Driving fatigue can be reflected from the state of the eyes, and eye positioning is the key to judging the state of the eyes. In order to solve the problems of real-time eye positioning and uncertainty of head rotation, this paper proposes an eye positioning method based on dual space: skin color features are used to quickly detect faces in color space, and on this basis, saturation and brightness features are used to quickly detect irises. For the problem of non-open eyes, a custom circular convolution template method is used in grayscale space to locate the eyes. Experimental results show that the positioning accuracy of this method is 94.67%, which can meet the real-time requirements and has certain robustness to wearing glasses, head rotation, and lighting changes. Keywords: skin color detection; iris detection; circular convolution template; eye positioning
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore