基于脑电同源样本捆绑法的情绪识别研究<sup>^</sup>

doi:10.3969/j.issn.0258-8021. 2016. 03.003

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (921 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要近年来,越来越多的研究者投入到基于脑电的情绪识别研究中,新理论方法的提出使得情绪的识别准确率有了进一步提高。然而,一旦走向实际应用,识别率大幅度下降,建立高精确率的情绪识别模型仍面临巨大的挑战。一个非常重要的原因是同源样本携带的非情绪信息使情绪识别模型更容易识别测试集的样本,使得传统的情绪识别方法出现识别正确率虚高的问题。为此,提出一种新的样本划分方法-同源样本捆绑法,并对10名被试开展情绪诱发实验,利用情绪图片和视频诱发平静、愉悦、难过和恶心等4种情绪状态。提取6个频段的脑电信号(θ∶4~8 Hz,α:8~12 Hz,β1: 13~18 Hz,β2:18~30 Hz,γ1:30~36 Hz和γ2:36~44 Hz)的功率谱特征,然后利用支持向量机进行情绪识别,并对比传统随机方法与同源样本捆绑法的情绪识别效果。结果显示,对于图片诱发任务,随机法四分类平均正确率为43.92%,而捆绑法只得到34.15%的平均正确率;在视频诱发任务下,随机法和捆绑法分别得到94.45%和37.88%的平均正确率,随机法显著高于捆绑法,证明传统的样本随机划分方法会带来虚高情绪识别率。最后利用基于递归筛选的支持向量机(SVM-RFE)算法剔除基于同源样本捆绑法下的非情绪特征,提高情绪识别正确率,图片和视频诱发任务分别得到76.22%和72.53%的平均识别正确率。综上,所提出的同源样本捆绑法可以剔除非情绪因素的影响,避免传统的样本划分方法带来的虚高情绪识别率,是情绪识别从理论研究走向实际应用重要且必要的一步。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	刘爽
	仝晶晶
	杨佳佳
	綦宏志
	明东

关键词 ：脑电, 情绪识别, 支持向量机(SVM), 同源样本捆绑法

Abstract：There are numerous studies measuring brain emotional status by analyzing EEGs under the emotional stimuli that have occurred, and acceptable accuracies were obtained in existing researches. However, emotion classification model would be challenged when it was applied in practical application. Shared non-emotional information in homologous samples may make the classification model easier to recognize the samples in the testing set, resulting in higher accuracies in EEG-based emotion recognition. In the pattern recognition, we proposed a new sample-divided method, named integrating homologous samples method, where the homologous samples were either used to build a classifier, or to be tested. In this paper, affective pictures and videos were used to elicit four emotional states of neutral, happy, sad, and disgust from 10 subjects, and EEG signals were recorded during the pictures or videos display. PSD were extracted from EEGs of 6 frequency bands (θ: 4-8 Hz, α: 8-12 Hz, β1: 13-18 Hz, β2: 18-30 Hz, γ1: 30-36 Hz and γ2: 36-44 Hz), and then sent to a SVM for classification. The results showed that the classification accuracy was much lower for the integrating homologous samples method (IHSM) than for the traditional dividing the samples randomly (TDSR). For the image evoked task, 43.92% and 34.15% were obtained by TDSR and IHSM, respectively. There were 94.45% and 37.88% for video evoked task. SVM-RFE was employed to select emotional features and improved the classification rates to 76.22% and 72.35% for these two tasks. The proposed method avoided the overinflated accuracies brought by the traditional method, and handle this problem is an important and necessary step from the laboratory to the practical application.

Key words： electroencephalography (EEG) emotion recognition support vector machine (SVM) integrating homologous samples method

收稿日期: 2016-01-10

PACS:

R318

基金资助:国家自然科学基金重大研究计划(91520205);国家自然科学基金(81571762, 31500865)

通讯作者: E-mail: richardming@tju.edu.cn

引用本文:

刘爽, 仝晶晶, 杨佳佳, 綦宏志, 明东. 基于脑电同源样本捆绑法的情绪识别研究^{^}[J]. 中国生物医学工程学报, 2016, 35(3): 272-277.
Liu Shuang, Tong Jingjing, Yang Jiajia, Qi Hongzhi, Ming Dong. Study on Emotion Recognition with Integrating EEG Homologous Samples Method. Chinese Journal of Biomedical Engineering, 2016, 35(3): 272-277.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021. 2016. 03.003 或 http://cjbme.csbme.org/CN/Y2016/V35/I3/272

[1] Blanchette I, Richards A. The influence of affect on higher level cognition: A review of research on interpretation, judgement, decision making and reasoning [J]. Cognition & Emotion, 2010, 24(4): 561-595.
[2] Picard RW, Vyzas E, Healey J. Toward machine emotional intelligence: Analysis of affective physiological state [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(10): 1175-1191.
[3] Liu S, Zhang D, Xu M, et al. Randomly dividing homologous samples leads to overinflated accuracies for emotion recognition [J]. International Journal of Psychophysiology Official Journal of the International Organization of Psychophysiology, 2015, 96(1): 29-37.
[4] Verma GK, Tiwary US. Multimodal fusion framework: A multiresolution approach for emotion classification and recognition from physiological signals [J]. NeuroImage, 2014, 102:162-172.
[5] Wang XW, Nie D, Lu BL. Emotional state classification from EEG data using machine learning approach [J]. Neurocomputing, 2014, 129:94-106.
[6] Khaliliardali Z. Emotion detection using brain and peripheral signals[C]// International Biomedical Engineering Conference. Cairo: IEEE 2008:1-4.
[7] Soleymani M, Pantic M, Pun T. Multimodal emotion recognition in response to videos [J]. IEEE Transactions on Affective Computing, 2012, 3(2): 211-223.
[8] Moltó J, Montaés S, Poy R, et al. Un nuevo método para el estudio experimental de las emociones: El International Affective Picture System (IAPS) [J]. Revista De Psicología General Y Aplicada, 1999, 52(1):55-87.
[9] Cohen B, Bravo-Fernandez E, Sances Jr A. Automated electroencephalographic analysis as a prognostic indicator in stroke [J]. Medical and Biological Engineering and Computing, 1977, 15(4): 431-437.
[10] 万柏坤, 朱欣, 杨春梅, 等. ICA去除EEG中眼动伪差和工频干扰方法研究 [J]. 电子学报, 2003, 31(10): 1571-1574.
[11] 陈海英. AR模型功率谱估计常用算法的性能比较 [J]. 漳州师范学院学报:自然科学版, 2009, 22(1): 48-52.
[12] Hidalgo-Mu Oz AR, L Pez MM, Santos IM, et al. Application of SVM-RFE on EEG signals for detecting the most relevant scalp regions linked to affective valence processing [J]. Expert Systems with Applications, 2013, 40(6): 2102-2108.
[13] Chang CC, Lin CJ. LIBSVM: A library for support vector machines [J]. Acm Transactions on Intelligent Systems & Technology, 2006, 2(3): 389-396.
[14] 白艳茹. 用于生物特征识别的多范式诱发脑电个体差异性研究 [D]. 天津:天津大学, 2012.
[15] Brown L, Grundlehner B, Penders J. Towards wireless emotional valence detection from EEG.[C]// International Conference of the IEEE Engineering in Medicine & Biology Society. Boston: IEEE, 2011:2188-2191.
[16] Nie D, Wang XW, Shi LC, et al. EEG-based emotion recognition during watching movies[C]// International IEEE/EMBS Conference on Neural Engineering. Antalya: IEEE Computer Society, 2011:667-670.
[17] Peng H, Long F, Ding C. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(8): 1226-1238.
[18] Ke Y, Qi H, Zhang L, et al. Towards an effective cross-task mental workload recognition model using electroencephalography based on feature selection and support vector machine regression [J]. International Journal of Psychophysiology, 2015, 98(2): 157-166.
[19] Christensen JC, Estepp JR, Wilson GF, et al. The effects of day-to-day variability of physiological data on operator functional state classification [J]. Neuroimage, 2012, 59(1): 57-63.