Professional Documents
Culture Documents
RoweTalkMSCS BiCubic
RoweTalkMSCS BiCubic
RoweTalkMSCS BiCubic
Department of
Biophysics
D.B. Rowe 1
MU MSCS Spring 2018
1.
Goal of Interpolation
1D Interpolation 1D Interpolation 2D Interpolation
2. Linear
3. Cubic
4. Cubic Spline
2D Interpolation
5.
BiLinear
6.
BiCubic
7.
BiCubic Spline MRI Example
8.
MRI Example
D.B. Rowe 2
MU MSCS Spring 2018
1. Goal of Interpolation
How do you determine how to get from one point to another?
D.B. Rowe 3
MU MSCS Spring 2018
1. Goal of Interpolation
How do you determine how to get from one point to another?
D.B. Rowe 4
MU MSCS Spring 2018
1. Goal of Interpolation
How do you determine how to get from one point to another?
true curve
D.B. Rowe 5
MU MSCS Spring 2018
2. 1D Linear Interpolation
Easiest to draw straight lines between points and use values
along the lines.
Interpolate: f (4.5)
Two points define a line. f (2.5) f (4)
f (1.5)
axis change
D.B. Rowe 6
MU MSCS Spring 2018
2. 1D Linear Interpolation
Easiest to draw straight lines between points and use values
along the lines.
f (0)
f (1)
axis change
D.B. Rowe 7
MU MSCS Spring 2018
2. 1D Linear Interpolation
Easiest to draw straight lines between points and use values
along the lines.
Model: f ( x) a0 x 0 a1 x1
x 0,1
f (0)
f (1)
D.B. Rowe 8
MU MSCS Spring 2018
2. 1D Linear Interpolation
Easiest to draw straight lines between points and use values
along the lines.
Model: f ( x) a0 x 0 a1 x1
x 0,1
Solve: (a0 , a1 )
f (0) a0 (1) a1 (0)
f (0)
f (1) a0 (1) a1 (1) f (1)
D.B. Rowe 9
MU MSCS Spring 2018
2. 1D Linear Interpolation
System of Equations: 2 equations, 2 unknowns f (0) a0 (1) a1 (0)
f (1) a0 (1) a1 (1)
D.B. Rowe 10
MU MSCS Spring 2018
2. 1D Linear Interpolation
Easiest to draw straight lines between points and use values
along the lines.
Model: f ( x) a0 x 0 a1 x1
x 0,1
Solve: (a0 , a1 )
Interpolate: .5 f (0)
f (1)
D.B. Rowe 13
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
values between.
Model: f ( x) a0 x 0 a1 x1
x 0,1
Solve: (a0 , a1 ) f (1)
Interpolate: .5
f (0)
D.B. Rowe 14
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
values between.
x 0,1
Solve: (a0 , a1 )
f (0)
Interpolate: .5
D.B. Rowe 15
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
values between.
x 0,1
f (1)
Solve: (a0 , a1 )
Interpolate: .5
D.B. Rowe 16
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
values between.
Model: f ( x) a0 x 0 a1 x1
f (0)
x 0,1
Solve: (a0 , a1 )
f (1)
Interpolate: .5
D.B. Rowe 17
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
values between.
Model: f ( x) a0 x 0 a1 x1 f (1)
x 0,1
Solve: (a0 , a1 )
f (0)
Interpolate: .5
D.B. Rowe 18
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
values between.
Model: f ( x) a0 x 0 a1 x1
x 0,1 f (0)
Solve: (a0 , a1 )
Interpolate: .5
D.B. Rowe 19
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
all values between.
Interpolate at 0<x<1
f ( x) a0 x 0 a1 x1
D.B. Rowe 20
MU MSCS Spring 2018
2. 1D Linear Interpolation
Repeat the process between all pairs of points to interpolate
all values between.
D.B. Rowe 21
MU MSCS Spring 2018
3. 1D Cubic Interpolation
Using adjacent points, we can estimate a cubic (third order
polynomial) between points.
Interpolate: f (4.5)
Four points define a f (2.5) f (4)
cubic equation.
f (3)
f ( x) a0 x 0 a1 x1 a2 x 2 a3 x 3 f (2)
f (3.5)
Find the coefficients of the
f (1)
cubic eqn. between points.
f (1.5)
D.B. Rowe 22
MU MSCS Spring 2018
3. 1D Cubic Interpolation
Using adjacent points, we can estimate a cubic (third order
polynomial) between points.
f (1)
f (1)
f (0)
D.B. Rowe 23
MU MSCS Spring 2018
3. 1D Cubic Interpolation
Using adjacent points, we can estimate a cubic (third order
polynomial) between points.
f (1)
f (0)
D.B. Rowe 24
MU MSCS Spring 2018
3. 1D Cubic Interpolation
Using adjacent points, we can estimate a cubic (third order
polynomial) between points. 4 equations, 4 unknowns
D.B. Rowe 25
MU MSCS Spring 2018 f (1) a0 (1)0 a1 (1)1 a2 (1) 2 a3 (1)3
f (0) a0 (0)0 a1 (0)1 a2 (0) 2 a3 (0)3
3. 1D Cubic Interpolation f (1) a0 (1)0 a1 (1)1 a2 (1) 2 a3 (1)3
f (2) a0 (2)0 a1 (2)1 a2 (2) 2 a3 (2)3
System of Equations: 4 equations, 4 unknowns
recycle matrix
D.B. Rowe 26
MU MSCS Spring 2018
3. 1D Cubic Interpolation
System of Equations: 4 equations, 4 unknowns
Solution Solution
a0 0 6 0 0 f ( 1)
a 2 3 6 1 f (0) a X 1 y
1 1
a2 6 3 6 3 0 f (1)
3
a 1 3 3 1 f (2)
Interpolate at 0<x<1 Interpolate at 0<x<1
a0
f ( x) a0 x 0 a1 x1 a2 x 2 a3 x 3 a
f ( x) [ x 0 , x1 , x 2 , x 3 ] 1
a2
a3
D.B. Rowe 27
MU MSCS Spring 2018
3. 1D Cubic Interpolation
System of Equations: 4 equations, 4 unknowns
D.B. Rowe 28
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
f (0)
D.B. Rowe 29
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
D.B. Rowe 30
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
D.B. Rowe 31
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
D.B. Rowe 32
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
Model: f ( x) a0 x 0 a1 x1 a2 x 2 a3 x 3 f (1)
x 1,0,1, 2 f (1)
Solve: (a0 , a1 , a2 , a3 )
Interpolate: f (0)
D.B. Rowe 33
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
interpolation
not done
Interpolation not done
-ends of cubic
-linear ends
-quadratic ends
D.B. Rowe 34
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
width
disparity interpolation
“kink” not done
D.B. Rowe 35
MU MSCS Spring 2018
3. 1D Cubic Interpolation
This involves 4 unknowns and 4 points to estimate them, then
points along the polynomial.
width
disparity interpolation
“kink” not done
D.B. Rowe 36
MU MSCS Spring 2018
Interpolate: f (4.5)
Two points plus two f (2.5) f (4)
derivatives. No “kinks?”
f (3)
Smooth transition through. f (2)
f (3.5)
Find the equation of the
f (1)
cubic eqn. between points. Only use two points
and their derivatives.
f (1.5)
D.B. Rowe 37
MU MSCS Spring 2018
f (1)
f (0)
D.B. Rowe 38
MU MSCS Spring 2018
f (0)
D.B. Rowe 39
MU MSCS Spring 2018
D.B. Rowe 40
MU MSCS Spring 2018
Derivative at x=0:
f (1)
f (1) f (1)
f '(0)
2
Derivative at x=1:
f (2) f (0) f (0)
f '(1)
2
D.B. Rowe 41
MU MSCS Spring 2018
recycle matrix
D.B. Rowe 42
MU MSCS Spring 2018
D.B. Rowe 43
MU MSCS Spring 2018
Interpolate: .5
D.B. Rowe 44
MU MSCS Spring 2018
Interpolate: .5
D.B. Rowe 45
MU MSCS Spring 2018
Interpolate: .5
D.B. Rowe 46
MU MSCS Spring 2018
D.B. Rowe 47
MU MSCS Spring 2018
D.B. Rowe 48
MU MSCS Spring 2018
interpolation
not done
D.B. Rowe 49
MU MSCS Spring 2018
interpolation
not done
no “kink”
D.B. Rowe 50
MU MSCS Spring 2018
interpolation
not done
no “kink”
D.B. Rowe 51
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Easiest to draw planes between points and use values within
the planes.
f (1,0)
f (0,0) f (1,1)
f (0,1)
D.B. Rowe 54
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Easiest to draw planes between points and use values within
the planes.
Interpolate:
f (1,0)
Find the equation of the
plane between points.
f (0,0) f (1,1)
f (0,1)
D.B. Rowe 55
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Easiest to draw planes between points and use values within
the planes.
f (0,0) f (1,1)
f (0,1)
D.B. Rowe 56
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Easiest to draw planes between points and use values within
the planes.
D.B. Rowe 57
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Easiest to draw planes between points and use values within
the planes.
D.B. Rowe 58
MU MSCS Spring 2018
recycle matrix
D.B. Rowe 59
MU MSCS Spring 2018
Solution a X 1 y Interpolate
0 1
a00 1 0 0 0 f (0,0) f ( x, y ) a00 a10 x a01 y a11 xy
a 1 1 0 0 f (1,0) 0
01 0<x<1, 0<y<1
a10 1 0 1 0 f (0,1) 1
a11 1 1 1 1 f (1,1)
f (.5,0) 1 .5 0 0
Interpolate one pixel yint X int a f (1,.5) 1 a00
1 .5 .5
f (.5,0) [1,.5,0,0]a a01
f (.5,.5) 1 .5 .5 .25
More rows to interpolate more pixels. a10
f (0,.5) 1 0 .5 0
f (.5,1) 1 .5 1 .5 a11
D.B. Rowe 60
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Easiest to draw planes between points and use values within
the planes.
f (0,0) f (.5,.5)
f (1,1)
f ( x, y ) a00 a10 x a01 y a11 xy f (0,.5)
f (.5,1)
f (0,1)
f (.5,0) f (1,.5)
f (.5,.5)
f (0,.5) f (.5,1)
D.B. Rowe 61
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Example: 8×8 interpolate 1 to 15×15
D.B. Rowe 62
MU MSCS Spring 2018
5. 2D BiLinear Interpolation
Example: 8×8 interpolate 1001 to 7015×7015
Expanded
D.B. Rowe 63
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Need to use adjacent points, estimate surfaces, and use values
within the surfaces.
Interpolate:
Sixteen points define a
2D bicubic surface.
Find the equation of the
surface between 4 points
using neighbors.
D.B. Rowe 64
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Need to use adjacent points, estimate surfaces, and use values
within the surfaces.
f (1,0)
f (0,0) f (1,1)
f (0,1)
D.B. Rowe 65
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Need to use adjacent points, estimate surfaces, and use values
within the surfaces.
j 0 i 0
f (1,0)
f (0,0) f (1,1)
x 1,0,1, 2 f (0,1)
D.B. Rowe 66
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Need to use adjacent points, estimate surfaces, and use values
within the surfaces.
j 0 i 0
f (1,0)
f (0,0) f (1,1)
x 1,0,1, 2 f (0,1)
Solve: aij
D.B. Rowe 67
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Need to use adjacent points, estimate surfaces, and use values
within the surfaces.
Interpolate:
Find the equation of the
surface between 4 points f (1,0)
f (0,0) f (1,1)
using neighbors and f (0,1)
D.B. Rowe 68
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
System of Equations: 16 equations, 16 unknowns
-1 0 1 2
System of Equations y Xa Image I ( x, y )
-1
16 equations from f ( x, y )
3 3
0
f ( x, y ) aij x i y j
i 0 j 0
1
x,y=-1,0,1,2
2
Simply insert all x,y combinations to get 16 equations.
D.B. Rowe 69
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Values from polynomial.
y Xa
f (1, 1) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a00
f (0, 1) 1
0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 a10 3 3
f (1, 1) 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a20
f ( x, y ) aij x i y j
f (2, 1) 1 2 4 8 1 2 4 8 1 2 4 8 1 2 4 8 a30 i 0 j 0
f (1,0) 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 a01
x,y=-1,0,1,2
f (0,0) 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a11
f (1,0) 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 a21
f (2,0) 1 2 4 8 0 0 0 0 0 0 0 0 0 0 0 0 a31
f (1,1) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a02
f (0,1) 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 a12
f (1,1) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a22
f (2,1) 1 2 4 8 1 2 4 8 1 2 4 8 1 2 4 8 a32
f (1, 2) 1 1 1 1 2 2 2 2 4 4 4 4 8 8 8 8 a03
f (0, 2) 1 0 0 0 2 0 0 0 4 0 0 0 8 0 0 0 a13
f (1, 2) 1 1 1 1 2 2 2 2 4 4 4 4 8 8 8 8 a23
f (2, 2) 1
2 4 8 2 4 8 16 4 8 16 32 8 16 32 64 a33
D.B. Rowe 70
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Estimate coefficient values. Almost Money Slide
a X 1I
a00 0 0 0 0 0 36 0 0 0 0 0 0 0 0 0 0 I (1, 1)
a 0 0 0 0 12 18 36 6 0 0 0 0 0 0 0 0 I (0, 1)
10
a20 0 0 0 0 18 36 18 0 0 0 0 0 0 0 0 0 I (1, 1)
a30 0 0 0 0 6 18 18 6 0 0 0 0 0 0 0 0 I (2, 1)
a01 0 12 0 0 0 18 0 0 0 36 0 0 0 6 0 0 I (1,0)
a11 4 6 12 2 6 9 18 3 12 18 36 6 2 3 6 1 I (0,0)
a 6 12 6 0 9 18 9 0 18 36 18 0 3 6 3 0 I (1,0)
21
a31 1 2 6 6 2 3 9 9 3 6 18 18 6 1 3 3 1 I (2,0)
a 36 0 18 0 0 0 36 0 0 0 18 0 0 0 0 0 0 I (1,1)
02
a12 6 9 18 3 12 18 36 6 6 9 18 3 0 0 0 0 I (0,1)
a22 9 18 9 0 18 36 18 0 9 18 9 0 0 0 0 0 I (1,1)
a32 3 9 9 3 6 18 18 6 3 9 9 3 0 0 0 0 I (2,1)
a03 0 6 0 0 0 18 0 0 0 18 0 0 0 6 0 0 I (1, 2)
a 2 3 6 1 6 9 18 3 6 9 18 3 2 3 6 1 I (0, 2)
13
a23 3 6 3 0 9 18 9 9 9 18 9 0 3 6 3 0 I (1, 2)
a 1 3 3 1 3 9 9 9 3 1 3 3 1 I (2, 2)
33 9 3 3
recycle matrix
D.B. Rowe 71
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Estimate coefficient values.
a X 1I
D.B. Rowe 72
MU MSCS Spring 2018
6. 2D BiCubic Interpolation
Example: 8×8 interpolate 1001 to 7015×7015
Expanded
* bilinear edges
Low Resolution Bilinear Interpolated BiCubic Interpolated
D.B. Rowe 73
MU MSCS Spring 2018
Interpolate:
Will use 4 points and 12
derivatives at those points
to define a bicubic surface.
D.B. Rowe 74
MU MSCS Spring 2018
f (1,0)
f (0,0) f (1,1)
f (0,1)
D.B. Rowe 75
MU MSCS Spring 2018
j 0 i 0
f (1,0)
f (0,0) f (1,1)
f (0,1)
D.B. Rowe 76
MU MSCS Spring 2018
j 0 i 0
f (1,0)
f (0,0) f (1,1)
f (0,1)
Solve: aij
D.B. Rowe 77
MU MSCS Spring 2018
D.B. Rowe 78
MU MSCS Spring 2018
-1 0 1 2
System of Equations y Xa Image I ( x, y )
-1
4 equations from f ( x, y )
3 3
0
f ( x, y ) aij x i y j
i 0 j 0
1
x,y=0,1
f (0,0) a00 2
f (1,0) a00 a10 a20 a30
f (0,1) a00 a01 a02 a03
f (1,1) a00 a10 a01 a11
D.B. Rowe 79
MU MSCS Spring 2018
-1 0 1 2
System of Equations y Xa Image I ( x, y )
-1
4 equations from f x ( x, y ) f ( x, y )
x 0
3 3
f x ( x, y ) aij ix i 1 y j
j 0 i 1 1
x,y=0,1
f x (0,0) a10 2
f x (1,0) 1a10 2a20 3a30
f x (0,1) a10 a11 a12 a13
f x (1,1) 1a10 2a20 3a31 1a11 2a21 3a31
1a12 2a22 3a32 1a13 2a23 3a33
D.B. Rowe 80
MU MSCS Spring 2018
-1 0 1 2
System of Equations y Xa Image I ( x, y )
-1
4 equations from f y ( x, y ) f ( x, y )
y 0
3 3
f y ( x, y ) aij jx i y j 1
j 1 i 0
1
x,y=0,1
f y (0,0) a01 2
f y (1,0) a01 a11 a21 a31
f y (0,1) 1a01 2a02 3a03
f y (1,1) 1a01 1a11 1a21 1a31
2a02 2a12 2a22 2a32
3a03 3a13 3a23 3a33
D.B. Rowe 81
MU MSCS Spring 2018
-1 0 1 2
System of Equations y Xa Image I ( x, y )
2 -1
4 equations from f xy ( x, y ) f ( x, y )
yx 0
3 3
f xy ( x, y ) aij ijx i 1 y j 1
j 0 i 0
1
x,y=0,1
f xy (0,0) a11 2
f xy (1,0) 1a11 2a21 3a31
f xy (0,1) 1a11 2a12 3a13
f xy (1,1) 1a11 2a21 3a31
2a12 4a22 6a32
3a13 6a23 9a33
D.B. Rowe 82
MU MSCS Spring 2018
-1 0 1 2
Need Image and Derivatives Image I ( x, y )
f ( x, y ) I ( x, y ) -1
f x ( x, y ) [ I ( x 1, y ) I ( x 1, y )] / 2 0
f y ( x, y ) [ I ( x, y 1) I ( x, y 1)] / 2 1
f xy ( x, y ) [ I ( x 1, y 1) I ( x 1, y ) I ( x, y 1) I ( x, y )] / 4 2
x,y=0,1
i.e. at (0,0)
Use the graph to reason out the derivatives. f x (0,0) [ I (1,0) I ( 1,0)] / 2
Only using surrounding points for derivatives. f y (0,0) [ I (0,1) I (0, 1)] / 2
f xy (0,0) [ I (1,1) I (1,0) I (0, 1) I (0,0)] / 4
D.B. Rowe 83
MU MSCS Spring 2018
y Xa 3 3
f ( x, y ) aij x i y j
f (0,0) 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a00
f (1,0)
1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 a10
i 0 j 0
f (0,1) 1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 a20
f (1,1) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a30
3 3
f x (0,0) 0
1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a01
f x ( x, y ) aij ix i 1 y j
f x (1,0) 0 1 2 3 0 0 0 0 0 0 0 0 0 0 0 0 a11
f (0,1) 0 j 0 i 1
1 0 0 0 1 0 0 0 1 0 0 0 1 0 0 a21
x 3 3
f x (1,1) 0 3 a31
f y ( x, y ) aij jx i y j 1
1 2 3 0 1 2 3 0 1 2 3 0 1 2
f (0,0)
y 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 a02
f y (1,0) 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 a12 j 1 i 0
f y (0,1) 0 0 0 0 1 0 0 0 2 0 0 0 3 0 0 0 a22 3 3
f y (1,1) 0
0 0 0 1 1 1 1 2 2 2 2 3 3 3 3 a32 f xy ( x, y ) aij ijx i 1 y j 1
f xy (0,0) 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 a03 j 0 i 0
f (1,0) 0 0 0 0 0 1 2 3 0 0 0 0 0 0 0 0 a13
xy
f xy (0,1) 0 0 0 0 0 1 0 0 0 2 0 0 0 3 0 0 a23
f (1,1) 0 9 a33
xy 0 0 0 0 1 2 3 0 2 4 6 0 3 6
x,y=0,1
D.B. Rowe 84
MU MSCS Spring 2018
y DI
f (0,0) 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 I (1, 1)
f (1,0) 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 I (0, 1)
f ( x, y ) I ( x, y )
f (0,1) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I (1, 1)
f (1,1) 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 I (2, 1)
f x (0,0) 0 0 0 0 2 0 2 0 I (1,0)
0 0 0 4 0 0 0 0
f x ( x, y ) [ I ( x 1, y )
f x (1,0) 0 0 0 0 0 2 0 2 0 0 0 0 0 0 0 0 I (0,0)
f (0,1)
x
0 0 0 0 0 0 0
0 2 0 2 0 0 0 0 0 I (1,0)
I ( x 1, y )] / 2
f x (1,1) 1 0 0 0 0 0 0 0 0 0 2 0 2 0 0 0 0 I (2,0)
f (0,0)
0 I (1,1) f y ( x, y ) [ I ( x, y 1)
y 4 0 2 0 0 0 0 0 0 0 2 0 0 0 0 0
f y (1,0) 0 0 2 0 0 0 0 0 I (0,1)
0 0 0 2 0 0 0 0
I ( x, y 1)] / 2
f y (0,1) 0 0 0 0 0 2 0 0 0 0 0 0 0 2 0 0 I (1,1)
f y (1,1)
0 0 0 0 0 0 2 0 0 0 0 0 0 0 2 0 I (2,1) f xy ( x, y ) [ I ( x 1, y 1)
f xy (0,0) 0 1 0 0 1 1 0 0 0 0 1 0 0 0 0 0 I (1, 2)
f (1,0) 0 0 1 0 0 1 1 0 0 0 0 1 0 0 0 0 I (0, 2) I ( x 1, y )
xy
f xy (0,1)
f (1,1)
0 0 0 0 0 1 0
0 0 0 0 0 0 1
0 1 1 0 0 0 0 1 0 I (1, 2) I ( x, y 1)
xy 0 0 1 1 0 0 0 0 1 I (2, 2)
x,y=0,1 I ( x, y )] / 4
D.B. Rowe 85
MU MSCS Spring 2018
recycle matrix
D.B. Rowe 86
MU MSCS Spring 2018
D.B. Rowe 87
MU MSCS Spring 2018
D.B. Rowe 88
MU MSCS Spring 2018
* bilinear edges
Low Resolution Bilinear Interpolated BiCubic Interpolated
D.B. Rowe 89
MU MSCS Spring 2018
8. MRI Example
Original 64×64
Bilinear
Cubic Spline
Cubic
Original 256×256
D.B. Rowe 90
MU MSCS Spring 2018
8. MRI Example
Original 256×256 Original 64×64
8. MRI Example
6301
Original 256×256 BiLinear 99
8. MRI Example
6301
Original 256×256 BiCubic 99
8. MRI Example
6301
Original 256×256 BiCubic Spline 99
8. MRI Example
-Original 256
●Original 64
-BiLinear
-BiCubic
-BiCubic Spline
“kink”
D.B. Rowe 95
MU MSCS Spring 2018
D.B. Rowe 96
MU MSCS Spring 2018
Thank You!
Questions?
1D 2D Example
D.B. Rowe 97