首页> 外文会议>International Conference on Advances in Computing and Information Technology >Electro-oculogram Based Classification of Eye Movement Direction
【24h】

Electro-oculogram Based Classification of Eye Movement Direction

机译:基于电镜的眼睛运动方向分类

获取原文

摘要

In this paper, the direction of eye movement is detected from Electro-oculographic signal using different types of support vector machine classifier. Here, a data acquisition system is designed to collect stimulated Electrooculographic signal. Discrete wavelet transform features of the signal are taken for classification. Eye movement in left and right direction is classified by support vector machine classifier with different kernels. Linear, quadratic, polynomial, radial basis function and multilayer perceptron kernels have been used. In comparison, all of them shown good results but multilayer perceptron performs the best. These classified signals may further be used for control application.
机译:在本文中,通过使用不同类型的支持向量机分类器从电视信号检测眼睛运动的方向。这里,设计数据采集系统以收集刺激的电划线信号。信号的离散小波变换特征是用于分类的。左右方向上的眼睛运动是通过带有不同内核的支持向量机分类器进行分类。已经使用了线性,二次,多项式,径向基函数和多层的Perceptron核。相比之下,所有这些都显示出良好的结果,但多层的感觉器表现了最好的。这些分类信号还可以用于控制应用。

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