Pavement type recognition based on deep learning

To obtain pavement type information on the basis of related theoretical knowledge on deep learning, this study constructs a deep convolutional neural network model that uses video information on ice, snow, asphalt, and cement pavements collected by a vehicle camera. The data are preprocessed to obtain training and test sets. The training set is used for neural network training, and the accuracy of the model is evaluated with the test set. Results show that the constructed network model can accurately classify four kinds of pavements with an accuracy of 99.5%.