Abstract:

This paper reports a face gesture recognition method based on EEG sensor, which has advantages of good sense of experience and high accuracy compared with traditional methods based on visible light, depth camera and EMG. By analyzing the interference signals characteristics of facial movements on EEG sensors, the system design is given, and a real-time pattern recognition and classification algorithm based on support vector machine is described. Finally, experimental results prove that high face gesture recognition rate can be obtained under the condition of few samples by use of the proposed method.

Key words: face gesture, Electroencephalogram（EEG）, Electromyogram（EMG）, pattern recognition, Support Vector Machine（SVM）

摘要：

提出了一种利用脑电传感器进行面部动作识别的方案。相比传统的可见光、深度相机、肌电方案，该方案具有体验感好、识别准确率高的特点。通过分析面部动作对脑电传感器产生的干扰信号特点，给出了系统设计，描述了基于支持向量机进行模式识别与分类的算法，最后通过实验验证，证明了该方案可在少样本的条件下，实现高精度的面部动作识别。

关键词: 面部动作, 脑电, 肌电, 模式识别, 支持向量机