加窗傅里叶滤波和相干增强扩散在条纹去噪中的比较分析  

Comparison of Windowed Fourier Filtering and Coherence Enhancing Diffusion for Fringe Pattern Denoising

作　　者：王海霞[1] 彭柔 钱克矛[2] 

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310014 [2]南洋理工大学计算机科学与工程学院,新加坡639798

出　　处：《实验力学》2017年第5期620-626,共7页Journal of Experimental Mechanics

基　　金：国家自然科学基金(61602414)

摘　　要：光学干涉测量具有非接触、高精度和全场测量的优点,能对形变、折射率、位移等信息进行测量。噪声滤除是光学干涉测量产生的条纹图像处理的一个关键问题。加窗傅里叶滤波(Windowed Fourier Filtering,WFF)与自适应加窗傅里叶滤波(Adaptive Windowed Fourier Filtering,AWFF)是有效的频域去噪算法。相干增强扩散(Coherence Enhancing Diffusion,CED)则是基于偏微分方程的空域去噪算法。针对条纹去噪问题,比较了WFF、AWFF和CED在不同密度和不同噪声类型的条纹图上的表现,分析了它们的适用条纹类型。Optical interferometry has the advantages of non-contact,high accuracy and full field measurement.Fringe patterns produced by various optical interferometric techniques encode the information of deformation,refractive index,vibration,etc.Noise is one of the key problems affecting fringe pattern processing. Windowed Fourier filtering（WFF）and adaptive windowed Fourier filtering（AWFF）are effective frequency domain denoising methods.Coherence enhancing diffusion（CED）is a spatial domain denoising method based on partial differential equation（PDE）.In this paper,the performances of WFF,AWFF and CED are compared by applying them to fringe patterns with different frequencies and different noise types.Applicable types of fringe pattern for WFF,AWFF and CED are identified respectively.

关 键 词：条纹图 相干增强扩散 加窗傅里叶滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]