NSSC OpenIR  > 其他部室
一种增强结构信息的图像平滑算法
Alternative TitleAn Image Smoothing Method with Structure Information Enhancing
张道兵; 张慧; 孙显; 刘波; 王宏琦; 北京8701信箱
Department火箭与气球探空技术研究室
Source Publication光子学报
2007
Volume26Issue:S1Pages:304-308
ISSN1004-4213
Language中文
Keyword图像处理 图像平滑 非线性扩散 一致性增强 结构张量
Abstract基于非线性扩散滤波提出了可以同时对同质区域进行平滑和对结构信息进行增强的算法.该算法在同质区域采用各项同性滤波而在边缘附近采用各项异性扩散滤波,避免了图像平滑和增强结构信息之间的矛盾.在实现时,根据一致性程度选择扩散矩阵的扩散系数,可确保偏微分方程在离散化时的简洁性和一致性.数字实验结果表明,可以有效的平滑一致性低的区域并增强一致性高的结构.
Other AbstractIt presents a numerical scheme for smoothing homogeneity with coherence enhanced based on nonlinear diffusion filter. Isotropic diffusion in homogenous areas and anisotropic diffusion in edge neighbors are uniformed into the scheme, which could avoid the paradox of smoothing homogenous areas and enhancing structure information. During implementing, switching diffusion coefficients in diffusion matrix according to coherence ensures the discretization of partial differential equation to be concision and consistency. Numerical examples show that it is efficient in smoothing low coherence areas and enhancing coherence structures.
Funding Project中国科学院空间科学与应用研究中心
Document Type期刊论文
Identifierhttp://ir.nssc.ac.cn/handle/122/1663
Collection其他部室
Corresponding Author北京8701信箱
Recommended Citation
GB/T 7714
张道兵,张慧,孙显,等. 一种增强结构信息的图像平滑算法[J]. 光子学报,2007,26(S1):304-308.
APA 张道兵,张慧,孙显,刘波,王宏琦,&北京8701信箱.(2007).一种增强结构信息的图像平滑算法.光子学报,26(S1),304-308.
MLA 张道兵,et al."一种增强结构信息的图像平滑算法".光子学报 26.S1(2007):304-308.
Files in This Item:
File Name/Size DocType Version Access License
200736S1304.pdf(906KB) 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张道兵]'s Articles
[张慧]'s Articles
[孙显]'s Articles
Baidu academic
Similar articles in Baidu academic
[张道兵]'s Articles
[张慧]'s Articles
[孙显]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张道兵]'s Articles
[张慧]'s Articles
[孙显]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.