A New Algorithm for Eye Location in Driver Fatigue Detection
In this paper, based on the background of driver fatigue detection, we propose a new algorithm to locate the eye in an image which is taken by infrared camera.Firstly, Median filtering is used to eliminate the noise points, then we make well use of the iterative threshold segmentation algorithm to change the gray image into binary image and improve its contrast ratio as well. The following procedure is to locate the face region by utilizing the connected components labeling algorithm, and then segment the face region with the method of minimum-area enclosing rectangle.Finally,a multi-resolution template matching approach is used to locate the eye after we construct a multi-resolution structure for the segmented face region and two-eyes template respectively. The simulation results show that this new algorithm is simple and practical; it not only avoids the influence of illumination on face location, but also possesses a faster speed and higher accuracy compared with the traditional template matching algorithm.
connected components labeling minimum areaenclosing rectangle multi-resolution template matching
Wangxin Huang Jin Tang Zhiqiang Xu
School of Information Science and Engineering Central South University Changsha, China
国际会议
成都
英文
1033-1036
2010-12-17（万方平台首次上网日期，不代表论文的发表时间）