Distracted driving recognition method based on deep convolutional neural network

In recent years, traffic accidents caused by the distracted driving have been on the rise with the popularization of smart phones. How to correctly identify whether the driver is in a distracted driving state and to provide the necessary warnings for the driver to avoid potential safety risks has become one of the most concerned issues. In this paper, a distracted driving recognition method based on deep convolutional neural network is proposed for the driving image data captured by the in-vehicle camera. This method uses the PCA technology to whiten the driving image, which reduces the redundancy and correlation of the pixel matrix. At the same time, a multi-layer CNN network is constructed in the model and the key parameters of the input layer, convolution layer, pooling layer, fully connected layer and output layer are optimized as well. The results of experimental analysis show that the accuracy of the proposed method can reach 97.31%, which is higher than that of the existing machine learning algorithms. Therefore, the proposed method is effective in improving the accuracy of distracted driving recognition.