Professional Documents
Culture Documents
Contents
a11
K
t
A
M O
t
aM 1
L
t
a1N
M
aMN
f
f
x f
K
x
x1
xN
g1
x
1
M
g M
x
1
K
O
L
g1
xN
g M
xN
A
x
x
If A is symmetric:
T
x Ax 2Ax
x
C. Nikou Digital Image Processing (E12)
2
T
Ax b 2A Ax b
x
10
Linear, Position-Invariant
Degradation
We now consider a degraded image to be
modelled by:
g ( x, y ) h ( x, y ) f ( x, y ) ( x, y )
where h(x, y) is the impulse response of the
degradation function ( i.e. point spread function
blurring the image).
The convolution implies that the degradation
mechanism is linear and position invariant (it
depends only on image values and not on
location).
C. Nikou Digital Image Processing (E12)
11
Linear, Position-Invariant
Degradation (cont...)
In the Fourier domain:
G (k , l ) H (k , l ) F (k , l ) N (k , l )
where multiplication is element-wise.
In matrix-vector form:
g = Hf +
where H is a doubly block circulant matrix and
f,g, and are vectors (lexicographic ordering).
C. Nikou Digital Image Processing (E12)
12
Linear, Position-Invariant
Degradation (cont...)
g ( x , y ) h ( x , y ) f ( x , y ) ( x, y )
G (k , l ) H (k , l ) F (k , l ) N ( k , l )
g = Hf +
If the degradation function is unknown the
problem of simultaneously recovering f(x,y)
and h(x,y) is called blind deconvolution.
C. Nikou Digital Image Processing (E12)
13
Linear Restoration
Using the imaging system
g = Hf +
we want to estimate the true image from the
degraded observation with known
degradation H.
A linear method applies an operator (a
matrix) P to the observation g to estimate
the unobserved noise-free image f:
f = Pg
C. Nikou Digital Image Processing (E12)
14
g = Hf
an obvious solution would be to use the
inverse filter:
P=H
-1
yielding
f = Pg = H -1g = H -1Hf = f
C. Nikou Digital Image Processing (E12)
15
H = WW
16
-1
where
H = WW
f = W Wg Wf = WW Wg
-1
F = G
C. Nikou Digital Image Processing (E12)
17
G
(
k
,
l
)
F = G F (k , l ) =
H (k , l )
-1
18
G (k , l )
, H (k , l ) 0
F (k , l ) = H (k , l )
0
, H (k , l ) 0
19
20
J (f ) Hf - g
J
min J (f )
0
f
f
f
Hf - g 0
f H H HT g
T
21
g ( x , y ) h ( x , y ) f ( x , y ) ( x, y )
G (k , l ) H (k , l ) F (k , l ) N ( k , l )
g = Hf +
22
G (k , l ) H (k , l ) F (k , l ) N (k , l )
N
(
k
,
l
)
F (k , l ) F (k , l )
H (k , l )
Even if we know H(k,l) we cannot recover F(k,l)
due to the second term.
If H(k,l) has small values the second term
dominates (it goes to infinity if H(k,l)=0!).
C. Nikou Digital Image Processing (E12)
23
24
25
26
(f f ) 2
min
E
27
E x1 y1
E xN y1
E x1 y2
M
E xN y2
E x1 y N
M
M
K
K E xN y N
K
28
min
E
(
f
f
)
29
J (f ) E (f f )
2
2
E f f
E f Pg
T
n
J (f ) E
p g
T
n
2
T
E f n p n g
30
J (f ) E f n p g f n p g
n
T
n
T
n
T
T
T
T
T
T
E f nf n p n gf n f n g p n p n gg p n
n
T
T
T
T
T
T
E f nf n p n E
gf n
E f n g p n p n E
gg p n
n
31
T
T
R fnfn 2p n R gfn p n R gg p n 0
p n
1
R gfn
2R gfn 2R gg p n 0 p n R gg
p R fn g R
T
n
1
gg
32
P R fg R
1
gg
(Hf Hf
R gg E gg E
)(
E Hff H Hf f H
T
HR ff HT HR f R f HT R
C. Nikou Digital Image Processing (E12)
33
Also,
f (Hf )T ...R Hff
R fg E fgT E
P R fg R R ff H
HR
H R
T
ff
34
ff
f H R H R
T
1 1
ff
HT R 1 g
35
g
No noise (R=0, g=f):
This is the inverse filter.
36
37
H W W
H
The columns of W-1 are the eigenvectors of any
circulant matrix H.
The corresponding eigenvalues are the DFT values
of the signal producing the circulant matrix.
Remember also that WT W
1 T
1
and that
W
W
38
H W W
H
H W W
H
W W
H
If H is real:
H H
T
T *
( NW)
*
H
1
W
N
1 *
W
*
W
W
*
H
*
H
1 *
39
g
1
( W W
)(
W W ) W
( W
)(
W W
)(
W W ) W
( W ) g
ff
H
ff
1
W
ff W W
1
*
H
*
H
(
H
ff
Wf ff
H ( H ff
*
H
*
H
*
H
) g
Wg
)
40
*
Wf
ff
H ( H ff
*
H
Wg
) 1
F (k , l )
S ff (k , l ) H (k , l )
2
S ff (k , l ) H (k , l ) S (k , l )
G (k , l )
41
SNR(k , l )
S ff (k , l )
S (k , l )
F (k , l )
H (k , l )
2
H (k , l ) SNR -1 (k , l )
C. Nikou Digital Image Processing (E12)
G (k , l )
42
S ff (k , l )
G (k , l )
MN
C. Nikou Digital Image Processing (E12)
43
44
Inverse
Wiener
45
46
J (f , ) Hf g Qf
Data fidelity term
Smoothness term
47
, f 0
48
F (k , l )
H * (k , l )
2
H (k , l ) Q(k , l )
G (k , l )
49
50
M 1 N 1
( m, n)
f
(
m
,
n
)
m 0 n 0
51
SNR
M 1 N 1
f (m, n) 2
m 0 n 0
( m, n)
f
(
m
,
n
)
m 0 n 0