中文摘要：

眼电图（electrooculogram，EOG）是一种最常用的眼球运动记录技术，但一直缺乏从EOG扫视信号中提取二维人眼位置信息的有效方法。本文提出了利用支持向量机对二维人眼位置信息进行提取的新方法。建立了相应的模型。研究结果表明，基于支持向量机的提取方法是准确、有效的，构筑EOG-SVR系统是可行的。

英文摘要：

EOG is widely used to record eye movement. However, an efficient means of extracting two-dimensional （2D） eye position data from EOG saccade signal is lacking. A new method to extract the 2D eye position data using support vector machine was proposed and the corresponding model was built. The results show that the extracting method based on support vector machine is accurate and efficient and constructing the EOG-SVR system is practical.