are the first channel used by humans to obtain various types of visual information from the outside world, especially for driver, 80-90% of information is received through eyes in driving. The eye movement behaviors are generally divided into six types, but attention is often paid to fixation, saccade, and smooth pursuit. Therefore, it is essential to classify eye movement behaviors accurately using an excellent classification algorithm. The classification of eye movement should be a complete process, including three steps of pre-processing, classification and post-processing. However, it is very uncommon that all these steps are included in eye tracking literature where eye movement classification is discussed. Therefore, first, this paper proposes a refined eye movement data pre-processing framework and the improved method of three steps are emphatically introduced. Second, an eye movement classification algorithm based on improved decision tree that is independent of the threshold setting and application environment is proposed, and a post-processing of merges adjacent fixations and discards short fixations is described. Finally, the application of the classified eye movement behavior in the driving field is described, including the estimation of preview time using fixation and the estimation of time-to-collision using smooth pursuit. Two important results are obtained in this paper. One is about the classification accuracy of eye movement behavior, the F1-score of fixation, saccade and smooth pursuit are respectively 92.63%, 93.46% and 65.29%, which is higher than that of the other algorithms. The other is about the application in driving, on the one hand, the preview time calculated by fixation is mostly distributed around 1-6s, which is more real compared with the traditional setting of 1s, at the same time, the regression relationship between preview time and road turning radius is also quantitatively analyzed and their regression function is obtained. On the other hand, the average estimated error of time-to-collision used by smooth pursuit is 7.37%. These results play an important role in the development of ADAS and the improvement of traffic safety.