Abstract: In recent days road accidents are increasing due to human errors such as drowsy driving, distractions and drunken driving have become the leading factors for the accidents. In this paper, an effort is made to build a real time system to detect drowsiness as well as the alcohol detection with subsequently alerting the car driver. The main aim of this proposed system is to reduce the number of accidents due to driver’s drowsiness and alcohol intake to increase the transportation safety. This proposed system contains a 5-megapixel digital USB camera, Raspberry-pi loaded with Raspbian-OS, Alcohol sensor (MQ-3) is used to detect the breath alcohol content in percentage. If the intoxication level is above the threshold, buzzer gets triggered ON with a message being sent to the authorized person and simultaneously the engine is stopped using a relay. This system will perform some tasks like the alarm notification and switching off the car power source. The detection of drowsiness is carried out using an OpenCV-source, Haar cascade classifiers and canny edge detection.

Keywords: Raspberry pi, MQ3 sensor, USB camera, Haar cascade classifier, Canny edge detection, buzzer, relay, External ADC.

| DOI: 10.17148/IARJSET.2021.8774