Drowsiness detection using Eye-Blink frequency and Yawn count for Driver Alert

Abstract

Several reasons can be sighted for the cause of these road accidents. Few of them include lack of sleep, drunk driving, violation of traffic rules, etc. Amongst them, the state of drowsiness and drunk driving alone contributes to 36% of accidents. Though a number of national schemes and traffic rules have been implemented to avoid these road accidents, it could only bring down the accident rate by 10%. As car accidents are one of the major issues of concern, this paper will be discussing mainly on Drunk driving or drowsiness. In these recent years, various methods have been proposed to implement drowsiness detection based on Hough transforms. Here, in this paper, we have determined a technique to detect drowsiness among car drivers and alert them whenever they tend to sleep. The algorithm is based on eye-blink and yawn frequency. It deals with an eye blink yawn frequency algorithm that uses eye coordinates to keep track of person and determine the open or closed state of the eye and generate an alarm if the driver is drowsy. The yawn count is determined by checking the frequency of yawn count with a minimum threshold value.

Proposal for a System for the Identification of the Concentration of Students Who Attend Online Educational Models

“…The algorithm proposed for the design of the blink counter is developed in Python. In this, several libraries are applied, among which MediaPipe Face Mesh stands out, from which 468 referential points distributed on the person's face are obtained when it is detected [35,36]. Of all the reference points, 12 points are taken to detect the eyes, six for the left eye and six for the right eye, as shown in Figure 3.…”