Monitoring of Micro-sleep and Sleepiness for the Drivers Using EEG Signal
Abstract: Nowadays sleepiness at the wheel is a problem that affects to the society at large rather more than it at first seemed. The purpose of this thesis is to detect sleepiness and micro-sleeps, which are the source that subsequently leads to drowsiness, through the study and analysis of EEG signals. Initially, raw data have some artifacts and noises which must be eliminated through a band pass filter. In this thesis EEG signals from different persons are analyzed and the feature extraction is carried out through the method Fast Fourier Transform (FFT). After that, the signals are classified to get the best result. To do this, the method Support Vector Machine (SVM) is used where the feature vectors, which have been extracted previously, are the input. The data are trained and tested to get a result with an accuracy of 77% or higher. It shows that EEG data could be used helping experts in the development of an intelligent system to classify different sleeping conditions i.e., micro-sleep and sleepiness.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)