基于共空间模式的运动想像信号特征提取方法研究(任务书,开题报告,论文说明书14000字)
摘要
大脑是人体的控制中枢，负责调节控制众多生理活动。通过研究大脑的构造与其运作机理，可以促进脑科学，人工智能，疾病的诊断与治疗等领域的发展。通过使用脑机接口识别大脑活动所产生的信号，从而建立大脑与外界的沟通桥梁，让大脑直接与外界设备进行交流，完成命令的传达。在脑机接口领域使用的最为广泛的大脑信号是脑电信号，脑电信号中使用最广泛的是运动想象信号。
本文针对运动想象信号的识别，建立了脑电信号预处理、脑电特征取与脑电特征分类等一系列处理步骤，完成了运动想象信号的特征提取和识别。本文使用巴特沃斯滤波器对脑电信号进行滤波，之后使用共空间模式和局部尺度特征分解和傅里叶变化提取空域和频域特征，最后通过线性判别式分类器和支持向量机分类器进行分类。
关键词：脑机接口运动想象空间模式局部尺度特征分解
Abstract
The brain is the control center of the human body, responsible for regulating and controlling many physiological activities. By studying the structure of the brain and its mechanism of operation, can promote brain science, artificial intelligence, disease diagnosis and treatment and other fields of development. Through the use of brain and brain interface to identify the brain generated by the signal, so as to establish a bridge between the brain and the outside world, so that the brain directly with the external equipment to communicate, to complete the command to convey. The most widely used brain signal used in the field of brain and brain interfaces is the EEG signal, and the most widely used in EEG signals is the motion imagination signal.
In this paper, a series of processing steps such as EEG signal preprocessing, EEG feature acquisition and EEG feature classification are established for the recognition of motion imagination signal, and the feature extraction and recognition of motion imagination signal are completed. In this paper, Bartwone filter is used to filter the EEG signal, and then the spatial and frequency domain features are extracted by using the co-spatial model and the local scale feature decomposition and Fourier transform. Finally, the linear discriminant classifier and the support vector machine classifier classification.
keyword:Brain-Computer Interface, Motor imagery, Common Spital Pattern, Local Characteristic-scale Decomposition
目录
摘要 i
Abstract ii
1绪论 1
1.1研究背景及意义 1
1.1.1研究背景 1
1.1.2 研究意义 2
1.2脑机接口技术 3
1.2.1脑机接口概述 3
1.2.2脑机接口国内外研究现状 4
1.3论文章节安排 6
2脑电信号概述 7
2.1脑电信号的产生与特点 7
2.2脑电信号分类 8
2.3脑电信号的分析方法 8
3数据描述与预处理 10
3.1数据描述 10
3.2数据预处理 10
3.2.1数据的限幅 11
3.2.2数据的带通滤波 11
4脑电信号特征的提取 12
4.1共空间模式 12
4.2局部尺度特征分解 14
5脑电信号特征的分类 18
5.1线性判别式分类器 18
5.2支持向量机分类器 20
6总结与展望 22
6.1总结 22
6.2展望 22
参考文献 24
致谢 26 |
