2D Coordinate Transformations Nick Battjes, Senior Student: Spatial Data Without Coordinates Control Points

2D Coordinate Transformations
Nick Battjes, Senior Student

B
l
u
e
l
a
k
e
G
r
e
e
n

P
a
r
k
D
a
r
k
R
i
v
e
r
R
e
d

H
i
l
l
s
W
G

S
t
a
t
i
o
n
Spatial data without coordinates Control Points
Blue
lake
Green Park
D
a
r
k
R
i
v
e
r
Red Hills
WG Station
Integrating of maps and spatial data in local coordinate system into a
world database system.
Note how the vector data (USGS Road layer) should be transformed
to match the raster data (Ikonos 1 meter res. Image)
Applications
Three different transformation primitives for the
Similarity transformation:

Translation- origin is moved, axes do not rotate i.e.:
Xn = X
0
DX
0
Y
n
= Y
0
DY
0

Scaling - both origin and axes are fixed, scale change
Xn = s
X
X
0
Y
n
= s
Y
Y
0

Rotation - origin fixed, axes move (rotate about origin)
Xn = X
0
cos(o) + Y
0
sin(o); Y
n
= - X
0
sin(o) + Y
0
cos(o)
2D Spatial Transformations
Two-Dimensional
Geographic Transformations
Xn
Yn
X0
Y0
Xn
Yn
X0
Y0
Xn
Yn
X
0
Y
0
Translation
Scaling Rotation
Moves and rotates objects in 2D and 3D space. Additionally,
you can scale the objects based on alignment points when
using the 2D option.

Conformal Transformation
Isogonal: having equal angles
Impose additional condition of equal scale (S = C
x
=
C
y
) yielding 4 parameters: S, o, DX
0
, DY
0
Moves and rotates and scale objects in 2D space.

Isogonal Affine Transformation
or Conformal/Similarity Transformation
cos sin
* * *
sin cos
Xn a b Xo c Xo DXo
S
Yn b a Yo d Yo DYo
o o
o o
( ( ( ( ( ( (
= + = +
( ( ( ( ( ( (

4 parameters: S scale, rotation, DX0, DY0 shifts in X and Y.
Xn,Yn are the transformed coordinates.
Xo, Yo are the original coordinates.
Two given points are required ( X1,Y1 and X2,Y2)
Moves and rotates and scale objects in 2D space.
Often called similarity Transformation since the basic shape
remain similar after the transformation

The formulas can be written in different forms
1. To compute the parameters given the coordinates
2. To compute the new coordinates given the parameters

( ) ( )
0 0 0
0 0 0
cos sin
sin cos
Xn S X S Y DX
Yn S X S Y DY
o o
o o
= + +
= + +
0 0
0 0
Xn a X b Y c
Yn b X a Y d
= + +
= + +
0 0
0 0
1 0
0 1
a
X Y Xn b
Y X Yn c
d
(
(
( (
(
=
( (
(

(

The formulas can be written in different forms
3. To compute the old coordinates given the parameters and new
coordinates (back substitution)

0 0
0 0
Xn a X b Y c
Yn b X a Y d
= + +
= + +
Conformal/Similarity Transformation
0 0
0 0
( )
( )
Xn c a X b Y
Yn d b X a Y
= +
= +
2
0 0
2
0 0
( )
( )
a Xn c a X ab Y
b Yn d b X ab Y
= +
=
Multiply Eq. 1 by a, Eq 2 by b
( ) ( )
( ) ( )
0
2 2
0
2 2
a Xn c b Yn d
X
a b
b Xn c a Yn d
Y
a b

=
+
+
=
+
Add equations to get X
0
The same operation to
obtain Y
0
Moves, rotates and scale objects in 2D space.

(
+
(
=
(
Cy
Cx
Yo
Xo
a b
b a
Yn
Xn
*
y
x
x
C a b
C b a
Cy a b
C b a
+ + =
+ + =
+ + =
+ + =
60 80 300
60 80 250
10 10 190
10 10 350
No X
o

(map)
Y
o

(map)
X
n

(ground)
Y
n

(ground)
1 10 10 350 190
2 80 60 250 300

(
(
(
(
(
(
(
(
=
(
(
(
(
y
x
C
C
b
a
*
1 0 80 60
0 1 60 80
1 0 10 10
0 1 10 10
300
250
190
350
Conformal Transformation: Example

Transform a given point X
0
=121.48, Y
0
=22.78

The point is transformed to Xn=310.59, Yn=373.6773

(
(
(
(
=
(
(
(
(
(
(
(
(
=
(
(
(
(

865 . 174
189 . 369
716 . 1
203 . 0
300
250
190
350
*
1 0 80 60
0 1 60 80
1 0 10 10
0 1 10 10
1
y
x
C
C
b
a
(
=
(
+
(

=
(
6773 . 373
59 . 310
865 . 174
189 . 369
78 . 22
48 . 121
*
203 . 0 716 . 1
716 . 1 203 . 0
Yn
Xn
Conformal Transformation:
Example
Used in photogrammetry for:
Transform comparator coordinates to photo coordinates and
used for correcting film distortion
Transform model coordinates to survey coordinates
Property
Carry parallel lines
into parallel lines
Does not have to
preserve orthogonality

y
1
y
2
x
1
x
2
|
c
t
2
t
1
Affine Transformation
Physical interpretation:

6 parameters: Cx, Cy, o, c, Dx
0
, Dy
0,
and in linear form:

( ) ( ) ( )
( ) ( ) ( ) ( )
0 0 0
0 0 0
cos( ) sin
sin cos
x y
x y
Xn C X C Y DX
Yn C X C Y DY
o o c
o o c
= + +
= + + +
( )
( ) ( )
0 0
cos sin
sin cos
x x
y y
a C d C
b C e C
c Dx f Dy
o o
o c o c
= =
= + = +
= =
Xn
Yn
X
0
Y
0
o
o+c
c
DX
DY
(
+
(
=
(
Cy
Cx
Yo
Xo
d c
b a
Yn
Xn
*
Affine Transformation
2-D Affine Transformation
The formulas for an affine transformation:

If n control points are measured, this Equation is reorganized as
follows:
s T A A A
s T A A A
x x a b c
y y d e f
( ( ( (
= +
( ( ( (

1 1 1
1 1 1
1 0 0 0
0 0 0 1
1 0 0 0
0 0 0 1
A
T s s
A
T s s
A
A
Tn sn sn
A
Tn sn sn
A
a
x x y
b
y x y
c
d
x x y
e
y x y
f
y
A
(
( (
(
( (
(
( (
(
( ( ~
(
( (
(
( (
(
( (
(

(

Projective or Polynomial Transformation
Instead of 4 or 6 parameters we have many parameters at
least 8 (8 would be the Bi-linear or projective
transformation)
With more parameters we need more known points to solve
the equations
N-equations and N unknowns.
...
...
2
0 5
2
0 4 0 0 3 0 2 0 1 0
2
0 5
2
0 4 0 0 3 0 2 0 1 0
+ + + + + + =
+ + + + + + =
Y b X b X Y b Y b X b b Yn
Y a X a X Y a Y a X a a Xn
General form

Alternatively,
+ + + + + + =
+ + + + + + =
xy b y b x b y bb x b b ' y
xy a y a x a y a x a a ' x
5
2
4
2
3 2 1 0
5
2
4
2
3 2 1 0
( ) ( )
( ) ( )
+ + + + =
+ + + + + =
xy 2 A y x A y A x A B ' y
xy 2 A y x A y A x A A ' x
3
2 2
4 1 2 0
4
2 2
3 2 1 0
Polynomial Transformation
Frequently used in photogrammetry
General form:
1 0 2 0 3
1 0 2 0
1 0 2 0 3
1 0 2 0
1
1
a X a Y a
Xn
d X d Y
b X b Y b
Yn
d X d Y
+ +
=
+ +
+ +
=
+ +
Projective Transformation
Impose condition of orthogonality (c = 0)
yielding 5 parameters: C
x
, C
y
, o, Ax, Ay

Orthogonal Affine Transformation
( ) ( )
( ) ( ) ( ) ( )
0 0 0
0 0 0
cos sin
sin cos
x y
x y
Xn C X C Y DX
Yn C X C Y DY
o o
o o
= + +
= + +
Condition: orthogonality and no scale change (C
x
=
C
y
= 1)

3 parameters: o, Ax, Ay

Rigid Body Transformation
0 0 0
0 0 0
cos sin
sin cos
Xn X Y DX
Yn X Y DY
o o
o o
= + +
= + +
Least Squares Adjustment Review
Model of adjustment of indirect observation ( Gauss-Markov model)

n: # of observations
m: # of parameters
f is a vector of given observation
V is the vector of residuals
B is the matrix of coefficients
W it the matrix of weights (Variance Covariance matrix)
o
0
is the reference variance
A is the vectors of parameters to be estimated

Number of observation is larger than number of parameters (redundant
observations). The solution that minimize the least-squares criterion
(v
T
wv) is:

) , 0 ( ~
2
0
W V V B f + A = o
n u rkB < =
f W B B W B
T T
= A
1
) (
Four fiducial marks (1 - 4) and two image points (a and

b) were measured on a comparator. The comparator
photo observations and the known values from the
camera calibration report are given in the following
spreadsheet.

Photo Coordinates Known Values
Point No. x y X Y
1 -111.734 -114.293 -113.007 -112.997
2 111.734 114.293 113.001 112.989
3 -114.289 111.699 -112.997 113.004
4 114.280 -111.749 112.985 -112.997
a 74.794 12.202
b -67.123 53.432
General 2D Conformal Transformation,
Example
4-Parameter Coordinate Transformation Program
______________________________________________________________________________________
Solution
Forming the B-matrix and f -matrix:
B
x
1
y
1
x
2
y
2
x
3
y
3
x
4
y
4
y
1
x
1
y
2
x
2
y
3
x
3
y
4
x
4
1
0
1
0
1
0
1
0
0
1
0
1
0
1
0
1
|
\
|
|
|
|
|
|
|
|
|
|
|
|
.
:= f
X
1
Y
1
X
2
Y
2
X
3
Y
3
X
4
Y
4
|
\
|
|
|
|
|
|
|
|
|
|
|
|
.
:=
General 2D Conformal Transformation,
Example
N B
T
B
( )
1
:=
The variance-covariance matrix is:
Q
XX
N :=
Q
XX
9.787E-006
0E+000
22.02E-009
122.332E-009
0E+000
9.787E-006
122.332E-009
22.02E-009
22.02E-009
122.332E-009
250E-003
0E+000
122.332E-009
22.02E-009
0E+000
250E-003
|
\
|
|
|
|
|
.
=
t B
T
f :=
t
102157.371
1161.611
0.018
0.001
|
\
|
|
|
|
|
.
=
General 2D Conformal Transformation, Example
The solution vector is:
A N t :=
A
0.99977
0.01137
0.00211
0.01222
|
\
|
|
|
|
|
.
=
The resisuals are
V B A f :=
V
0.002
0.013
0.004
0.019
0.002
0.020
0.004
0.013
|
\
|
|
|
|
|
|
|
|
|
|
.
=
The reference variance f or the adjustment is
o
V
T
V
4
:= o 0.0003 ( ) =
The Transf ormed coordinates become:
X
a
A
1
x
a
A
2
y
a
+ A
3
+ := X
a
74.913 =
Y
a
A
2
x
a
A
1
y
a
+ A
4
+ := Y
a
11.361 =
X
b
A
1
x
b
A
2
y
b
+ A
3
+ := X
b
66.502 =
Y
b
A
2
x
b
A
1
y
b
+ A
4
+ := Y
b
54.195 =
Normally shown as

Unique solution if
0 0
0 0
Xn a X b Y c
Yn d X e Y f
= + +
= + +
0
a b
d e
=
General 2D Affine Transformation
0 0
0 0
1 0 0 0
0 0 0 1
a
b
X Y Xn c
X Y Yn d
e
f
(
(
(
(
( (
=
(
( (

(
(
(

Four fiducial marks (1 - 4) and two image points (a and
b) were measured on a comparator. The comparator
photo observations and the known values from the
camera calibration report are given in the following
spreadsheet.

Photo Coordinates Known Values
Point No. x y X Y
1 -111.734 -114.293 -113.007 -112.997
2 111.734 114.293 113.001 112.989
3 -114.289 111.699 -112.997 113.004
4 114.280 -111.749 112.985 -112.997
a 74.794 12.202
b -67.123 53.432
General 2D Affine Transformation,
Example
_____________________________________________________________________________________
y
b
53.432 := x
b
67.123 :=
y
a
12.202 := x
a
74.794 :=
The measured points are:
Y
4
112.997 := X
4
112.985 := y
4
111.749 := x
4
114.280 :=
Y
3
113.004 := X
3
112.997 := y
3
111.699 := x
3
114.289 :=
Y
2
112.989 := X
2
113.001 := y
2
114.293 := x
2
111.734 :=
Y
1
112.997 := X
1
113.007 := y
1
114.293 := x
1
111.734 :=
Input Values: Note that lower case values represent observed comparator coordinates while the upper
case represents the known camera calibration coordinates f or the respective f iducial values
______________________________________________________________________________________
6-Parameter Coordinate Transformation Program
General Affine Transformation, Example
B
x
1
0
x
2
0
x
3
0
x
4
0
y
1
0
y
2
0
y
3
0
y
4
0
1
0
1
0
1
0
1
0
0
x
1
0
x
2
0
x
3
0
x
4
0
y
1
0
y
2
0
y
3
0
y
4
0
1
0
1
0
1
0
1
|
\
|
|
|
|
|
|
|
|
|
|
|
|
.
:= f
X
1
Y
1
X
2
Y
2
X
3
Y
3
X
4
Y
4
|
\
|
|
|
|
|
|
|
|
|
|
|
|
.
:=
N B
T
B
( )
1
:=
Solution: forming the B matrix and f vector

The variance-covariance matrix is:
Q
XX
N :=
Q
XX
19.573E-006
1.603E-009
44.019E-009
0E+000
0E+000
0E+000
1.603E-009
19.573E-006
244.661E-009
0E+000
0E+000
0E+000
44.019E-009
244.661E-009
250E-003
0E+000
0E+000
0E+000
0E+000
0E+000
0E+000
19.573E-006
1.603E-009
44.019E-009
0E+000
0E+000
0E+000
1.603E-009
19.573E-006
244.661E-009
0E+000
0E+000
0E+000
44.019E-009
244.661E-009
250E-003
|
\
|
|
|
|
|
|
|
.
=
t B
T
f :=
t
51079.018
583.52
0.018
578.092
51078.353
0.001
|
\
|
|
|
|
|
|
|
.
=
The solution vector is:
A N t :=
A
0.99977
0.01134
0.00211
0.01140
0.99977
0.01222
|
\
|
|
|
|
|
|
|
.
=
The resisuals are
V B A f :=
V
0.001
0.016
0.001
0.016
0.001
0.016
0.001
0.016
|
\
|
|
|
|
|
|
|
|
|
|
.
=
The reference variance f or the adjustment is
o
V
T
V
2
:= o 0.001 ( ) =
The Transf ormed coordinates become:
X
a
A
1
x
a
A
2
y
a
+ A
3
+ := X
a
74.913 =
Y
a
A
4
x
a
A
5
y
a
+ A
6
+ := Y
a
11.359 =
X
b
A
1
x
b
A
2
y
b
+ A
3
+ := X
b
66.504 =
Y
b
A
4
x
b
A
5
y
b
+ A
6
+ := Y
b
54.197 =

2D Coordinate Transformations Nick Battjes, Senior Student: Spatial Data Without Coordinates Control Points

Uploaded by

Copyright:

Available Formats

2D Coordinate Transformations Nick Battjes, Senior Student: Spatial Data Without Coordinates Control Points

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2D Coordinate Transformations Nick Battjes, Senior Student: Spatial Data Without Coordinates Control Points

Uploaded by

Copyright:

Available Formats

2D Coordinate Transformations

Nick Battjes, Senior Student

Four fiducial marks (1 - 4) and two image points (a and

You might also like