Hilbert Space Geometry in Wavelet Image Compression Algorithms
Hilbert Space Geometry in Wavelet Image Compression Algorithms
Hilbert Space Geometry in Wavelet Image Compression Algorithms
Figure 1: A diagram demonstrating the steps in wavelet based image compression and reconstruction.
Wavelets
For each n, k Z , define n ,k ( x) by
n ,k ( x) 2 n / 2 (2 n x k )
Now, then let {Vn } be an MRA with scaling function (x) and scaling filter h(k ) ,
x
where h(k ) 2 1 / 2 ( ), ( x k ) .
2
Then the wavelet filter g (k ) is defined by
g (k ) (1) k h(1 k )
and the wavelet by
( x) g (k ) 2 (2 x k ) .
kZ
L2 ( R) L2 (T ) l 2 ( Z )
How Wavelets Work
The first two conditions may be expressed in terms of the filter sequences h, g which
respectively define H, G:
k h(k )h(k 2n) (n) k g (k ) g (k 2n) (2.2)
k g ( k ) h( k 2n) 0 k h( k ) g ( k 2n)
Discrete Wavelet Transform
(Orthogonal Wavelet)
Assume that we have function s Vn
s ( x ) sn , k n , k ( x ) ,
k
s Pn 1 s Qn 1 s
Lemma:
n 1, j , n ,k hk 2, j ,
n 1, j , n ,k g k 2, j
Discrete Wavelet Transform
(Orthogonal Wavelet) (cont’d)
• Algorithm:
The original signal is s {s }. n n ,k
( g ( ) sn ) j g ( j k ) s n , k ,
k
which are followed by downsampling:
sn 1 ( 2)( h( ) * sn ) ,
d n 1 ( 2)( g () * sn ) .
Then the reconstruction step consists of upsampling, followed by two convolutions,
namely,
sn h * ( 2) sn 1 g * ( 2)d n 1 .
Discrete Wavelet Transform
(Orthogonal Wavelet) (cont’d)
s
s h h h h
n 1, 1 1 0 1 2 n , 1
sn 1, 0 h1 h0 h1 h2 sn , 0
s
n 1,1 h1 h0 h1 h2 s
n ,1
H and H’ are the low-pass filter and G and G’ are the high-pass filter.
Biorthogonal Wavelets (cont’d)
The first two conditions may be expressed in terms of the filter sequences h, h’, g, g’
which respectively define H, H’, G, G’:
k h' ( k ) h( k 2n) ( n) k g ' ( k ) g ( k 2n)
k g ' ( k ) h( k 2n) 0 k h' ( k ) g ( k 2n)
Notice the difference in (2.1) and (2.3) that self-duality no longer holds in (2.3) and the
conditions are weakened.
h( k ) 2 ;
k
g (2k ) g (2k 1) ;
k k
h' ( k )
k
2; g ' (2k ) g ' (2k 1) .
k k
Having four operators gives plenty of freedom to construct filters with special properties,
such as symmetry.
Discrete Wavelet Transform
(Biorthogonal Wavelet)
Assume that we have function s Vn
s( x) sn*, kn ,k ( x) ,
k
s Pn 1 s Qn 1 s
Lemma:
~
n 1, j , n ,k hk 2, j ,
n 1, j ,~n , k g k 2, j ,
~ ~
n 1, j , n ,k hk 2, j ,
~ , g~
n 1, j n,k k 2, j
Discrete Wavelet Transform
(Biorthogonal Wavelet) (cont’d)
• Algorithm:
The original signal is sn {snk }.
The decomposition is done as follows:
~ g~
s h s
n 1, j
d
k 2 j nk
s n 1, j k 2 j nk
k
k
( g~ () sn ) j g~( j k ) sn ,k ,
k
which are followed by downsampling:
~
s n 1 ( 2)( h () * s n ) ,
d ( 2)( g~ () * s ) .
n 1 n
Then the reconstruction step consists of upsampling, followed by two convolutions,
namely,
sn h* * ( 2) sn 1 g * * ( 2)d n 1 .
Discrete Wavelet Transform
(Biorthogonal Wavelet) (cont’d)
~ ~ ~ ~
s n 1, 1 h1 h0 h1 h2 s n , 1
~ ~ ~ ~
s n 1, 0 h1 h0 h1 h2 s n , 0
~ ~ ~ ~
s
n 1,1 h 1 h 0 h1 h2 s
n ,1
fB1,1 fB 2 ,1 fB M ,1
fB1, 2 fB 2 , 2 fB M , 2
fB
fB fB 2 , N fB M , N
1, N
Wavelets on Images
• A discrete image can be viewed as a
single column consisting of M signals
having length N, f M
f
f
2
f
1
1 A. Skodras, C. Christopoulos, and T. Ebrahimi, “ JPEG 200 Still Image Compression Standard,”
IEEE Signal Processing Magazine, vol. 18, pp. 36-58, Sept,
33
Implementation (cont’d)
In the case of lossless compression equations (3) and (4) were used.
R 2G b Ur Vr
Yr 4
Yr 4
G
Vr
R G (3) R Vr G
(4)
Ur BG
B Ur G
1-level Wavelet Decomposition
• A 1-level wavelet transform of an NxM image can be represented
as
a | h
1 1 a 1
V m
1
V 1
n h 1
V 1
m W n
1
a 1
( x , y ) ( x ) ( y ) h 1
H
( x, y ) ( x) ( y)
f
v1 | d1 v1 Wm1 Vn1 d 1 Wm1 Wn1
v1 V ( x, y ) ( x) ( y ) d 1 D ( x, y ) ( x) ( y )
where the subimages h1, d1, a1, and v1 each have the dimension
of N/2 by M/2.
a1 : first averaged image, which consists of average intensity
values of the original image.
v1 : the first detail image of vertical components.
h1 : the first detail image of horizontal components.
d1 : the first detail image of diagonal components.
1-level Wavelet Decomposition
(cont’d)
a 1 ( x, y ) ( x) ( y ) hi h j (2 x i) (2 y j ) Vm1 Vn1
i j
ch1
ca1
H1H
1
cv1 cd1
Octagon
Wavelet Decomposition
Wavelet Decomposition
Wavelet Decomposition
Test Image
Wavelet Decomposition
Wavelet Decompostion
Strange Circle
ca1 ch1
H 1
cv1 cd1
From the definition MRA, any wavelet (x) that comes from an MRA must satisfy
( x)dx 0
(*).
The integral (*) is referred to as the zeroth moment of (x)
If (*) holds, we say that (x) has its zeroth moment vanishing.
The integral
x k ( x) dx is referred to as the kth moment of (x) .
If x k ( x)dx 0 , we say that (x ) has its kth moment vanishing.
Vanishing Moments
• Artifacts such as wave-shaped
irregularities in the reconstructed image
could occur with the number of vanishing
moments.
• Results in lots of zero values for the
wavelet coefficients, which leads to
efficient coding.
Size of Filters
• Long analysis filters
- greater computation time for the wavelet
transform.
- can create unpleasant artifacts in the
compressed image .
Lena
lena.png
1
ratio = X * Y * 3 ( L2 norm _ re cov ery _ in % * X * Y * 3 / 100)
X *Y * 3
n1 j 1 i1
3 Y X
rms =
( f i , j , n g i , j , n ) 2
X *Y * 3
Results (cont’d)
n1 j 1 i1
3 Y X
( f i , j , n gi , j , n ) 2
D=
Y X 2
n 1 j 1 i 1
f i, j, n
255
PSNR 20 log
rms
Thr=10
Results(cont’d)
Thr=20
Wavelet ratio rms D PSNR Wavelet ratio rms D PSNR
Coif1 37.0173 6.3527 0.0450 32.0716 Coif1 93.0199 8.3876 0.0594 29.6580
Coif3 26.8321 6.1489 0.0436 32.3549 Coif3 51.7268 8.0346 0.0569 30.0316
Coif5 17.9320 6.1404 0.0435 32.3669 Coif5 30.4181 8.0154 0.0568 30.0523
bior1.1 34.2627 6.8373 0.0484 31.4331 bior1.1 95.8129 9.1988 0.0652 28.8562
bior2.2 30.2723 6.0565 0.0429 32.4863 bior2.2 73.2145 7.7818 0.0551 30.3092
bior5.5 43.9978 6.8879 0.0488 31.3691 bior5.5 104.6662 9.3134 0.0660 28.7487
sym2 38.4340 6.3633 0.0451 32.0571 sym2 100.6590 8.3909 0.0594 29.6546
sym5 34.3523 6.1740 0.0437 32.3195 sym5 76.9412 8.0751 0.0572 29.9878
sym8 28.2633 6.1511 0.0436 32.3518 sym8 56.9009 8.0363 0.0569 30.0296
db1 34.2627 6.8373 0.0484 31.4331 db1 95.8129 9.1988 0.0652 28.8562
db2 38.4340 6.3633 0.0451 32.0571 db2 100.6590 8.3909 0.0594 29.6546
db3 38.2236 6.2446 0.0442 32.2208 db3 94.7768 8.1916 0.0580 29.8635
References
[1] Ingrid Daubechies, Ten Lectures on Wavelets, SIAM, 1992.
[2] R.C. Gonzalez, R.E. Woods and S. L. Eddins, Digital Image Processing
Using MATLAB, . Prentice Hall, 2004.
[3] A. Skodras, C. Christopoulos, and T. Ebrahimi, “JPEG 2000 Still Image
Compression Standard,” IEEE Signal Processing Magazine, vol. 18, pp. 36-
58, Sept, 2001.
[4] James S. Walker, A Primer on Wavelets and their Scientific Applications,
Chapman Hall, CRC, 1999.
[5] David F. Walnut, An Introduction to Wavelet Analysis, Birkhäuser, 2002.
[6]MATLAB Wavelet Toolkit User Guide
http://www.mathworks.com/access/helpdesk/help/pdf_doc/wavelet/wavelet_
ug.pdf
[7] Fritz Keinert, Wavelets and Multiwavelets, Chapman & Hall/CRC, 2004.
[8] Myung-Sin Song, “Wavelet Image Compression”, Proceedings of the 2005
Symposium on Great Plains Operator Theory, AMS Contemporary
Mathematics book series
Daubechies Wavelet
• Daubechies wavelets are compactly-
supported, orthogonal wavelets which are
energy or norm preserving.
• It has moment-vanishing condition.
Haar Wavelet
• The Haar wavelet can be described as a step
function f(x) with
Haar Wavelet (cont’d)
• A discrete signal is expressed in the form
f f 1, f 2,......, fN
a1 (a1 , a 2 ,....., a N / 2 )
d1 (d1 , d 2 ,....., d N / 2 )
Haar Wavelet (cont’d)
• The l-level Haar transform H1 of discrete signal f
is defined as