2. 22.058 - lecture 4, Convolution and
Fourier Convolution
Convolution
Fourier Convolution
Outline
· Review linear imaging model
· Instrument response function vs Point spread function
· Convolution integrals
• Fourier Convolution
• Reciprocal space and the Modulation transfer function
· Optical transfer function
· Examples of convolutions
· Fourier filtering
· Deconvolution
· Example from imaging lab
• Optimal inverse filters and noise
school.edhole.com
3. Instrument Response Function
The Instrument Response Function is a conditional mapping, the form of the map
depends on the point that is being mapped.
{ }
IRF(x,y|x0,y0)=Sd(x-x0 )
22.058 - lecture 4, Convolution and
Fourier Convolution
This is often given the symbol h(r|r’).
Of course we want the entire output from the whole object function,
¥
òò
-¥
¥
òò
-¥
and so we need to know the IRF at all points.
d(y-y0 )
E(x,y)=
¥
òò
-¥
I(x,y)S{d(x-x0)d(y-y0)}dxdydx0dy0
E(x,y)=
¥
òò
-¥
I(x,y)IRF(x,y|x0,y0)dxdydx0dy0
school.edhole.com
4. IRF(x,y|x0,y0)=PSF(x-x0,y-y0 )
22.058 - lecture 4, Convolution and
Fourier Convolution
Space Invariance
Now in addition to every point being mapped independently onto the detector,
imaging that the form of the mapping does not vary over space (is independent of
r0). Such a mapping is called isoplantic. For this case the instrument response
function is not conditional.
The Point Spread Function (PSF) is a spatially invariant approximation of the IRF.
school.edhole.com
5. 22.058 - lecture 4, Convolution and
Fourier Convolution
Space Invariance
Since the Point Spread Function describes the same blurring over the entire sample,
The image may be described as a convolution,
or,
IRF(x,y|x0,y0)ÞPSF(x-x0,y-y0 )
¥
òò
-¥
E(x,y)=
I(x0,y0)PSF(x-x0,y-y0)dx0dy0
Image(x,y)=Object(x,y)ÄPSF(x,y)+noise
school.edhole.com
6. Convolution Integrals
Let’s look at some examples of convolution integrals,
¥ òg(x')h(x-x')dx'
f(x)=g(x)Äh(x)=
So there are four steps in calculating a convolution integral:
22.058 - lecture 4, Convolution and
Fourier Convolution
#1. Fold h(x’) about the line x’=0
#2. Displace h(x’) by x
#3. Multiply h(x-x’) * g(x’)
#4. Integrate
-¥
school.edhole.com
7. Convolution Integrals
22.058 - lecture 4, Convolution and
Fourier Convolution
Consider the following two functions:
#1. Fold h(x’) about the line x’=0
#2. Displace h(x’) by x x
school.edhole.com
11. Some Properties of the Convolution
22.058 - lecture 4, Convolution and
Fourier Convolution
commutative:
associative:
multiple convolutions can be carried out in any order.
distributive:
fÄg=gÄf
fÄ(gÄh)=(fÄg)Äh
fÄ(g+h)=fÄg+fÄh
school.edhole.com
12. Recall that we defined the convolution integral as,
¥ò
One of the most central results of Fourier Theory is the convolution
theorem (also called the Wiener-Khitchine theorem.
22.058 - lecture 4, Convolution and
Fourier Convolution
where,
Convolution Integral
fÄg=f(x)g(x'-x)dx
-¥
Á{fÄg}=F(k)×G(k)
f(x)ÛF(k)
g(x)ÛG(k)
school.edhole.com
13. Convolution Theorem
In other words, convolution in real space is equivalent to
multiplication in reciprocal space.
22.058 - lecture 4, Convolution and
Fourier Convolution
Á{fÄg}=F(k)×G(k)
school.edhole.com
14. Convolution Integral Example
We saw previously that the convolution of two top-hat functions
(with the same widths) is a triangle function. Given this, what is
the Fourier transform of the triangle function?
15 ?
22.058 - lecture 4, Convolution and
Fourier Convolution
15
10
5
-3 -2 -1 1 2 3
=
10
5
-3 -2 -1 1 2 3
school.edhole.com
15. Proof of the Convolution Theorem
¥ò
fÄg=f(x)g(x'-x)dx
¥ò
F(k)eikxdk
-¥
¥ò
G(k')eik'(x'-x)dk'
-¥
22.058 - lecture 4, Convolution and
Fourier Convolution
The inverse FT of f(x) is,
-¥
f(x)=1
2p
and the FT of the shifted g(x), that is g(x’-x)
g(x'-x)=1
2p
school.edhole.com
16. Proof of the Convolution Theorem
So we can rewrite the convolution integral,
¥ò
fÄg=f(x)g(x'-x)dx
¥ò
G(k')eik'(x'-x)dk'
-¥
¥ò
¥ò
dkF(k)dk'G(k')eik'x'1
eix(k-k')dx
-¥ 14 4 2 4 4 3
22.058 - lecture 4, Convolution and
Fourier Convolution
as,
-¥
¥ò
fÄg=1
¥ò
dxF(k)eikxdk
4p2
-¥
-¥
change the order of integration and extract a delta function,
fÄg=1
2p
2p
-¥
d(k-k')
¥ò
-¥
school.edhole.com
17. Proof of the Convolution Theorem
¥ò
¥ò
¥ò
dkF(k)dk'G(k')eik'x'1
eix(k-k')dx
2p
-¥ 14 -¥
4 2 4 4 3
d(k-k')
¥ò
¥ò
dkF(k)dk'G(k')eik'x'd
(k-k')
¥ò
dkF(k)G(k)eikx'
-¥
22.058 - lecture 4, Convolution and
Fourier Convolution
or,
fÄg=1
2p
-¥
fÄg=1
2p
-¥
-¥
Integration over the delta function selects out the k’=k value.
fÄg=1
2p
school.edhole.com
18. Proof of the Convolution Theorem
¥ò
dkF(k)G(k)eikx'
-¥
This is written as an inverse Fourier transformation. A Fourier
transform of both sides yields the desired result.
22.058 - lecture 4, Convolution and
Fourier Convolution
fÄg=1
2p
Á{fÄg}=F(k)×G(k)
school.edhole.com
20. 22.058 - lecture 4, Convolution and
Fourier Convolution
Reciprocal Space
real space reciprocal space
21. 22.058 - lecture 4, Convolution and
Fourier Convolution
Filtering
We can change the information content in the image by
manipulating the information in reciprocal space.
Weighting function in k-space.
school.edhole.com
22. 22.058 - lecture 4, Convolution and
Fourier Convolution
Filtering
We can also emphasis the high frequency components.
Weighting function in k-space.
school.edhole.com
23. Modulation transfer function
i(x,y) = o(x,y)ÄPSF(x,y) + noise
c c c c
I(kx,ky)=O(kx,ky)×MTF(kx,ky)+Á{noise}
22.058 - lecture 4, Convolution and
Fourier Convolution
¥
òò
-¥
E(x,y)=
¥
òò
-¥
I(x,y)S{d(x-x0)d(y-y0)}dxdydx0dy0
¥
òò
-¥
E(x,y)=
¥
òò
-¥
I(x,y)IRF(x,y|x0,y0)dxdydx0dy0
school.edhole.com
24. 22.058 - lecture 4, Convolution and
Fourier Convolution
Optics with lens
‡ Input bit mapped image
sharp =Import @" sharp . bmp" D;
Shallow @InputForm @sharp DD
Graphics@Raster@<<4>>D,Rule@<<2>>DD
s =sharp @@1, 1DD;
Dimensions @s D
8480,640<
ListDensityPlot @s , 8PlotRange school.edhole.com ÆAll , Mesh ÆFalse <D