首页 | 本学科首页   官方微博 | 高级检索  
     

一种改进的自适应保细节中值滤波算法
引用本文:朱菊华,杨新,李俊.一种改进的自适应保细节中值滤波算法[J].计算机工程与应用,2001,37(3):93-95.
作者姓名:朱菊华  杨新  李俊
作者单位:上海交通大学图象处理与模式识别研究所
基金项目:国家自然科学基金!(项目编号:69931010)的部分资助。
摘    要:中值滤波是常用的降噪算法,它可以保留比较尖锐的边界,但是却容易模糊图象的细节。尽管已经有一些改进算法,但效果并不十分理想。文章介绍了一种保细节中值滤波算法,采用了多尺度多方向的窗口,根据图象各部分特性自适应地选择窗口进行中值滤波。实验结果证明,文章的算法优于其他几种常用的中值滤波算法。

关 键 词:图象处理  中值滤波  自适应滤波  多尺度
修稿时间:2000年8月1日

The Improved Adaptive Minutiae Preserving Median Filter
Zhu Juhua,Yang Xin,Li Jun.The Improved Adaptive Minutiae Preserving Median Filter[J].Computer Engineering and Applications,2001,37(3):93-95.
Authors:Zhu Juhua  Yang Xin  Li Jun
Abstract:Median filter is a widely used and effective nonlinear filter in image processing. However,it can not be ignored that median filter blurs the minutiae such as points and short and thin lines. Although there're several improved median smoothers,the results are not satisfying. In this paper,we propose an improved adaptive median filter using multiscale and multidirectional window. This method,to some extent,is effective for both smoothing and minutiae preserving, mainly because it selects different window according to the feature of different part of the image. Compared with other methods,this algorithm has better performance.
Keywords:image processing  median filtering  adaptive filtering  multiscale
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号