Development of the Imaging System Based on the Human Visual System.

我们知道人类的视觉系统并没有很好的光学元件，在大脑中会进行恢复，但我们可以获得清晰的图像，这是因为大脑会恢复视网膜上的退化图像。因此，我们提出了具有与人类视觉系统相同功能的成像系统，并在本研究中开发了基于人类视觉系统的高质量成像系统。(1)We提出与视觉系统功能相似的原理，并通过计算机模拟进行检验。我们使用GARDS（Generalized Analytic Reconstruction from Discrete Samples）方法进行复原。该方法用灵敏度分布函数代替点扩散函数来描述系统的特性。该方法的优点是将对象空间处理为连续的，能够解析地恢复空间变退化，特别是能够用灵敏度函数描述系统中出现退化的所有因素。(2)If有人使用简单的透镜来获得图像，所获得的图像由于模糊或失真而劣化。然后，可以通过使用关于光学系统的属性的知识进行后处理来恢复它们。但是，通常情况下，光学系统的特性很难得到。因此，我们提出了一种方法来学习功能参数与选择适当的模式（基函数）。(3)we证实了该方法对空变成像系统的有效性。我们在工作站上制作了一个光学系统（监视器、单透镜和CCD相机）与后处理相结合的系统并进行了实验。在拍摄了一幅退化图像后，利用该方法进行了恢复.并且，实现了组合液晶透镜和恢复处理的变焦功能。

We can know htat a human visual system does not have good optical elements and restoration will be made in the brain, bue we can obtain a clear image.This reason is that the brain restores degraded images taken on the retina.So we propose the imaging system that has same functions as teh human visual system.and we develop the high quality imaging system based on the human visual system in this research.(1)We propose principles which are similar to functions of the visual system and examine them through computer simulations. We use the method that called GARDS(Generalized Analytic Reconstruction from Discrete Samples) for restoration. In this method the characteristics of the system is described with sensitivity distribution functions instead of point spread functions. Merits of this method are to deal the object space as continuous, to be able to restore space-variant degradation analytically, and especially to be able to describe all factors of the degradation appeared in the system by sensitivity functions.(2)If somebody uses a simple lens to get and image, the obtains image is degraded by blur or distortion. Then one can restore them through post-processing by using knowledge about properties of the optical system. But usually, it is difficult to get characteristics of optical systems. So we propose a method to learn functional parameters with selecting proper patterns(base functions).(3)we have confirmed effectiveness of our method for the space-variant imaging system. we made and experiment on the system combined an optical system (a monitor, a single lens and CCD-camare) with post-processing which is on a workstation. After taking a degraded image, we restored in by the method. And also, A zooming function that combines a liquid crystal lens and restoring process is realized.