人眼屈光度校正算法的研究现状

Status of Study on the Human Eye Diopter's Correction Algorithm

人类能够与用户图像界面进行交互是依赖于一个清晰正常的视觉。然而,屈光不正(例如近视、远视、散光等)能使人眼视网膜中所显示的图像质量退化。图像质量的好坏严重影响了我们在军事、医学、娱乐领域所取得的研究成果。目前,对于人眼屈光度校正算法的研究主要有五种方法:1.应用一个维纳滤波的预补偿算法来校正人眼屈光度;2.用一个全变差的方法绘制图像对人眼屈光度进行校正;3.用视觉分析系统来校正人眼屈光度;4.使用多层显示器来校正视觉畸变;5.使用可计算的光场显示器来校正视觉畸变。通过研究这几种算法的优缺点提出现有算法的改进以及新的算法研究。

Human ability to interact with graphical user interfaces （GUIs） is dependent on a clear normal vision. However, refractive errors such as myopia, hyperopia or astigmatism, can degrade the image quality showed in the human retina. The image quality showed in the human retina can strongly affect the research result we have in military, medical, and entertainment. Now, there are mostly five approach in the research of the human eye diopter＇s correction algorithm： 1. a pre-compensation algorithm to correct the human eye diopter＇s using a wiener filter. 2. a total variation approach for customizing imagery to correct the human eye diopter＇s algorithm. 3. use a vision analysis system to correct the human eye diopter＇s algorithm. 4. use multilayer displays to correct visual aberrations. 5. use computational light field displays to correct visual aberrations. This paper makes a conclusion by study the advantages and disadvantages of three algorithms, which is the improvement of current algorithms and the research of new algorithm.