The Drowsy Testing System According to Deep-Learning

Abstract

The danger of drowsy driving to public transportation is merely inferior to drunk driving; therefore this research principally defines a testing method of drowsy driving. Thus, this paper optimizes the algorithm according to the YOLO v3. Through attaching an attention device, the network can concentrate on the eyes and mouth of the person appropriately, and improve the network composition to advance the testing speed. Via measuring the rate of closed eyes, the frequency of yawning, and the detection of closed eyes various times in a unit time, the three criteria utilized to discover whether the driver has a drowsy driving condition. After examination, associated with other testing techniques, the method in this article has a noticeable improvement in testing precision and rate.