二维数字图像恢复算法的研究

Study of Digital Image Restoration Algorithm in 2D

现代数字技术使对多维数字信号的处理成为可能,从简单的数字环路到更高级的平行计算机系统。数字图像的处理可以分为以下三个步骤:图像预处理,图像分析,图像认知。讨论二维图像处理,其所用到的大多概念和技术均可扩充到三维或更多维的图像处理。通过将各种滤波算法作用于一幅模糊成信噪比为30dB的图像。得出这样的结论:基于线性滤波和最小均方误差(或均方根值误差)准则的维纳滤波是一种解决图像恢复问题的较好方法。

Modern digital technology has made it possible to manipulate multi-dimensional signals with systems that range from simple digital circuits to advanced parallel computers. The manipulation of digital image can be divided into three steps： image processing, image analysis, image understanding. This paper only discusses 2-D image process beeause most of the concepts and techniques that are to be described can be extended easily to three or more dimension. In the example given in the paper, a variety of image restoration algorithms applied to a image of which the SNR is 30dB. The result shows that, the Wiener filtering is a better solution to the restoration problem based upon the hypothesized use of a linear filtering and the minimum mean-square error criterion.