Driving is one of the daily activities that requires concentration. Many road accidents are said to be caused by a driver’s tiredness, drowsiness, inattention, or distraction. An electroencephalogram (EEG) is a recording of electrical activity in the brain made with electrodes inserted on the head. One of the most successful approaches for identifying drowsiness is the classification of electroencephalogram (EEG) signals. This project aims to alert drowsy drivers by utilizing a brain-computer interface comprised of a brain sensor and a mobile interface. Initially, the sensor’s recorded brain signals will go through several phases, including feature extraction and classification using learning-based algorithms. The outcome will then be turned into visual and audible feedback via a mobile device.

Accurately identify drowsy drivers in order to avoid fatal vehicle accidents.

Detect and classify sleepy drivers from their emitted EEG signals.

Get high classification accuracy using one signal channel.

Denoise raw EEG signals for improved signal processing.

Dataset of the brain signals (alpha-theta) signals responsible for detecting drowsiness. Those signals are acquired using an EEG brain sensor, which measures the electrical activity of the cerebral cortex.

Select the features needed and exclude unneeded features. This will help to minimize classification time and computation.

Improve performance, experiment with various machine learning and deep learning methodologies.

Using a hardware component, collect and prepossess an EEG dataset, then compare the two datasets.