Note: This page has been translated by MathWorks. Please click here

To view all translated materials including this page, select Japan from the country navigator on the bottom of this page.

To view all translated materials including this page, select Japan from the country navigator on the bottom of this page.

2-D adaptive noise-removal filtering

**The syntax wiener2(I,[m n],[mblock nblock],noise) has been removed.
Use the wiener2(I,[m n],noise) syntax instead.**

`J = wiener2(I,[m n],noise)`

`[J,noise_out] = wiener2(I,[m n])`

filters the grayscale image `J`

= wiener2(`I`

,`[m n]`

,`noise`

)`I`

using a pixel-wise adaptive
low-pass Wiener filter. `[m n]`

specifies the size
(`m`

-by-`n`

) of the neighborhood used to
estimate the local image mean and standard deviation. The additive noise (Gaussian
white noise) power is assumed to be `noise`

.

The input image has been degraded by constant power additive noise.
`wiener2`

uses a pixelwise adaptive Wiener method based on
statistics estimated from a local neighborhood of each pixel.

`wiener2`

estimates the local mean and variance around each pixel.

$$\mu =\frac{1}{NM}{\displaystyle \sum _{{n}_{1},{n}_{2}\in \eta}a({n}_{1},{n}_{2})}$$

$${\sigma}^{2}=\frac{1}{NM}{\displaystyle \sum _{{n}_{1},{n}_{2}\in \eta}{a}^{2}({n}_{1},{n}_{2})-{\mu}^{2}},$$

where $$\eta $$ is the *N*-by-*M* local
neighborhood of each pixel in the image `A`

. `wiener2`

then creates a pixelwise Wiener filter using these estimates,

$$b({n}_{1},{n}_{2})=\mu +\frac{{\sigma}^{2}-{\nu}^{2}}{{\sigma}^{2}}(a({n}_{1},{n}_{2})-\mu ),$$

where ν^{2} is the noise variance. If the noise variance is
not given, `wiener2`

uses the average of all the local estimated
variances.

[1] Lim, Jae S., *Two-Dimensional Signal and Image
Processing*, Englewood Cliffs, NJ, Prentice Hall, 1990, p. 548, equations
9.26, 9.27, and 9.29.

Was this topic helpful?