Combined Linear Regression and Quadratic Classification Approach for an EEG-Based Prediction of Driver Performance
Electroencephalography (EEG) has been used to reliably and non-invasively detect fatigue in drivers. In fact, linear relationships between EEG power-spectral estimates and indices of driver performance have been found during simplified driving tasks. Here we sought to predict driver performance using linear regression in a more complex paradigm. Driver performance varied widely between participants, often varying greatly within a single driving session. We found that a non-selective linear regression model did not generalize well between periods of stable and erratic driving, yielding large errors. However, prediction errors were significantly reduced by training a linear regression model on stable driving for each participant. To provide a confidence estimate for the stable driving model, a quadratic discriminate classifier was trained to detect the transition from stable to erratic driving from the EEG power-spectra. Combined, the regression model and classifier yielded significantly lower prediction errors and provided improved discrimination of poor driving.
KeywordsEEG Regression Driving Fatigue Power Spectral Density
Unable to display preview. Download preview PDF.
- 1.Treat, J.R., Tumbas, N.S., McDonald, S.T., Shinar, D., Hume, R.D., Mayer, R.E., Stanisfer, R.L., Castellan, N.J.: Tri-level study of the causes of traffic accidents. Report No. DOT-HS-034-3-535-77, TAC (1977)Google Scholar
- 3.National Sleep Foundation. Sleep in America Poll, http://www.sleepfoundation.org/article/sleep-america-polls/2005-adult-sleep-habits-and-styles
- 4.Smith, P., Shah, M., da Vitoria Lobo, N.: Monitoring head/eye motion for driver alertness with one camera. In: Proc.15th International Conference on Pattern Recognition (ICPR 2000), Barcelona, Spain, vol. 4, pp. 636–642 (September 2000)Google Scholar
- 5.Perez, C.A., Palma, A., Holzmann, C.A., Pena, C.: Face and eye tracking algorithm based on digital image processing. In: Proc. IEEE International Conference on Systems, Man, and Cybernetics (SMC 2001), Tucson, Ariz, USA, vol. 2, pp. 1178–1183 (October 2001)Google Scholar
- 6.Popieul, J.C., Simon, P., Loslever, P.: “Using driver’s head movements evolution as a drowsiness indicator. In: Proc. IEEE International Intelligent Vehicles Symposium (IV 2003), Columbus, Ohio, USA, pp. 616–621 (June 2003)Google Scholar
- 12.Desmond, P.A., Matthews, G.: Task-induced Fatigue Effects and Simulated Driving. Quart. Journal of Experimental Psychology 55(2), 659–686 (2002)Google Scholar
- 13.Peiris, M.T.R., Davidson, P.R., Bones, P.J., Jones, R.D.: Detection of lapses in responsiveness from the EEG. Journal of Neural Engineering 8 (2011)Google Scholar
- 14.Stikic, M., Johnson, R.R., Levendowski, D.J., Popovic, D.P., Olmstead, R.E., Berka, C.: EEG-derived estimators of present and future cognitive function. Frontiers in Human Neuroscience 5 (2011)Google Scholar
- 19.Lin, C.T., Wu, R.C., Jung, T.P., Liang, S.F., Huang, T.Y.: Estimating Driving Performance Based on EEG Spectrum Analysis. EURASIP Journal on Applied Signal Processing 19, 3165–3174 (2005a)Google Scholar
- 21.Chuang, S.W., Ko, L.W., Lin, Y.P., Huang, R.S., Jung, T.P., Lin, C.T.: Co-modulatory spectral changes of independent brain processes are correlated with task performance. Neyroimage 62, 1467–1477 (2012)Google Scholar
- 24.Yang, G., Lin, Y., Bhattacharya, P.: A driver fatigue recognition model based on information fusion and dynamic Bayesian network. Information Sciences 108, 1942–1954 (1942)Google Scholar