上海大学学报(自然科学版) ›› 2012, Vol. 18 ›› Issue (6): 555-560.

• 通信与信息工程 • 上一篇    下一篇

一种新的自适应维纳滤波方法

李东兵,李国平,滕国伟,赵海武,王国中,李萍   

  1. 1.上海大学 通信与信息工程学院,上海 200072; 2.上海国茂数字技术有限公司,上海 201204
  • 收稿日期:2012-03-02 出版日期:2012-12-28 发布日期:2012-12-28
  • 通讯作者: 李国平(1974—),男,副教授,博士,研究方向为音视频压缩多媒体通讯. E-mail:liguoping@shu.edu.cn
  • 作者简介:李国平(1974—),男,副教授,博士,研究方向为音视频压缩多媒体通讯. Email:liguoping@shu.edu.cn
  • 基金资助:

    上海大学创新基金资助项目(A10010710003)

A New Method of Adaptive Wiener Filtering

LI Dong-bing,LI Guo-ping,TENG Guo-wei,ZHAO Hai-wu,WANG Guo-zhong,LI Ping   

  1. 1. School of Communication and Information Engineering, Shanghai University, Shanghai 200072, China;
    2. Shanghai GMT Digital  Technology Co.〖KG-*3〗, Ltd.〖KG-*3〗, Shanghai 201204, China
  • Received:2012-03-02 Online:2012-12-28 Published:2012-12-28

摘要: 针对经典维纳滤波器存在的不足,提出一种新的自适应维纳滤波器.该滤波器能够根据不同的图像特性在给定的多个模板之间自适应地选择模板,使得滤波效果更加理想;并且相对于传统的维纳滤波方法,可以较完整地保存图像的轮廓信息.试验结果表明,新的自适应维纳滤波方法在主观和客观性能方面比传统的维纳滤波方法均有较大的改善.

关键词: 经典维纳滤波, 自适应维纳滤波, 自适应选择

Abstract: Aimed at improving the insufficient result of classic Wiener filtering, a new adaptive Wiener filter is proposed. The new filter can adaptively select the filter template from different given templates, giving a better filtering result. The traditional Wiener filter seeks to suppress noise in images at the cost of losing some edge information, while the proposed method can overcome the problems. Experimental results show that the proposed filter gives improved performance both in subjective and objective evaluations as compared to the classic Wiener filtering.

Key words: adaptive selection, adaptive Wiener filter, classical Wiener filter

中图分类号: