Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform...
-
date post
22-Dec-2015 -
Category
Documents
-
view
217 -
download
2
Transcript of Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform...
![Page 1: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/1.jpg)
Multiscale transforms : wavelets, ridgelets, curvelets, etc.
Outline :• The Fourier transform
• Time-frequency analysis and the Heisenberg principle
• Cauchy Schwartz inequality
• The continuous wavelet transform
• 2D wavelet transform
• Anisotropic frames : Ridgelets, curvelets, etc.
![Page 2: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/2.jpg)
The Fourier transform (1)
• Diagonal representation of shift invariant linear transforms.
• Truncated Fourier series give very good approximations to smooth functions.
• Limitations : – Provides poor representation of non stationary signals or image.
– Provides poor representations of discontinuous objects (Gibbs effect)
![Page 3: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/3.jpg)
The Fourier transform (2)
• A Fourier transform is a change of basis.• Each dot product assesses the coherence between the signal and the basis
element.
• Cauchy-Schwartz :
• The Fourier basis is best for representing harmonic components of a signal!
![Page 4: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/4.jpg)
• Computational harmonic analysis seeks representations of s signal as linear combinations of basis, frame, dictionary, element :
• Analyze the signal through the statistical properties of the coefficients
• The analyzing functions (frame elements) should extract features of interest.
• Approximation theory wants to exploit the sparsity of the coefficients.
What is good representation for data?
basis, framecoefficients
![Page 5: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/5.jpg)
Seeking sparse and generic representations
• Sparsity
• Why do we need sparsity?– data compression – Feature extraction, detection– Image restoration
sorted index
few big
many small
![Page 6: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/6.jpg)
Candidate analyzing functions for piecewise smooth signals
• Windowed fourier transform or Gaborlets :
• Wavelets :
![Page 7: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/7.jpg)
Heisenberg uncertainty principle
• Different tilings in time frequency space :
• Localization in time and frequency requires a compromise
![Page 8: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/8.jpg)
Windowed/Short term Fourier transform
• Invertibility condition :
• Reconstruction :
• Decomposition :
with
( with a gaussian window w, this is the Gabor transform)
![Page 9: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/9.jpg)
The Continuous Wavelet Transform
• decomposition
• reconstruction
• admissible wavelet :
• simpler condition : zero mean wavelet
The CWT is a linear transform. It is covariant under translation and scaling. Verifies a Plancherel-Parceval type equation.
![Page 10: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/10.jpg)
Continuous Wavelet Transform
• Example : The mexican hat wavelet
![Page 11: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/11.jpg)
2D Continuous Wavelet transform• either a genuine 2D wavelet function (e.g. mexican hat)or a separable wavelet i.e. tensor product of two 1D wavelets.
• example :
Images obtained using the nearly isotropic undecimated wavelet transform obtained with the a trous algorithm.
![Page 12: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/12.jpg)
Wavelets and edges
• many wavelet coefficients are needed to account for edges ie singularities along lines or curves :
• need dictionaries of strongly anisotropic atoms :
ridgelets, curvelets, contourlets, bandelettes, etc.
![Page 13: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/13.jpg)
Continuous Ridgelet Transform
Ridgelet function:
The function is constant along lines. Transverse to these ridges, it is a wavelet.
Ridgelet Transform (Candes, 1998):
€
R f a,b,θ( ) = ψ a,b,θ∫ x( ) f x( )dx
€
ψa,b,θ x( ) = a1
2ψx1 cos(θ) + x2 sin(θ) − b
a
⎛
⎝ ⎜
⎞
⎠ ⎟
![Page 14: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/14.jpg)
The ridgelet coefficients of an object f are given by analysis
of the Radon transform via:
€
R f (a,b,θ) = Rf (θ, t)ψ (t − b
a∫ )dt
![Page 15: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/15.jpg)
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Example application of Ridgelets
![Page 16: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/16.jpg)
SNR = 0.1
![Page 17: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/17.jpg)
![Page 18: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/18.jpg)
Undecimated Wavelet Filtering (3 sigma)
![Page 19: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/19.jpg)
Ridgelet Filtering (5sigma)
![Page 20: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/20.jpg)
Local Ridgelet Transform
The ridgelet transform is optimal to find only lines of the size of the image.To detect line segments, a partitioning must be introduced. The image isdecomposed into blocks, and the ridgelet transform is applied on each block.
Image
Partitioning
Ridgelet transform
![Page 21: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/21.jpg)
In practice, we use overlap to avoid blocking artifacts.
The partitioning introduces a redundancy, as a pixel belongs to 4 neighboringblocks.
Smooth partitioning
Image
Ridgelettransform
![Page 22: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/22.jpg)
Edge Representation
Suppose we have a function f which has a discontinuity across a curve, andwhich is otherwise smooth, and consider approximating f from the best m-terms in the Fourier expansion. The squarred error of such an m-termexpansion obeys:
f–f m
F 2 m12 ,mŒƒ
f–f m
W 2 m 1 ,mŒƒ
In a wavelet expansion, we have
f–f m
C 2 log m 3m 2 ,mŒƒ
In a curvelet expansion (Donoho and Candes, 2000), we have
Width = Length^2
![Page 23: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/23.jpg)
The Curvelet Transform for Image Denoising, IEEE Transaction on Image Processing, 11, 6, 2002.
Numerical Curvelet Transform
![Page 24: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/24.jpg)
The Curvelet Transform
The curvelet transform opens us the possibility to analyse an image with different block sizes, but with a single transform.
The idea is to first decompose the image into a set of wavelet bands, andto analyze each band by a ridgelet transform. The block size can be changedat each scale level.
- à trous wavelet transform-Partitionning-ridgelet transform . Radon Transform . 1D Wavelet transform
![Page 25: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/25.jpg)
The Curvelet Transform
J.L. Starck, E. Candès and D. Donoho,"Astronomical Image Representation by the Curvelet Transform,Astronomy and Astrophysics, 398, 785--800, 2003.
![Page 26: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/26.jpg)
NGC2997
![Page 27: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/27.jpg)
A trous algorithm:
€
I(k, l) = cJ ,k,l + w j,k,lj=1
J
∑
![Page 28: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/28.jpg)
PARTITIONING
![Page 29: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/29.jpg)
CONTRAST ENHANCEMENT
Curvelet coefficient
Modifiedcurvelet coefficient
€
˜ I = CR yc CT I( )( )
€
yc (x,σ ) =x − cσ
cσ
m
cσ
⎛
⎝ ⎜
⎞
⎠ ⎟p
+2cσ − x
cσ€
yc (x,σ ) =1
€
yc (x,σ ) =m
x
⎛
⎝ ⎜
⎞
⎠ ⎟p
€
yc (x,σ ) =m
x
⎛
⎝ ⎜
⎞
⎠ ⎟s
if
if
if
if
€
x < cσ
€
x < 2cσ
€
2cσ ≤ x < m
€
x > m
€
{
![Page 30: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/30.jpg)
Contrast Enhancement
![Page 31: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/31.jpg)
F
![Page 32: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/32.jpg)
![Page 33: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/33.jpg)
![Page 34: Multiscale transforms : wavelets, ridgelets, curvelets, etc. Outline : The Fourier transform Time-frequency analysis and the Heisenberg principle Cauchy.](https://reader030.fdocuments.in/reader030/viewer/2022032523/56649d7e5503460f94a61d18/html5/thumbnails/34.jpg)
Critical Sampling Redundant Transforms
Pyramidal decomposition (Burt and Adelson) (bi-) Orthogonal WT Undecimated Wavelet Transform Lifting scheme construction Isotropic Undecimated Wavelet Transform Wavelet Packets Complex Wavelet Transform Mirror Basis Steerable Wavelet Transform Dyadic Wavelet Transform Nonlinear Pyramidal decomposition (Median)
Multiscale Transforms
New Multiscale Construction
Contourlet RidgeletBandelet Curvelet (Several implementations)Finite Ridgelet TransformPlatelet(W-)Edgelet Adaptive Wavelet