Deep Residual Learning-Based Convolutional Variational Autoencoder For Driver Fatigue Classification

Driving under the influence of fatigue often results in uncontrollable vehicle dynamics, which causes severe and fatal accidents. Therefore, early warning on onset is crucial to avoid occurrences such kind a disaster. In this paper, authors have investigated novel semi-supervised convolutional variational autoencoder-based classification approach classify state driver. A autoencoder generative network. The proposed discriminative model using autoencoders residual learning. This calculates an intermediate loss base deep features network addition label information training. obtained by method helps training be more effective leads better accuracy driver classification. trained has managed with higher (97%) than other successful models taken into comparison, proving that practical for computing currently available methods.