A Partial Least Squares Regression-Based Fusion Model for Predicting the Trend in Drowsiness

This paper proposes a new technique of modeling driver drowsiness with multiple eyelid movement features based on an information fusion technique - partial least squares regression (PLSR), with which to cope with the problem of strong collinear relations among eyelid movement features and, thus, predicting the tendency of the drowsiness. With a set of electro- oculogram signals measured in an experiment conducted in Sweden, 14 typical eyelid movement features are first extracted. Then, statistical analyses from 20 subjects indicate that the eyelid movement parameters can characterize a driver's degree of drowsiness. The intrinsic quantitative relationships between eyelid movement features and driver drowsiness degree are modeled by PLSR analysis. The developed model provides a framework for integrating multiple sleepiness features together and defining the contribution of each feature to the decision and prediction result. The predictive precision and robustness of the model thus established are validated, which show that it provides a novel way of fusing multifeatures together for enhancing our capability of detecting and predicting the state of drowsiness.