EEG Based Driver Drowsiness Detection system

日期 : Sep 2014 - May 2015

In the recent world, race of survival and competency has caused shortage of time .Because of this, mental fatigue became common in almost every adult .Mental fatigue causes many problems and of those many, drowsiness is major one. The term “drowsy” is synonymous with sleepy, which simply means an inclination to fall asleep, so it is "Transition from awake to asleep". In recent years, drowsiness has been one of the major causes of road accidents and air crashes which lead to severe physical injuries, deaths and significant economic losses. Statistics indicated the need of a reliable driver drowsiness detection system. Although many companies already worked on it but, those systems are too much expensive. We choose EEG (electroencephalogram) signals to detect driver’s drowsiness in order to avoid fatal accidents. Our purpose also involved to develop a low cost, small size, flexible and robust system. Our project title was “EEG-Based Driver Drowsiness Detection system”. When human Brain is in drowsy state its EEG signal contains 4 Hz to 8Hz frequency band. We developed system with the help of instrumentation amplifiers and electrodes to capture that frequency band. Electrodes were used to get signal from brain. Electrodes placement involved standard 10-20 system, used widely for Brain-computer interface (BCI). We used three stages of amplification to amplify obtained signal. After each amplification, a high-pass filter was used, which was designed to remove DC offsets. It's cut off frequency was about 0.16Hz. After third stage of amplification ,there existed digital part. Digital part contained Arduino Mega 2560 microcontroller, which observed continuously incoming signal with baud rate of 9600. It was set to trigger an alarm on the frequency band of 4 Hz to 8 Hz.