A patch-based non-local means method for image denoising

Shaobo Dang, Yanning Zhang, Dong Gong

科研成果: 书/报告/会议事项章节会议稿件同行评审

3 引用 (Scopus)

摘要

In this paper, a revised version of non-local means denoising method is proposed. Different from the original non-local means method in which the algorithm is processed on a pixel-wise basis, the proposed method using image patches to implement non-local means denoising. Given that some details, texture and structure information will be smoothed out when performing weighted averaging, we carry out a pre-processing procedure to classify the image patches into several clusters according to their feature similarities. Later, the weights needed in non-local means algorithm is calculated between image patches in the same cluster. By this means, the above mentioned detail, texture and structure can be preserved due to the redundancy in similar image patches. We illustrate the overall algorithm's performance via several experiments. The results indicate the effectiveness of the proposed method.

源语言英语
主期刊名Intelligent Science and Intelligent Data Engineering - Third Sino-Foreign-Interchange Workshop, IScIDE 2012, Revised Selected Papers
582-589
页数8
DOI
出版状态已出版 - 2013
活动3rd Sino-Foreign-Interchange Workshop on Intelligent Science and Intelligent Data Engineering, IScIDE 2012 - Nanjing, 中国
期限: 15 10月 201217 10月 2012

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
7751 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议3rd Sino-Foreign-Interchange Workshop on Intelligent Science and Intelligent Data Engineering, IScIDE 2012
国家/地区中国
Nanjing
时期15/10/1217/10/12

指纹

探究 'A patch-based non-local means method for image denoising' 的科研主题。它们共同构成独一无二的指纹。

引用此