DOI QR코드

DOI QR Code

감성판별을 위한 생체신호기반 특징선택 분류기 설계

The Design of Feature Selection Classifier based on Physiological Signal for Emotion Detection

  • 이지은 (연세대학교 일반대학원 생체공학협동과정) ;
  • 유선국 (연세대학교 의과대학 의학공학교실)
  • Lee, JeeEun (Graduate School of Biomedical Engineering, Yonsei University) ;
  • Yoo, Sun K. (Department of Medical Engineering, Yonsei University College of Medicine)
  • 투고 : 2013.07.26
  • 발행 : 2013.11.25

초록

감성은 학습, 행동, 의사결정, 상호대화를 포함한 인간의 일상생활에 중요한 요소이다. 본 논문에서는 시스템의 복잡도를 줄이기 위하여 생체신호로부터 최소한의 중요한 특징만을 추출하여 사용하는 감성 분류기를 설계하고자 한다. 생체신호는 맥파, 피부온도, 피부전도도, 뇌파신호(전두엽, 두정엽)를 사용하였으며, 4가지 감정(보통, 슬픔, 공포, 행복)은 영화 관람을 통하여 유도하였다. 측정한 생체신호로부터 추출한 24개의 특징으로부터 최적의 특징 집합의 결정은 서포트벡터머신 기반 적합도 함수를 사용하는 유전알고리즘을 적용하였다. 최적의 4감정 분류 정확도는 96.4%이었으며, 서포트벡터머신만을 사용하였을 경우보다 17% 높았다. 선택된 최소에러 특징은 맥파 심박변이도의 평균, NN50, 맥파 유도 맥파 전달 시간의 평균, 피부전도도의 평균과 두정엽 뇌파의 ${\delta}$, ${\beta}$ 주파수 대역에너지였다. 실험을 통하여 두정엽 뇌파, 맥파, 피부전도도의 조합이 고정밀 감정 장비에 적합하였으며, 79% 성능을 보인 맥파와 피부전도도의 조합이 간단한 감성장비에 적절하게 적용할 수 있다.

The emotion plays a critical role in human's daily life including learning, action, decision and communication. In this paper, emotion discrimination classifier is designed to reduce system complexity through reduced selection of dominant features from biosignals. The photoplethysmography(PPG), skin temperature, skin conductance, fontal and parietal electroencephalography(EEG) signals were measured during 4 types of movie watching associated with the induction of neutral, sad, fear joy emotions. The genetic algorithm with support vector machine(SVM) based fitness function was designed to determine dominant features among 24 parameters extracted from measured biosignals. It shows maximum classification accuracy of 96.4%, which is 17% higher than that of SVM alone. The minimum error features selected are the mean and NN50 of heart rate variability from PPG signal, the mean of PPG induced pulse transit time, the mean of skin resistance, and ${\delta}$ and ${\beta}$ frequency band powers of parietal EEG. The combination of parietal EEG, PPG, and skin resistance is recommendable in high accuracy instrumentation, while the combinational use of PPG and skin conductance(79% accuracy) is affordable in simplified instrumentation.

키워드

참고문헌

  1. Mu Li and Bao-Liang Lu, "Emotion Classification Based on Gamma-band EEG", 31th Annual International Conference of the IEEE EMBS, pp.1323-1326, Minnesota, USA, Sep 2009.
  2. Jerritta Selvaraj, Murugappan Murugappan, Khairunizam Wan and Sazali Yaacob, "Classification of emotional states from electrocardiogram signals: a non-linear approach based on hurst," BioMedical Engineering Online, 2013.
  3. 안형철, 최진영, "[특집]지능로봇의 감성행동 기술 동향," Journal of the IEEK, vol.32(1), pp.50-59, 2005.
  4. Jonghwa K and Ande E, "Emotion Recognition Based on Physiological Changes in Music Listening," Pattern Anal. Mach. Intell., IEEE Transact, pp.2067-2083 2008.
  5. Ekman P. and Friesen WV, "Universals and Cultural Differences in the Judgments of FAcial Expressions of Emotion," J pers Soc Psychol, pp.712-714, 1987.
  6. Lang PJ, "The Emotion Probe: Studies of Motivation and Attention," Am Psychol, pp.372-385, 1995.
  7. LIU Guang-Yuan and HAO Min, "Emotion Recognition of Physiological Signals Based on Adaptive Hierarchical Genetic Algorithm," 2009 World Congress on Computer Science and Information Engineering, pp.670-674, 2009.
  8. Xiaowei Niu, Liwan Chen and Qiang Chan, "Research on Genetic Algorithm based on Emotion recognition using physiological signals," ICCP Proceedings, pp.614-618, 2011.
  9. WooJin Chio, "A classification analysis of negative emotion based on PPG signal using Fuzzy-GA,", July, 2007.
  10. John Aleen, "Photoplethysmography and its application in clinical physiological measurement," Physiological measurement, vol.28(3), pp.1-39, 2007. https://doi.org/10.1088/0967-3334/28/1/001
  11. Pierre Rainville, Bechara A, Naqvi N and Damasio AR, "Basic emotions are associated with distince patterns of cardiorespiratory activity," International Journal of Psychophysiology, pp.5-18, 2006.
  12. Foteini Agrafioti, "ECG Pattern Analysis for Emotion Detection", IEEE Transactions on Affective Computing, vol.3, 2012.
  13. Malik M., "Measurement of heart rate variability," Heart Rate Variability. Armonk, pp.33-132, 1995.
  14. Zhai, J. and Barreto, A., "Stress Detection in Computer Users Based on Digital Signal Processing of Noninvasive Physiological Variables," The 28th Annual International Conference Engineering in Medicine and Biology Society, pp.1355-1358, 2006.
  15. Guanghua Wu, Guangyuan Liu and Min Hao, "The analysis of emotion recognition from GSR based on PSO," 2010 International Symposium on Intelligence Information Processing and Trusted Computing, pp.360-363, 2010.
  16. Murugappan Murugappan, Nagarajan Ramachandran and Yaacob Sazali, "Classification of human emotion from EEG using discrete wavelet transform," J. Biomedical Science and Engineering, vol.3, pp.390-396, April, 2010. https://doi.org/10.4236/jbise.2010.34054
  17. Edgar Osuna, Robert Freund, and Federico Girosi, "Support Vector Machines: Training and Applications," C.B.C,I. Paper No.144, March, 1997.
  18. Chih-Chung Chang and Chih-Jen Lin, "Libsvm: a library for support vector machines," ACM Transactions on Intelligent Systems and Technology (TIST), vol.2, pp.3-27, 2011.
  19. David E. Goldberg, "Genetic Algorithm in Search, Optimization, and Machine Learning," Addison_Wesley Professional, 1 edition, 1989.
  20. Melanie Dumas, "Emotional Expression Recognition using Support Vector Machines," In Proceedings of International Conference on Multimodal Interfaces, 2001.
  21. 권오상, "감성로봇 현황과 추세," Journal of the IEEK, vol.28(12), pp.18-25, 2001.

피인용 문헌

  1. 심박 정보 기반 위치 정보 융합형 감정 추론 어플리케이션 개발 vol.8, pp.8, 2013, https://doi.org/10.15207/jkcs.2017.8.8.083