Eye location is the precondition to monitor drivers fatigue state with the help of image processing technique.In order to achieve the purpose of detecting eyes location , a method was put forward to locate a driver ’ s eyes based on infrared difference and gray integral projection .Firstly, two images for bright and dark pupils were collected for the pupil ’ s different condition under the adaxial and abaxial infrared irradiation .Secondly , the collected parity frame images contained only binary pupil images by means of differential morphological processing .Finally, precise horizontal and vertical coordinates of the left and right eyes were obtained based on the vertical and horizontal projection on the left and right eyes respectively .Good results were achieved .The experimental results show that the method has high accuracy and real-time performance .This result is of great help to the next step of driver fatigue detection .%眼睛位置的快速准确定位是利用机器视觉方法对驾驶员的状态进行监测以及预警的前提。为了检测出眼睛的位置，采用一种基于红外差分和灰度积分投影的方法进行了驾驶员眼睛的定位。首先利用瞳孔在近轴和远轴红外光分别照射下的不同状态，采集到瞳孔亮暗程度不同的两幅图像，然后将采集到的奇偶帧图像进行差分等相关形态学处理，获得了只包含瞳孔的二值化图像，最后分别对左右眼进行垂直和水平投影，获得左右眼的精确横竖坐标。对该方法进行了理论分析和实验验证，取得了很好的效果。结果表明，该方法具有较高的准确率和实时性。这一结果对下一步驾驶员的疲劳检测是有很大帮助的。
展开▼
