Table of Contents Author Guidelines Submit a Manuscript
Mathematical Problems in Engineering
Volume 2015 (2015), Article ID 832621, 13 pages
http://dx.doi.org/10.1155/2015/832621
Research Article

Blink Number Forecasting Based on Improved Bayesian Fusion Algorithm for Fatigue Driving Detection

1School of Information and Control, Nanjing University of Information Science & Technology, Nanjing 210044, China
2Jiangsu Key Laboratory of Big Data Analysis Technology, Nanjing 210044, China
3School of Computer and Software, Nanjing University of Information Science & Technology, Nanjing 210044, China
4The NEXTRANS Center, Purdue University, West Lafayette, IN 47907, USA
5School of Electronic and Information Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, China

Received 9 January 2015; Accepted 1 May 2015

Academic Editor: Yakov Strelniker

Copyright © 2015 Wei Sun et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

An improved Bayesian fusion algorithm (BFA) is proposed for forecasting the blink number in a continuous video. It assumes that, at one prediction interval, the blink number is correlated with the blink numbers of only a few previous intervals. With this assumption, the weights of the component predictors in the improved BFA are calculated according to their prediction performance only from a few intervals rather than from all intervals. Therefore, compared with the conventional BFA, the improved BFA is more sensitive to the disturbed condition of the component predictors for adjusting their weights more rapidly. To determine the most relevant intervals, the grey relation entropy-based analysis (GREBA) method is proposed, which can be used analyze the relevancy between the historical data flows of blink number and the data flow at the current interval. Three single predictors, that is, the autoregressive integrated moving average (ARIMA), radial basis function neural network (RBFNN), and Kalman filter (KF), are designed and incorporated linearly into the BFA. Experimental results demonstrate that the improved BFA obviously outperforms the conventional BFA in both accuracy and stability; also fatigue driving can be accurately warned against in advance based on the blink number forecasted by the improved BFA.