A New Method of Extracting Vigilant Feature from Electrooculography Using Wavelet Packet Transform
1 The Teaching and Research Section of Anesthesia and Medical Imaging Equipments, Wannan Medical College, Wuhu, 241002，China
2 Center for Brainlike Computing and Machine Intelligence, Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China
3 MOEMicrosoft Key Lab for Intelligent Computing and Intelligent Systems, Shanghai Jiao Tong University, Shanghai 200240, China
4 Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai 200240, China
5 Department of Mechanical Engineering and Science, Graduate School of Engineering, Kyoto University Kyoto 6293558, Japan
Abstract：Vigilance refers to the sensitivity when a person concentrates on executing an assignment. To ensure safety, vigilant estimation and prediction is necessary for many kinds of posts, such as highspeed railway drivers and dangerous goods transport drivers. The vigilant estimation and prediction based on physiological signals such as EEG and EOG is an important subject in vigilant research. How to get the best vigilant feature is one of the kernel problems. In this article, the wavelet packet transform was applied for extracting energy ratio in frequency domain from horizontal EOG in order to get the features closely related with vigilance. We discussed energy ratio features in 16 different segmentations and adopted moving average and linear dynamic system to denoise the acquired features. Experimental results show that the demarcation at （0~150 Hz）/（150~3125 Hz） is the best condition. From 35 data sets, the average correlation coefficient is 0742 and the standard deviation is 0151, which is better than the existing 11 features such as slow eye movement, rapid eye movement and blink. The average correlation coefficient increases 555% and the standard deviation decreases 662% than the best feature in the literature.
〖KH*2D〗
〖WTHZ〗Key words：〖WTBZ〗
［1］Ji Qiang, Yang Xiaojie. Realtime eye, gaze, and face pose tracking for monitoring driver vigilance ［J］. RealTimeImaging, 2002, 8(5):357-377.
［2］Grace R, Byrne VE, Legrand JM, et al. A machine vision based drowsy driver detection system for heavy vehicles［C］ //Proceedings of Conference on Ocular Measures of Driver Alertness. Washington DC: FHWA,1999:75-86.
［3］Makeig S, Jung TP. Changes in alertness are a principal component of variance in the EEG spectrum ［J］.Neuroreport,1995,7(1):213-216.
［4］Kenneth P, Wright Jr, Joseph HT, et al. Relationship between alertness, performance, and body temperature in humans ［J］. American Journal of Physiology, 2002, 283(6): 1370-1377.
［5］傅佳伟, 石立臣,吕宝粮. 基于EEG的警觉度分析与估计研究综述 ［J］.中国生物医学工程学报,2009,28(4):589-596.
［6］Wright N, McGown A. Involuntary sleep during civil air operations: wrist activity and the prevention of sleep ［J］. Aviation, Space, and Environmental Medicine, 2004,75(1):37-45.
［7］Hanke S, Zeitlhofer J, Wiest G, et al. Automated vigilance classification based on EOG signals: preliminary results［C］ //Proceedings of World Congress on Medical Physics and Biomedical Engineering. Munich: IFMBE, 2009: 428-431.
［8］Magosso E, Provini F, Montagna P. A wavelet based method for automatic detection of slow eye movements: A pilot study ［J］ Medical Engineering & Physics, 2006, 29(9): 860-875.
［9］Khushaba RN, Kodagoda S, Lal S, et.al.Driverdrowsiness classification using fuzzy wavelet-packetbased featureextraction algorithm ［J］. IEEE Transaction on Biomedical Engineering, 2011, 58(1):121-131.
［10］Cai Haoyu, Ma Jiaxin, Shi Lichen, et al. A novel method for EOG features extraction from the forehead［C］ //Proceedings of 33rd annual International Conference of the IEEE Engineering in Medicine and Biology Society. Boston: IEEE, 2011: 3075-3078.
［11］Ma Jiaxin, Shi Lichen， Lu BaoLiang. Vigilance Estimation by Using Electrooculographic Features［C］ //Proceedings of 32nd annual International Conference of the IEEE Engineering in Medicine and Biology Society. Buenos Aires: IEEE, 2010: 6591-6594.
［12］马家昕.基于眼电信号的警觉度估计模型研究［D］. 上海：上海交通大学，2010.
［13］杨建国.小波分析及其工程应用 ［M］.北京：机械工业出版社，2005:63-74.
［14］Shi Lichen, Lu Baoliang. Offline and online vigilance estimation based on linear dynamical system and manifold learning［C］ // Proceedings of 32nd annual International Conference of the IEEE Engineering in Medicine and Biology Society. Buenos Aires: IEEE, 2010: 6587-6590.
