Wiener deconvolution

Wiener deconvolution is a method used to improve the details of signals and increase the signal-to-noise ratio and resolution of estimated signals. It is often adopted in linear deconvolution to effectively diminish noise in each image. If the imaging system satisfies the linear shift-invariant condition, deconvolution methods such as Wiener deconvolution and Richardson-Lucy deconvolution can improve the image spatial resolution, although the improvement is limited by the noise level in the image.From: Testing and Measurement: Techniques and Applications [2015], Imaging resolution and properties analysis of super resolution microscopy with parallel detection under different noise, detector and image restoration conditions [2018], Thermal image enhancement through the deconvolution methods for low-cost infrared cameras [2018]

Least Mean-Square Error Predictors

View Chapter

Purchase Book

Published in Arthur David Snider, Random Processes for Engineers, 2017

Arthur David Snider

The so-called Wiener deconvolution, implemented through Fourier transforms as indicated, has been very successful in such applications as clarifying audio recordings and photographs (using a two-dimensional version); see the sources in the following. Note that for white noise Re(k)=σe2δk0,Se(f)=σe2, and the normal equations 5.30 resemble 5.27.

For-EMD regularized deconvolution in acoustic measurement

View Chapter

Purchase Book

Published in Kennis Chan, Testing and Measurement: Techniques and Applications, 2015

Y. Dai, J.H. Yang, X.H. Cao, H. Hou

shrinkage of ideal deconvolution in EMD domain. This method improves the details of signals after Wiener deconvolution, increasing SNR and resolution of estimated signals. The method is self-adaptive and robust because it doesn't require any regularization parameters setting.

Imaging resolution and properties analysis of super resolution microscopy with parallel detection under different noise, detector and image restoration conditions

View Article

Journal Information

Published in Journal of Modern Optics, 2018

Zhongzhi Yu, Shaocong Liu, Shiyi Sun, Cuifang Kuang, Xu Liu

Linear DCV is a method that performs DCV on each detector’s image individually and then sums the images. In parallel detection, the size of each individual detector is relatively small compared to the pinhole in conventional confocal microscopy, and therefore the noise in each image captured by an individual detector is relatively high. Therefore, with linear deconvolution, we adopt Wiener deconvolution to diminish noise in each image more effectively. The purpose of DCV is to find some g(t) that can estimate object’s original signal x(t) as follows:

Wiener deconvolution

Explore chapters and articles related to this topic

Least Mean-Square Error Predictors

For-EMD regularized deconvolution in acoustic measurement

Imaging resolution and properties analysis of super resolution microscopy with parallel detection under different noise, detector and image restoration conditions