Title: BAE 790I BMME 231 Fundamentals of Image Processing Class 18
1BAE 790I / BMME 231Fundamentals of Image
ProcessingClass 18
- Restoration Filtering
- Inverse Filters Properties
- Wiener Filters
- MSE
- Properties
2Image Restoration
- Objective To quantitatively estimate the true
image from its degraded measurement.
System H
Restoration process
g
f
n
An estimate of f
3Noise Effects
- Problem The inverse filter generally magnifies
the noise (note Fourier-domain example).
4Inverse Filter with Noise
24.2 dB
1.7 dB
High-frequency noise is amplified by the inverse
filter.
5Noise Effects
- Consider the error image between the true image
and the inverse-filter estimate - The estimate is unbiased. (On average, it is
correct.)
6Noise Effects
- Consider the autocorrelation of the error image
for the inverse filter - If H has some small eigenvalues, H-1 has some BIG
eigenvalues. - If Rnn has content in the corresponding
eigenimages, those elements of the noise get BIG.
7Mean Square Error
- Consider the mean square error (MSE) between the
true image and the inverse-filter estimate - This is a scalar quantity that gives a measure of
how close the estimate is, on average. - It is not zero.
8Restoration Filtering
- Objective To quantitatively estimate the true
image from its degraded measurement.
System H
Filter Q
g
f
n
An estimate of f
9Noise Effects
- Consider the error image between the true image
and the restoration filter estimate - This is the case for any linear system Q.
10Mean Square Error for Q
- First, we compute an expression for the square
error when using any linear system
11Mean Square Error for Q
- Second, we apply the expectation operator to get
the mean. Only the noise is random, so the
operator only applies to terms with n
12Mean Square Error for Q
- Second, we apply the expectation operator to get
the mean. Only the noise is random, so the
operator only applies to terms with n
Terms with En go to zero since n is zero mean
and uncorrelated with f.
13Mean Square Error for Q
- Second, we apply the expectation operator to get
the mean. Only the noise is random, so the
operator only applies to terms with n
Terms with En go to zero since n is zero mean
Terms with En go to zero since n is zero mean
and uncorrelated with f.
14Mean Square Error for Q
- Third, we want to find Q to minimize the MSE.
- Take the derivative with respect to Q
15Mean Square Error for Q
- The derivative operator is a square matrix that
takes the derivative with respect to every
element of Q.
16Aside Matrix Derivative Relations
17Mean Square Error for Q
- The derivative operator is a square matrix that
takes the derivative with respect to every
element of Q.
18Mean Square Error for Q
- Fourth, set the derivative to zero and solve for
Q.
19Linear Wiener Filter
- The filter Q gives the minimum MSE for the linear
case. - The only stipulation we made was that n is
zero-mean and uncorrelated with f. - This is the linear version of the Wiener filter.
20LSI Wiener Filter
- For LSI systems, in the Fourier domain, this
becomes
21LSI Wiener Filter
- Note the similarity between the linear and LSI
forms
22LSI Wiener Filter
- Note that the noise-to-signal ratio appears in
the denominator - What happens at frequencies where noise is low?
- Frequencies where noise is high?
23Wiener Filter
- The Wiener filter automatically cuts off the
filter at frequencies where noise becomes
significantly higher than signal. - It also restores some of the blurring.
24Wiener Filter - Blur Sigma 2
Original
Blur noise
SNR24.2 dB NMSE .0027
Wiener - noise-free
Wiener - noisy
SNR27.3 dB NMSE .0015
25Wiener Filter - Blur Sigma 5
Original
Blur noise
SNR23.5 dB NMSE .0074
Wiener - noise-free
Wiener - noisy
SNR26.0 dB NMSE .0041
26Wiener Filter Characteristics
Constant SNR 200 (26 dB)
27Wiener Filter Characteristics
Constant Gaussian Blur sigma 2
28Wiener Filter Bias
- The bias of the Wiener Filter result
- This is generally not zero. The Wiener filter
estimate will be biased.
29Linear Wiener Filter
- Normally, this would require inversion of a large
matrix - Some approximations may help.
30Generalized Wiener Filter
- Implement the Wiener filter in another transform
domain - Hopefully, the inversion is easier to accomplish
this way.
Unitary Transform A
Wiener Filter M
Inverse Transform AT
g
f