 The proposed method uses a combination of EEG and iBlink data to detect driver fatigue. It employs a moving standard deviation algorithm to identify iBlink intervals, which are then used to filter out iBlink-related features from the EEG signal. This filtered EEG signal is further decomposed into subbands, and various linear and non-linear features are extracted. These features are then selected by the neighborhood components analysis and fed to a classifier to distinguish between fatigue and alert driving. The proposed method was tested on two data sets, and its performance was compared against other methods. The results indicate that the proposed method is reliable for detecting driver fatigue in real-world scenarios. This article was authored by Muhammad Shababdi, Martin Bermbrandt, Erfan Nasiri, and others.