首页> 外文期刊>International Journal of Innovative Computing Information and Control >PERSONAL IDENTIFICATION BY STEADY-STATE VISUAL EVOKED POTENTIALS BASED ON VOTING PROCESS BY MAHALANOBIS DISTANCE
【24h】

PERSONAL IDENTIFICATION BY STEADY-STATE VISUAL EVOKED POTENTIALS BASED ON VOTING PROCESS BY MAHALANOBIS DISTANCE

机译:基于Mahalanobis距离的投票过程稳态视觉诱发电位的个人识别

获取原文
获取原文并翻译 | 示例
           

摘要

In this study, we attempt personal authentication using steady-state visual evoked potentials (SSVEPs), a type of evoked electroencephalographic brain activity generated by external visual stimuli. We recorded SSVEPs induced by a 5-Hz light stimulus from 16 participants. Recordings were from four electrodes, and multivariate autore-gressive coefficients were used as features for authentication. Subsequently, classification based on the Mahalanobis distance was performed using the multivariate autoregressive coefficients. The authentication was performed under two conditions: 1) considering only those registered in the database and 2) considering unregistered third persons. Furthermore, we examined the effectiveness of the voting process, in which the person is authenticated on the basis of voting on multiple test samples. When authentication proceeded using only registrants, accuracy was higher when four electrodes were used than when two electrodes were used, and the voting process improved identification accuracy to 96.4%. Authentication using unregistered third persons was also more accurate with four electrodes than with two electrodes. The accuracy was 75.8% when voting was not applied and reached 90.4% when voting was applied. Our method, which can authenticate an individual in just a few seconds, is considered effective for realizing personal authentication by SSVEP.
机译:在本研究中,我们尝试使用稳态视觉诱发电位(SSVEP),一种由外部视觉刺激产生的诱发脑电图脑活动的个人认证。我们从16名参与者录制了由5 Hz光刺激引起的SSVEPS。录制来自四个电极,多变量自动加法系数用作认证的功能。随后,使用多变量自回归系数来执行基于Mahalanobis距离的分类。认证在两个条件下执行:1)仅考虑在数据库中注册的那些,而2)考虑未注册的第三人。此外,我们检查了投票过程的有效性,其中人在多次测试样本上投票进行了认证。当使用仅注册人进行认证时,使用四个电极的使用比使用两个电极时,精度更高,并且投票过程将识别精度提高至96.4%。使用未注册的第三人员的认证也比使用两个电极更准确。当申请投票时,准确性为75.8%,达到投票时达到90.4%。我们的方法可以在几秒钟内验证个人,被认为是通过SSVEP实现个人认证的有效性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号