Automatic classification methods for detecting drowsiness using wavelet packet transform extracted time-domain features from single-channel EEG signal

•EEG sub-bands like Delta, Theta, Alpha, Beta, and Gamma are extracted in time-domain using the WPT signal processing method, where the best wavelet function is used and appropriate decomposition levels.•A method has been proposed to select the potential features for the detection of onset of drowsiness.•Performance of multiple classifiers have been compared in detecting the onset of drowsiness and the result is validated with the subject wise, cross subject wise, and combine subject wise data set.•A detailed qualitative and quantitative analysis for the performance of classifiers is carried out using standard EEG sleep dataset and virtually simulated driving driver dataset.