Scribd
Upload
78 views4 pages

Unit Iii Image Restoration

This document discusses various methods for image restoration, including inverse filtering, Wiener filtering, and constrained least squares filtering. Inverse filtering attempts to recover the original image by dividing the degraded image by the degradation function. Wiener filtering treats images and noise as random variables and estimates the original based on minimizing the mean square error. Constrained least squares filtering finds the minimum of a criterion function subject to a constraint to estimate the original image. The document also covers geometric spatial transformations which modify pixel spatial relationships through coordinate transformation and intensity interpolation.

Uploaded by

Nivethithaa Dhanraj
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
78 views4 pages

Unit Iii Image Restoration

This document discusses various methods for image restoration, including inverse filtering, Wiener filtering, and constrained least squares filtering. Inverse filtering attempts to recover the original image by dividing the degraded image by the degradation function. Wiener filtering treats images and noise as random variables and estimates the original based on minimizing the mean square error. Constrained least squares filtering finds the minimum of a criterion function subject to a constraint to estimate the original image. The document also covers geometric spatial transformations which modify pixel spatial relationships through coordinate transformation and intensity interpolation.

Uploaded by

Nivethithaa Dhanraj
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

UNIT III Image Restoration

Image Restoration degradation model g(x,y) f(x,y) Degradation Function H + Restoration Filter(s)

f(x,y)

Noise n(x,y) Degradation Restoration

In spatial domain g(x,y) = h(x,y) * f(x,y) + n(x,y)

In frequency domain G(u,v) = H(u,v) F(u,v) + N(u,v)

where h(x,y) is the spatial representation of the degradation function, * indicates the convolution.

Unconstrained restoration

Inverse filtering In this an estimate F(u,v) = G(u,v)

----------H(u,v)

Substituting for G(u,v)

F(u,v) = F(u,v) +

N(u,v) ----------H(u,v)

If we know the degradation function we cannot recover the undegraded image because N (u, v ) is not known.

To avoid the degradation function has zero or small values the filter frequencies has been chosen away from the origin.

Removal of blur caused by linear motion Wiener filtering

Considering images and noise as random variables f-uncorrupted image, f-estimate error = E(f-f) 2

F(u,v) =

H*(u,v)sf (u,v) --------------------------------------s f (u,v)H(u,v) 2 + s (u,v)

G(u,v)

H*(u,v) --------------------------------------H(u,v) 2 + s (u,v) /sf (u,v)

G(u,v)

1 ------H(u,v)

H(u,v) 2

G(u,v)

--------------------------------------H(u,v) 2 + s (u,v) /sf (u,v)

The result is known as Wiener filter or minimum mean square filter .

Constrained Least Square Filtering:

In this method find the minimum of criterion function M-1 N-1 C= [2 f(x,y) ] 2

x=0 y=0 subject to the constraint g-Hf 2 = 2

where W 2 W T W is the Euclidean vector norm, fis the estimate of the undegraded image. The frequency domain solution to this optimization problem is given by the expression

f(u,v) =

H(u,v) 2 H(u,v) 2 +

G(u,v) P(u,v) 2

---------------------------------------

where is a parameter that must be adjusted so that the constraint is satisfied and P(u,v) is the Fourier transform of the function

0 p(x,y) = -1 0

-1 4 -1

0 -1 0

Geometric spatial transformation:

Geometric transformations modify the spatial relationship between the pixels in an image. It consists of i) spatial transformation of coordinates ii) intensity interpolation that assigns intensity values to the spatially transformed pixels. The transformation of coordinates may be expressed as (x,y) = T{ (v,w ) } where (v,w ) are pixel coordinates in the original image and (x,y) are the corresponding pixel coordinates in the transformed image. One of the most commonly used spatial coordinate transformations is the affine transform which can scale ,rotate, translate or sheer set of points. The transformations relocate pixels on an image to new location. Assigning intensity values to new location is called interpolation.

You might also like