视线跟踪系统中CCD摄像机的自适应调节

分析了视线跟踪系统中光源、摄像机、眼睛角膜及光斑的几何位置关系,提出了眼睛跟踪策略和一种调光、变焦、聚焦相结合的CCD物镜自动调节方法。对捕获图像进行处理,依据图像灰度均值控制光圈,以适应外部环境光强变化;利用SMD算子大步幅快速粗聚焦,提取图像中光斑信息控制云台瞄准、跟踪眼睛,并进行变焦控制,可在用户头肩图像中自动定位眼睛,适应用户头部位置变化;在眼睛区域自动确定检测窗,在检测窗内使用FSWM聚焦算子判断图像清晰度,实现自动聚焦。若图像中能提取瞳孔信息,则依据瞳孔边缘梯度均值实时调整聚焦。实验表明,在不同外部环境光强及用户在不同使用位置情况下,CCD摄像机通过自适应调节,可实时跟踪眼睛并捕获到清晰眼睛图像。

Adaptive regulation of CCD camera in eye gaze tracking system

An eye-gaze tracking method and an automatic adjusting method for object lens of CCD camera in combining the aperture regulation with zooming and focusing are proposed by analyzing the geometrical positions of two LEDs, a camera, an eye cornea and its glints. The aperture of CCD lens is controled to adapt the system to external lighting circumstances based on the average brightness values of the processed images. A SMD operator is used for focusing roughly and quickly in a large scale for obtaining the eye glints to control a pan-tilt unit for eye aiming and eye tracking, and regulating the focus of the lens to locate the eye and adapt the head movements in a head-shoulder image. The FSWM operator is applied in the determined window in the eye region to focus automatically for getting clear images. If the pupil can be extracted from the image, the average gradient of the edge points of the pupil can be used for real-time focusing. Experiments show that this system can track the eye and acquire clear eye images in real time by the adaptive regulation of the CCD camera in different external lightings and user position conditions.