Abstract
Aiming at the problem of serious traffic accident caused by driver fatigue driving, a method of identifying facial fatigue expression by using convolutional neural network is proposed in this paper. Combined with image processing technology, the fatigue expression features are trained and learned to realize the identification of driver fatigue expression. Compared with the traditional detection method based on visual feature, the neural network model constructed in this paper can extract the fatigue expression features more fully. In the training test of the sample pictures, the accuracy rate can reach 95.4%, and the running time is only 0.89s. It is confirmed that the method proposed in this paper is effective and feasible.