Computer Graphics

Two-Dimensional
Transformations
Edited by
Dr. Md. Manjur Ahmed
Faculty of Computer Systems and Software
Engineering
[email protected]
 Chapter Description

• Aims
– Basic of Computer Graphics.

• Expected Outcomes
– Understand the basic concept of computer graphics. (CO1: Knowledge)
– Ability to use the computer graphics technology. (CO1: Knowledge)

• References
– Dr. Masudul Ahsan, Dept. Of CSE, Khulna University of Engineering and Technology
(KUET), Bangladesh.
– Computer Graphics by Zhigang Xiang, Schaum’s Outlines.
– Donald Hearn & M. Pauline Baker, Computer Graphics with OpenGL, 4th Edition, Boston :
Addison Wesley, 2011.

2
 Modeling Transformations
• Simulate the manipulation of objects in space
• Two contrary points of view for describing
object
– Geometric transformation –
• Relative to a stationary coordinate system
• Changes in orientation, size and shape
– Coordinate Transformation – Keeping the object
stationary while coordinate system is transformed
with respect to the stationary object.
 Geometric transformation
– Basic Transformations
• Translation
• Rotation
• Scaling
• Shear
• Mirror reflection
 Translation – 2D Transformation
(about the Origin)
• Object is moved to a new position
 Considering distance and direction P’
• Let, new point P(x,y) is found by P
adding translation distance (tx,ty) to P(x,y).
Then displacement vector is
y
x  x  t x and y  y  t y - - - - (1)
Let’s column vector

 x  x  t x 
 
P    P   T  
 y y  t y 
Therefore, Eqn (1) can be rewritten as P  P  T
 Rotation – 2D
• Object is rotated along a circular path using
rotation angle ()
– Rotation angle ()
• Counter clockwise, + 
• Clockwise, - 
– Consider center (0,0) of rotation means that origin
as pivot point
 Rotation – 2D y
 x r cos  
P

P    
 y  r sin  

r sin( + )
r
P

r sin
 x  r cos(   )  r
P       
 
y    
 r sin( )  r cos( +  ) x
r cos
r cos  cos   r sin  sin  
  
 r cos  sin   r sin  cos  
 x cos   y sin  
  
 x sin   y cos  
 Rotation – 2D
Matrix form
 x   x  cos   y  sin   cos   sin    x 
 y   x  sin   y  cos     sin  cos     y 
       

Therefore,
P  R  P

R is called Rotation
Matrix
 Scaling – 2D
• Consider the dimensions of an object, either Expand or compress
• Scaling factors for XY Plane (2D) are defined as sx and sy.
magnification if sx (or sy) > 1
reduction if sx (or sy) < 1

x  sx  x , y  s y  y
In matrix form where S is scaling matrix.
P  S P
  
 x  sx  sx 0   x
ys    0   
sy   y
 y 
 Scaling – 2D
• After scaling,
– Centroid can be changed and new object will be located at a different
position relative to origin
scale factors, s  2, s 1
x y 2

2 2

1 1
1/2

Types of Scaling:
Differential ( sx != sy )
Uniform ( sx = sy )
 Homogeneous Co-ordinates
• Non-homogeneously: In matrix form, translation, scaling and
rotation are defined as:
– translation: P = P + T,  22 Matrix form
– Scale: P = S · P  33 Matrix form
– Rotate: P = R · P  33 Matrix form
• Composition Transformations (??)---- more than one
transformations at a time
– But translation not expressed as a matrix multiplication method, thus
difficult to determine
• Homogeneously: Allow all three transformation by the
multiplication of 33 matrices
• Each Cartesian position (x,y) is represented by a triple (xh, yh,
h), where,
xh yh
x y
h h
 Homogeneous Co-ordinates
• h can have any non zero value, better to use h = 1
• allows all transformation eq as matrix multiplication,
and
1)Transformation matrices for translation
x  x  t x , y  y  t y
 x   1 0 t x   x 
      
 y  0 1 t y  *  y 
 1  0 0 1   1 

T(t x ,t y )
 Homogeneous Co-ordinates
2) Transformation matrices for rotation
x  x  cos   y  sin  , y  x  sin   y  cos 
 x   cos  sin  0  x 
       
 y   sin cos 0 *  y 
 1   0 0 1  1 

R( )
3) Transformation matrices for Scaling
x  sx  x , y  s y  y
 x   s x 0 0  x 
      
y   0 sx 0 *  y 
 1   0 0 1  1 
S( sx ,s y )
 Mirror Reflection
Y Y

(1,1)
(-1,1) (1,1)

X X

(1,-1)

Reflection about X - axis Reflection about Y - axis

x  x y    y x   x y   y
1 0 0   1 0 0
M x  0  1 0 M y   0 1 0
0 0 1  0 0 1
Shearing Transformation
1 a 0 1 0 0 1 a 0
SH x  0 1 0 SH y  b 1 0 SH xy  b 1 0
0 0 1 0 0 1 0 0 1

Sheared in X Sheared in Y Sheared in both X

unit cube
direction direction and Y direction
 Composite Transformation
 Composite transformation matrix based on the concatenation
of transformation matrices
 It leads to different results if the change in the order of
transformation. Thus, the matrix, [A] ∙ [B] ≠ [B] ∙ [A]

 For example, in order to rotate an object around an arbitrary

point P(h,k):
1) Translate P(h,k) to the origin.
2) Rotate it around the origin.
3) To finish, translate the center of rotation back as it
was previously.
 General Pivot point, P(h,k),
Rotation of  : R,P
Step 1: Translate A(h,k) to origin
B3(x’+h, y’ +k)
Step 2: Rotate  w.r.t to origin
Step 3: Translate (0,0) to A(h,k)
A3(h,k)
R,P= T(h ,k) * R* T(-h ,-k)

B(x,y)

B1(x’,y’) B2(x’,y’)
A(h,k)

A1 (0,0) A2 (0,0)
Fixed point P(h,k) Scaling : Ssx,sy,p
Step 1: Translate P(h,k) to origin
Step 2: Scale S(sx,sy) w.r.t origin
(7,2)
Step 3: Translate (0,0) to P(h,k)
(1,1) (7,1)
Ssx,sy,P= T(h ,k) * S(sx,sy)* T(-h ,-k)
T(1,1)

(4,3)
(4,2)
(6,1)
(1,1) (4,1)
(0,0) (4,0) (0,0) (6,0)
S3/2,1/2,(1,1) T(-1,-1) S(3/2,1/2)
18