Detection of Drowsiness from Facial Images in Real-Time Video Media using Nvidia Jetson Nano

Drowsiness affects people everyday life that leads to inefficient work or car accidents due to insomnia resulting damage to life and property. Especially, it can become worse when it happens to public transportation. As the author works in Department of Land Transport (DLT), drowsiness is a factor that causes car accident. Therefore, due to this concern, effective detection of drowsiness system is performed. The purpose of this paper is to develop Algorithm to analyze face structure from taking video media and detecting drowsiness. Also, the system is conducted to examine Algorithm's efficiency. The result shows that using facial landmarks can help to generate eyes and mouth component effectively, which can help to create equations to analyze drowsiness correctly by using Nvidia Nano Jetson. Nvidia Jetson Nano is a tool that accurately evaluates image by tracking closing eye motions more than 35 FPS or 1.5 seconds, and yawning or opening mouth motion more than 50 FPS equals to 2 seconds. It indicates the possibility of sleepiness which notifies to drivers. Thus, it is suggested that government should support the system and take to Thailand policy in order to reduce car accident from insomnia.