一种用于综合电脑验光仪的快速人眼识别算法

针对嵌入式系统中所采用的Hough和随机Hough变换方法来确定人眼位置的方法存在的运算量较大问题，结合综合电脑验光仪所采集图像的特点，本文提出了一种对硬件系统要求较低，且识别速度较快，同时可以保证精度的识别定位人眼瞳孔定位方法。该方法通过获取图像中对象的轮廓，寻找连通区域与边界，然后根据圆度与设定阈值的比值的大小来识别人眼瞳孔并定位其中心位置实现自动跟踪。结果表明该方法可以准确确定人眼位置实现自动跟踪，且运算复杂度较低，运算速度相比于传统Hough变换算法有较大提升，可降低硬件成本。

A Fast Eye Recognition Algorithm for Integrated Refractometer

Aiming at the computation problem for determining the human eye location using Hough and random Hough transform method in embedded system, this paper presents a method that can recognize the human eye localization, which can ensure the fast and precision recognition by using the relatively lower hardware system. The method finds the connected region and boundary by acquiring the contour of the object in the image, and then it can recognize the eye pupil and locate its central location to achieve automatic tracking according to the comparison of the ratio of the roundness and the set threshold. The results show that this method can accurately locate human eyes to achieve automatic tracking, and the computation complexity is low. Compared with the traditional Hough transform algorithm, the algorithm can greatly improve the computation speed and reduce the cost of hardware.