Comparative study on existing driver drowsiness detection systems

What is it about?

Driver drowsiness results in thousands of fatalities every year. After the advent of AI, there is a regenerated interest in the field to detect the alertness levels of drivers. This study compares all the available systems from a user’s frame of reference–low cost, easy to use and no extra setup. We look at all the previous contributions and list out important common parameters that should be included in data inputs for contact and non-contact systems and draw out comparisons to which would serve better with contextual information.

Featured Image

Photo by Thái An on Unsplash

Photo by Thái An on Unsplash

Why is it important?

This paper's major goal is to give end users a scale of uniformity and relevance so they can analyse and evaluate the systems using the provided metrics. Contact techniques include EEG, EOG, HRV, PRV among others. Non-contact methods majorly include CNN, deep learning, SVM, linear regression, kernelized correlation filters. We highlight the most significant aspects of any good system

Perspectives