您的位置:首页  > 论文页面

基于各向异性扩散的电子散斑图像去噪

发表时间:2010-08-15  浏览量:1379  下载量:699
全部作者: 阮江涛,佟景伟,王世斌
作者单位: 天津大学机械工程学院
摘 要: P-M模型是图像平滑中的经典模型,但在这个传统各向异性扩散方程中,通常只用到了中心点的4邻域方向,漏掉了一些图像信息。采用基于中心点8邻域方向即包含对角线方向的改进各向异性扩散方程对电子散斑图像进行滤波处理。并讨论了去相关最优迭代次数。实验结果表明:该方法能有效滤除散斑干涉条纹图的散斑颗粒噪声,并保持了边缘信息,有利于下一步实验数据的处理。
关 键 词: 固体力学;各向异性扩散;电子散斑干涉;去噪;方向;去相关
Title: ESPI image denoising based on anisotropic diffusion
Author: RUAN Jiangtao, TONG Jingwei, WANG Shibin
Organization: School of Mechanical Engineering, Tianjin University
Abstract: P-M model is a classical model for image smoothing, but in conventional anisotropic diffusions, usually 4�neighborhood directions are used. Therefore, some image details are lost. To overcome the shortcoming of the classical anisotropic diffusion equation, an improved anisotropic diffusion method based on 8�neighborhood orientations of the center pixel including the diagonal directions is proposed. Meanwhile, optimal iteration times based on decorrelation is discussed. Experimental results show that the proposed algorithm can effectively reduce speckle noise in the patterns while preserving edge details. It is beneficial to the following processing of experimental data.
Key words: solid mechanics; anisotropic diffusion; electronic speckle pattern interferometry; denoising; orientation; decorrelation
发表期数: 2010年8月第15期
引用格式: 阮江涛,佟景伟,王世斌. 基于各向异性扩散的电子散斑图像去噪[J]. 中国科技论文在线精品论文,2010,3(15):1578-1583.
 
0 评论数 0
暂无评论
友情链接