Fourier Transform and Image Filtering - SCI Home
Transcript of Fourier Transform and Image Filtering - SCI Home
![Page 1: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/1.jpg)
Fourier Transform and
Image Filtering
CS/BIOEN 6640Lecture Marcel Prastawa
Fall 2010
![Page 2: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/2.jpg)
The Fourier Transform
![Page 3: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/3.jpg)
Fourier Transform
• Forward, mapping to frequency domain:
• Backward, inverse mapping to time domain:
![Page 4: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/4.jpg)
Fourier Series
• Projection or change of basis• Coordinates in Fourier basis:
• Rewrite f as:
![Page 5: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/5.jpg)
Example: Step FunctionStep function as sum of infinite sine waves
![Page 6: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/6.jpg)
Discrete Fourier Transform
![Page 7: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/7.jpg)
Fourier Basis
• Why Fourier basis?
• Orthonormal in [-pi, pi]• Periodic• Continuous, differentiable basis
![Page 8: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/8.jpg)
FT Properties
![Page 9: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/9.jpg)
Common Transform PairsDirac delta - constant
![Page 10: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/10.jpg)
Common Transform PairsRectangle – sinc
sinc(x) = sin(x) / x
![Page 11: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/11.jpg)
Common Transform PairsTwo symmetric Diracs - cosine
![Page 12: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/12.jpg)
Common Transform Pairs
Comb – comb (inverse width)
![Page 13: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/13.jpg)
Common Transform PairsGaussian – Gaussian (inverse variance)
![Page 14: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/14.jpg)
Common Transform Pairs Summary
![Page 15: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/15.jpg)
Quiz
What is the FT of a triangle function?
Hint: how do you get triangle function from the functions shown so far?
![Page 16: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/16.jpg)
Triangle Function FTTriangle = box convolved with box
So its FT is sinc * sinc
![Page 17: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/17.jpg)
Fourier Transform of Images
![Page 18: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/18.jpg)
• Forward transform:
• Backward transform:
• Forward transform to freq. yields complex values (magnitude and phase):
2D Fourier Transform
![Page 19: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/19.jpg)
2D Fourier Transform
![Page 20: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/20.jpg)
Fourier Spectrum
Fourier spectrumOrigin in corners
Retiled with originIn center
Log of spectrum
Image
![Page 21: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/21.jpg)
Fourier Spectrum–Rotation
![Page 22: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/22.jpg)
Phase vs Spectrum
Image Reconstruction fromphase map
Reconstruction fromspectrum
![Page 23: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/23.jpg)
Fourier Spectrum Demo
http://bigwww.epfl.ch/demo/basisfft/demo.html
![Page 24: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/24.jpg)
Low-Pass Filter• Reduce/eliminate high frequencies• Applications
– Noise reduction• uncorrelated noise is broad band• Images have sprectrum that focus on low
frequencies
86%
88%
90%
92%
94%
96%
98%
100%
0% 10% 20% 30% 40% 50% 60% 70% 80%
![Page 25: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/25.jpg)
Ideal LP Filter – Box, Rect
Cutoff freq Ringing – Gibbs phenomenon
![Page 26: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/26.jpg)
Extending Filters to 2D (or higher)
• Two options– Separable
• H(s) -> H(u)H(v)• Easy, analysis
– Rotate• H(s) -> H((u2 + v2)1/2)• Rotationally invariant
![Page 27: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/27.jpg)
Ideal LP Filter – Box, Rect
![Page 28: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/28.jpg)
Ideal Low-Pass Rectangle With Cutoff of 2/3
Image Filtered Filtered + HE
![Page 29: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/29.jpg)
Ideal LP – 1/3
![Page 30: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/30.jpg)
Ideal LP – 2/3
![Page 31: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/31.jpg)
Butterworth Filter
Control of cutoff and slopeCan control ringing
![Page 32: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/32.jpg)
Butterworth - 1/3
![Page 33: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/33.jpg)
Butterworth vs Ideal LP
![Page 34: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/34.jpg)
Butterworth – 2/3
![Page 35: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/35.jpg)
Gaussian LP FilteringILPF BLPF GLPF
F1
F2
![Page 36: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/36.jpg)
High Pass Filtering
• HP = 1 - LP– All the same filters as HP apply
• Applications– Visualization of high-freq data (accentuate)
• High boost filtering– HB = (1- a) + a(1 - LP) = 1 - a*LP
![Page 37: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/37.jpg)
High-Pass Filters
![Page 38: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/38.jpg)
High-Pass Filters in Spatial Domain
![Page 39: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/39.jpg)
High-Pass Filtering with IHPF
![Page 40: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/40.jpg)
BHPF
![Page 41: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/41.jpg)
GHPF
![Page 42: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/42.jpg)
HP, HB, HE
![Page 43: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/43.jpg)
High Boost with GLPF
![Page 44: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/44.jpg)
High-Boost Filtering
![Page 45: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/45.jpg)
Band-Pass Filters
• Shift LP filter in Fourier domain by convolution with delta
LP
BPTypically 2-3 parameters-Width-Slope-Band value
![Page 46: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/46.jpg)
Band Pass - Two Dimensions
• Two strategies– Rotate
• Radially symmetric– Translate in 2D
• Oriented filters
• Note:– Convolution with delta-pair in FD is
multiplication with cosine in spatial domain
![Page 47: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/47.jpg)
Band Bass Filtering
![Page 48: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/48.jpg)
SEM Image and Spectrum
![Page 49: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/49.jpg)
Band-Pass Filter
![Page 50: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/50.jpg)
Radial Band Pass/Reject
![Page 51: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/51.jpg)
Band Reject Filtering
![Page 52: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/52.jpg)
Band Reject Filtering
![Page 53: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/53.jpg)
Band Reject Filtering
![Page 54: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/54.jpg)
Aliasing
![Page 55: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/55.jpg)
Discrete Sampling and Aliasing
• Digital signals and images are discrete representations of the real world – Which is continuous
• What happens to signals/images when we sample them?– Can we quantify the effects? – Can we understand the artifacts and can we limit
them?– Can we reconstruct the original image from the
discrete data?
![Page 56: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/56.jpg)
A Mathematical Model of Discrete Samples
Delta functional
Shah functional
![Page 57: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/57.jpg)
A Mathematical Model of Discrete Samples
Discrete signal
Samples from continuous function
Representation as a function of t• Multiplication of f(t) with Shah
• Goal– To be able to do a continuous Fourier
transform on a signal before and after sampling
![Page 58: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/58.jpg)
Fourier Series of A Shah Functional
u
![Page 59: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/59.jpg)
Fourier Transform of A Discrete Sampling
u
![Page 60: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/60.jpg)
Fourier Transform of A Discrete Sampling
u
Energy from higher freqs gets folded back down into lower freqs –Aliasing
Frequencies get mixed. The original signal is not recoverable.
![Page 61: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/61.jpg)
What if F(u) is Narrower in the Fourier Domain?
u
• No aliasing!• How could we recover the original
signal?
![Page 62: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/62.jpg)
What Comes Out of This Model
• Sampling criterion for complete recovery
• An understanding of the effects of sampling– Aliasing and how to avoid it
• Reconstruction of signals from discrete samples
![Page 63: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/63.jpg)
Shannon Sampling Theorem
• Assuming a signal that is band limited:
• Given set of samples from that signal
• Samples can be used to generate the original signal– Samples and continuous signal are
equivalent
![Page 64: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/64.jpg)
Sampling Theorem• Quantifies the amount of information in
a signal– Discrete signal contains limited frequencies– Band-limited signals contain no more
information then their discrete equivalents• Reconstruction by cutting away the
repeated signals in the Fourier domain– Convolution with sinc function in
space/time
![Page 65: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/65.jpg)
Reconstruction
• Convolution with sinc function
![Page 66: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/66.jpg)
Sinc Interpolation Issues
• Must functions are not band limited• Forcing functions to be band-limited can
cause artifacts (ringing)
f(t) |F(s)|
![Page 67: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/67.jpg)
Sinc Interpolation Issues
Ringing - Gibbs phenomenonOther issues:
Sinc is infinite - must be truncated
![Page 68: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/68.jpg)
Aliasing• High frequencies appear as low
frequencies when undersampled
![Page 69: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/69.jpg)
Aliasing
16 pixels8 pixels
0.9174pixels
0.4798pixels
![Page 70: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/70.jpg)
Overcoming Aliasing
• Filter data prior to sampling– Ideally - band limit the data (conv with sinc
function)– In practice - limit effects with fuzzy/soft low
pass
![Page 71: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/71.jpg)
Antialiasing in Graphics
• Screen resolution produces aliasing on underlying geometry
Multiple high-res samples get averaged to create one screen sample
![Page 72: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/72.jpg)
Antialiasing
![Page 73: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/73.jpg)
Interpolation as Convolution
• Any discrete set of samples can be considered as a functional
• Any linear interpolant can be considered as a convolution– Nearest neighbor - rect(t)– Linear - tri(t)
![Page 74: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/74.jpg)
Convolution-Based Interpolation
• Can be studied in terms of Fourier Domain• Issues
– Pass energy (=1) in band– Low energy out of band– Reduce hard cut off (Gibbs, ringing)
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6
![Page 75: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/75.jpg)
Fast Fourier Transform
With slides from Richard Stern, CMU
![Page 76: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/76.jpg)
DFT
• Ordinary DFT is O(N2)• DFT is slow for large images
• Exploit periodicity and symmetricity• Fast FT is O(N log N)• FFT can be faster than convolution
![Page 77: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/77.jpg)
Fast Fourier Transform
• Divide and conquer algorithm• Gauss ~1805• Cooley & Tukey 1965
• For N = 2K
![Page 78: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/78.jpg)
The Cooley-Tukey Algorithm• Consider the DFT algorithm for an integer power of 2,
• Create separate sums for even and odd values of n:
• Letting for n even and for n odd, we obtain
N = 2ν
X[k] =n=0
N−1∑ x[n]WN
nk =n=0
N−1∑ x[n]e− j2πnk / N ; WN = e− j2π / N
X[k] = x[n]WNnk +
n even∑ x[n]WN
nk
n odd∑
n = 2r n = 2r +1
X[k] = x[2r]WN2rk +
r=0
N / 2( )−1∑ x[2r +1]WN
2r+1( )k
r=0
N /2( )−1∑
![Page 79: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/79.jpg)
The Cooley-Tukey Algorithm• Splitting indices in time, we have obtained
• But andSo …
N/2-point DFT of x[2r] N/2-point DFT of x[2r+1]
X[k] = x[2r]WN2rk +
r=0
N / 2( )−1∑ x[2r +1]WN
2r+1( )k
r=0
N /2( )−1∑
WN2 = e− j2π2 / N = e− j2π /(N / 2) = WN / 2 WN
2rkWNk = WN
kWN / 2rk
X[k] =n=0
(N/ 2)−1∑ x[2r]WN /2
rk + WNk
n=0
(N/ 2)−1∑ x[2r +1]WN / 2
rk
![Page 80: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/80.jpg)
Example: N=8
• Divide and reuse
![Page 81: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/81.jpg)
Example: N=8, Upper Part
• Continue to divide and reuse
![Page 82: Fourier Transform and Image Filtering - SCI Home](https://reader036.fdocuments.in/reader036/viewer/2022090906/613ca34cf046235e845ce80f/html5/thumbnails/82.jpg)
Two-Point FFT• The expression for the 2-point DFT is:
• Evaluating for we obtain
which in signal flowgraph notation looks like ...
X[k] =n=0
1∑ x[n]W2
nk =n=0
1∑ x[n]e− j2πnk / 2
k = 0,1X[0] = x[0]+ x[1]
X[1] = x[0] + e− j2π1/ 2x[1] = x[0]− x[1]
This topology is referred to as thebasic butterfly