文章摘要

文章信息/Info

Title:: Wiener Filter Image Restoration based on Global ?Variance and Noise Estimation

Author(s):: LI Dongsheng ; CHEN Chunxiao; WANG Zhangli ; YANG Junhao; Department of Biomedical Engineering, Nanjing University of Aeronautics ?and Astronautics, Nanjing 211106，China

Keywords:: Image restoration; Wiener filter; Global variance; Noise estimation; Structural similarity index

摘要:: 图像复原是从降质图像中利用图像处理技术恢复接近真实图像的方法。维纳滤波在抑制噪声和恢复模糊图像方面具有广泛的应用。然而传统的维纳滤波方法没有充分利用降质图像的信息，在失真图像复原过程中仍存在复原质量不高的问题。本研究将图像的全局方差和噪声功率谱引入维纳滤波，提出了基于全局方差和噪声估计的二次维纳滤波复原方法。实验表明，本研究提出的方法可以增强图像复原算法的稳定性，能够恢复更多的图像细节信息，图像复原质量较高。

Abstract:: Image restoration is a method using image processing techniques to recovery the degenerate image in order to gain an approximate image which is similar to the real image. Wiener filtering is widely used in suppressing noise and recovering blurring image. However, the information of the degenerate image hadn’t been maken full use, there were still many defects of the traditional Wiener filtering method, such as hard to gain a high quality image. In this paper, the global variance and noise power spectrum were introduced into Wiener filtering, and a new Wiener filter restoration method which based on global variance and noise estimation was proposed. Experiments show that the method proposed can enhance the stability of the image restoration algorithm, acquire more details of restored image and obtain a high quality image finally.