Design and Implementation of Drowsiness Detection System Based on Standard Deviation of Lateral Position

Abstract

Drowsiness can decrease concentration while driving. The decreased concentration while driving makes a slower response in making decisions about road conditions. Roads with monotonous characteristics and low traffic density can also reduce driver alertness. This condition can lead to traffic accidents. However, drivers sometimes do not realize that they are sleepy. Therefore, a system is needed to detect drowsiness and provide warnings to drivers. To overcome that problem, we propose a system for detecting and alerting a driver's sleepiness based on the deviation from the lane. This system is designed using the Python programming language and the OpenCV library. Based on the results of experiments that we have done, the calculation of the Standard Deviation of the Lateral Position accuracy reaches 96.54% during the day and 91.08% at night.