基于Emotiv Epoc+的眼动信号采集与识别方法研究
|
摘要
使用Emotiv Epoc+脑电波采集头戴设备,对10位被测人员的上、下、左、右、眨眼2次和眨眼3次6种眼球扫视运动信号进行采集,每人每个动作采集5组数据,共获得300组实验数据.对14通道的信号进行带通滤波后,采用短时能量计算方法优选2个最敏感通道,然后对优选出的通道信号采用归一化极值方法进行特征提取,最后采用特征匹配法对信号进行识别.实验结果表明:水平方向眼动识别率达到99.9%,垂直方向眼动识别率达到97.3%,平均识别率达到98.6%.
Electro-Oculogram(EOG)recognition is a research hotpot and one of the main methods in the field of human-computer interface(HCI).EOG data is collected from 10 healthy participants with the headset Emotiv Epoc+,300 groups of experimental data for 6 kinds of eye movement including moving up,moving down,moving left,moving right,blinking twice and blinking 3 times are obtained with 5 groups of data for each eye movement of each person.After filtering the 14 channels signal with bandpass filter,two most sensitive channels are chosen by calculating short-time energy for each channel.Then normalized extremum are extracted as the feature of the EOG.Finally eye movement recognitions are implemented by feature matching.The experiment result indicates that the recognition rate of horizontal direction reaches 99.9%,vertical direction reaches 97.3%,and the average rate reaches 98.6%.
Electro-Oculogram(EOG)recognition is a research hotpot and one of the main methods in the field of human-computer interface(HCI).EOG data is collected from 10 healthy participants with the headset Emotiv Epoc+,300 groups of experimental data for 6 kinds of eye movement including moving up,moving down,moving left,moving right,blinking twice and blinking 3 times are obtained with 5 groups of data for each eye movement of each person.After filtering the 14 channels signal with bandpass filter,two most sensitive channels are chosen by calculating short-time energy for each channel.Then normalized extremum are extracted as the feature of the EOG.Finally eye movement recognitions are implemented by feature matching.The experiment result indicates that the recognition rate of horizontal direction reaches 99.9%,vertical direction reaches 97.3%,and the average rate reaches 98.6%.
引文
[1]BAREA R,BOQUETE L,MAZO M,et al.System for assisted mobility using eye movements based on electrooculography[J].IEEE Transactions on Neural Systems&Rehabilitation Engineering,2002,10(4):209-218.
[2]VENKATARAMANAN S,PRABHAT P,CHOUDHURY S R,et al.Biomedical instrumentation based on electrooculogram(EOG)signal processing and application to a hospital alarm system[C]//International Conference on Intelligent Sensing and Information Processing.Chennai,India:IEEE,2005:535-540.
[3]郑敏敏,高小榕.基于眼电的字符输入系统研究[J].中国生物医学工程学报,2012,31(6):801-806.
[4]TSUJI M.Development of a communication support device controlled by eye movements and voluntary eye blink[J].Ieice Transactions on Information&Systems,2006,89(6):1790-1797.
[5]POSTELNICU C C,GIRBACIA F,TALABA D.EOG-based visual navigation interface development[J].Expert Systems with Applications,2012,39(12):10857-10866.
[6]LYU Z,WU X P,LI M,et al.A novel eye movement detection algorithm for EOG driven human computer interface[J].Pattern Recognition Letters,2010,31(9):1041-1047.
[7]周婷婷,罗志增.基于小波包变换的眼电信号特征提取及分类[J].杭州电子科技大学学报,2014(6):71-75.
[8]KUMAR D,POOLE E.Classification of EOG for human computer interface[J].2002(1):64-67.
[9]HORI J,SAKANO K,SAITOH Y.Development of a communication support device controlled by eye movements and voluntary eye blink[C].Oxford:Oxford University Press,2006:4302-5.
[10]OSTER M,LICHTSTEINER P,DELBRUCK T,et al.A spike-based saccadic recognition system[C].IEEE International Symposium on Circuits and Systems.New Orleans:IEEE,2007:3083-3086.
[11]赵力.语音信号处理[M].北京:机械工业出版社,2003:21-36.
[12]王君,吴小培,吕钊.基于EOG的眼动信息提取与分类研究[J].电子测量技术,2010,33(5):62-65.
[13]杨淑莹.模式识别与智能计算:Matlab技术实现[M].北京:电子工业出版社,2009:33-39.
[14]刘薇.时间序列分析在吉林省GDP预测中的应用[D].长春:东北师范大学,2008.
[15]李盛丹.基于VC++和MATLAB的车型分类及车辆计数系统[D].长春:东北师范大学,2009.
[2]VENKATARAMANAN S,PRABHAT P,CHOUDHURY S R,et al.Biomedical instrumentation based on electrooculogram(EOG)signal processing and application to a hospital alarm system[C]//International Conference on Intelligent Sensing and Information Processing.Chennai,India:IEEE,2005:535-540.
[3]郑敏敏,高小榕.基于眼电的字符输入系统研究[J].中国生物医学工程学报,2012,31(6):801-806.
[4]TSUJI M.Development of a communication support device controlled by eye movements and voluntary eye blink[J].Ieice Transactions on Information&Systems,2006,89(6):1790-1797.
[5]POSTELNICU C C,GIRBACIA F,TALABA D.EOG-based visual navigation interface development[J].Expert Systems with Applications,2012,39(12):10857-10866.
[6]LYU Z,WU X P,LI M,et al.A novel eye movement detection algorithm for EOG driven human computer interface[J].Pattern Recognition Letters,2010,31(9):1041-1047.
[7]周婷婷,罗志增.基于小波包变换的眼电信号特征提取及分类[J].杭州电子科技大学学报,2014(6):71-75.
[8]KUMAR D,POOLE E.Classification of EOG for human computer interface[J].2002(1):64-67.
[9]HORI J,SAKANO K,SAITOH Y.Development of a communication support device controlled by eye movements and voluntary eye blink[C].Oxford:Oxford University Press,2006:4302-5.
[10]OSTER M,LICHTSTEINER P,DELBRUCK T,et al.A spike-based saccadic recognition system[C].IEEE International Symposium on Circuits and Systems.New Orleans:IEEE,2007:3083-3086.
[11]赵力.语音信号处理[M].北京:机械工业出版社,2003:21-36.
[12]王君,吴小培,吕钊.基于EOG的眼动信息提取与分类研究[J].电子测量技术,2010,33(5):62-65.
[13]杨淑莹.模式识别与智能计算:Matlab技术实现[M].北京:电子工业出版社,2009:33-39.
[14]刘薇.时间序列分析在吉林省GDP预测中的应用[D].长春:东北师范大学,2008.
[15]李盛丹.基于VC++和MATLAB的车型分类及车辆计数系统[D].长春:东北师范大学,2009.