一种用于脑机接口的模式识别方法

基于脑电的脑机接口(BCI)是在人脑和计算机或其他电子设备之间建立的全新对外信息交流和控制技术,是一种不依赖于常规大脑信息输出通路(外围神经和肌肉组织)的脑机通讯系统.及时有效地提取和识别与运动想象有关的脑电模式可以帮助运动功能受损的病人建立一种与外界沟通的新途径.论文基于传统的特征提取方法--时频特征组合法,经过滑动窗优化,获取最佳时间段的时域均值和最佳频率段的频域功率谱均值,以此作为特征向量.基于该特征向量,用径向基概率神经网络对脑电信号进行分类.实验结果表明,该方法能够有效地提高脑电识别率,具有应用价值.

A pattern recognition method in brain-computer interface

The EEG-based brain-computer interface (BCI) is a novel technology, which provides a wholly new channel between human being brains and computers or electronic equipments instead of the normal output pathways of peripheral nerves and muscles. Quick and correct classification of these event-related EEG pattern can be used to provide patients with severe paralysis a new communication channel between patients and environments. In this paper, based on the traditional method, EEG features from time domain and frequency domain are extracted, and these features are then optimized by the way of sliding window. Finally, a radial basis probabilistic neural network (RBPNN) is used to classify testing samples based on these optimized features. According to the analysis and experiment, this method improves the correct rate of classification effectively, and has the practicability in the application of brain-computer interface technique.