Transcript Recap of Friday linear Filtering convolution differential filters filter types boundary conditions. Review: questions 1. Write down a 3x3 filter that returns a positive value if the.

Recap of Friday
linear Filtering
convolution
differential filters
filter types
boundary conditions.
Review: questions
1. Write down a 3x3 filter that returns a positive
value if the average value of the 4-adjacent
neighbors is less than the center and a
negative value otherwise
2. Write down a filter that will compute the
gradient in the x-direction:
gradx(y,x) = im(y,x+1)-im(y,x) for each x, y
Slide: Hoiem
Review: questions
3. Fill in the blanks:
a)
b)
c)
d)
_
A
F
_
=
=
=
=
D
_
D
D
*
*
*
*
Filtering Operator
B
A
_
_
D
B
E
G
C
F
H
I
D
Slide: Hoiem
The Frequency Domain (Szeliski 3.4)
Somewhere in Cinque Terre, May 2005
Slides from Steve Seitz
and Alexei Efros
CS129: Computational Photography
James Hays, Brown, Spring 2011
Salvador Dali
“Gala Contemplating the Mediterranean Sea,
which at 30 meters becomes the portrait
of Abraham Lincoln”, 1976
A nice set of basis
Teases away fast vs. slow changes in the image.
This change of basis has a special name…
Jean Baptiste Joseph Fourier (1768-1830)
had crazy idea (1807):
Any periodic function
can be rewritten as a
weighted sum of sines
and cosines of different
frequencies.
Don’t believe it?
• Neither did Lagrange,
Laplace, Poisson and
other big wigs
• Not translated into
English until 1878!
But it’s true!
• called Fourier Series
A sum of sines
Our building block:
Asin(x   
Add enough of them to get
any signal f(x) you want!
How many degrees of
freedom?
What does each control?
Which one encodes the
coarse vs. fine structure of
the signal?
Fourier Transform
We want to understand the frequency  of our signal. So,
let’s reparametrize the signal by  instead of x:
f(x)
Fourier
Transform
F()
For every  from 0 to inf, F() holds the amplitude A
and phase  of the corresponding sine Asin(x   
• How can F hold both? Using complex numbers.
F ( )  R( )  iI ( )
2
2
1 I ( )
A   R( )  I ( )
  tan
R( )
We can always go back:
F()
Inverse Fourier
Transform
f(x)
Time and Frequency
example : g(t) = sin(2pf t) + (1/3)sin(2p(3f) t)
Time and Frequency
example : g(t) = sin(2pf t) + (1/3)sin(2p(3f) t)
=
+
Frequency Spectra
example : g(t) = sin(2pf t) + (1/3)sin(2p(3f) t)
=
+
Frequency Spectra
Usually, frequency is more interesting than the phase
Frequency Spectra
=
=
+
Frequency Spectra
=
=
+
Frequency Spectra
=
=
+
Frequency Spectra
=
=
+
Frequency Spectra
=
=
+
Frequency Spectra

1
= A sin(2 kt )
k 1 k
Frequency Spectra
Extension to 2D
in Matlab, check out: imagesc(log(abs(fftshift(fft2(im)))));
Man-made Scene
Can change spectrum, then reconstruct
Low and High Pass filtering
The Convolution Theorem
• The Fourier transform of the convolution of two
functions is the product of their Fourier transforms
F[ g  h]  F[ g ] F[h]
• Convolution in spatial domain is equivalent to
multiplication in frequency domain!
2D convolution theorem example
|F(sx,sy)|
f(x,y)
*
h(x,y)
|H(sx,sy)|
g(x,y)
|G(sx,sy)|
Filtering in frequency domain
FFT
FFT
=
Inverse FFT
Slide: Hoiem
FFT in Matlab
• Filtering with fft
im = double(imread(‘…'))/255;
im = rgb2gray(im); % “im” should be a gray-scale floating point image
[imh, imw] = size(im);
hs = 50; % filter half-size
fil = fspecial('gaussian', hs*2+1, 10);
fftsize = 1024; % should be order of 2 (for speed) and include
im_fft = fft2(im, fftsize, fftsize);
% 1)
fil_fft = fft2(fil, fftsize, fftsize);
% 2)
image
im_fil_fft = im_fft .* fil_fft;
% 3)
im_fil = ifft2(im_fil_fft);
% 4)
im_fil = im_fil(1+hs:size(im,1)+hs, 1+hs:size(im, 2)+hs); % 5)
padding
fft im with padding
fft fil, pad to same size as
multiply fft images
inverse fft2
remove padding
• Displaying with fft
figure(1), imagesc(log(abs(fftshift(im_fft)))), axis image, colormap jet
Slide: Hoiem
Fourier Transform pairs
Low-pass, Band-pass, High-pass filters
low-pass:
High-pass / band-pass:
Edges in images
What does blurring take away?
original
What does blurring take away?
smoothed (5x5 Gaussian)
High-Pass filter
smoothed – original
Image gradient
The gradient of an image:
The gradient points in the direction of most rapid change in intensity
The gradient direction is given by:
• how does this relate to the direction of the edge?
The edge strength is given by the gradient magnitude
Effects of noise
Consider a single row or column of the image
• Plotting intensity as a function of position gives a signal
How to compute a derivative?
Where is the edge?
Solution: smooth first
Where is the edge? Look for peaks in
Derivative theorem of convolution
This saves us one operation:
2D edge detection filters
Laplacian of Gaussian
Gaussian
derivative of Gaussian
is the Laplacian operator:
Campbell-Robson contrast sensitivity curve
Depends on Color
R
G
B
Lossy Image Compression (JPEG)
Block-based Discrete Cosine Transform (DCT)
Using DCT in JPEG
The first coefficient B(0,0) is the DC component,
the average intensity
The top-left coeffs represent low frequencies,
the bottom right – high frequencies
Image compression using DCT
DCT enables image compression by
concentrating most image information in the
low frequencies
Lose unimportant image info (high frequencies)
by cutting B(u,v) at bottom right
The decoder computes the inverse DCT – IDCT
•Quantization Table
3
5
7
9
11 13 15 17
5
7
9
11 13 15 17 19
7
9
11 13 15 17 19 21
9
11 13 15 17 19 21 23
11 13 15 17 19 21 23 25
13 15 17 19 21 23 25 27
15 17 19 21 23 25 27 29
17 19 21 23 25 27 29 31
JPEG compression comparison
89k
12k
Things to Remember
Sometimes it makes sense to think of
images and filtering in the frequency
domain
• Fourier analysis
Can be faster to filter using FFT for
large images (N logN vs. N2 for autocorrelation)
Images are mostly smooth
• Basis for compression
Remember to low-pass before sampling
Summary
Frequency domain can be useful for
Analysis
Computational efficiency
Compression