4.
9 Matrix Transformations from Rn to Rm
A matrix transformation T: Rn Rm is a mapping of the form
⃗ ) = A𝒙
T (𝒙 ⃗ ,
⃗ vectors in Rn and an m x n matrix A.
for all 𝒙
The matrix transformations are precisely the linear transformations from Rn to Rm,
that is, the transformations with the linearity properties
𝑇(𝒖 + 𝒗) = 𝑇(𝒖) + 𝑇(𝒗) and 𝑇(𝑘𝒖) = 𝑘𝑇(𝒖)
We will use these two properties as the starting point for defining more general
linear transformations.
Remark:
It is important to note that a linear transformation is a special kind of function.
The input and output are both vectors.
⃗ ) by 𝒚
If we denote the output vector T (𝒙 ⃗ we can write
⃗ = A𝒙
𝒚 ⃗
Example 1:
1
Consider the letter L in figure, made up of the vectors (1, 0) or [ ] and (0, 2) or
0
0
[ ], show that the effect of the linear transformation
2
0 −1
⃗)=[
T (𝒙 ] ⃗𝒙
1 0
on this letter, describe the transformation.
Solution: As ⃗ ) = A𝒙
T (𝒙 ⃗
1 0 −1 1 0
T[ ]=[ ][ ]=[ ]
0 1 0 0 1
0 0 −1 0 −2
T[ ]= [ ][ ]=[ ]
2 1 0 2 0
4 Original
0 1 4
[ ] Translated
3
−1 0
3
0
2 [ ]
0 1
[] 2
2 1
1
0
0 1 2 3 4 0
-4 -3 -2 -1 0
1 −2
[] []
0
0
The effect of transformation on the L is rotated through an angle of 900 in the
anticlockwise direction.
Work to do:
2 0 1 0
Q1. Consider the matrices A= [ ], B=[ ],
0 2 0 0
−1 0 0 1
C=[ ], D=[ ],
0 1 −1 0
0 0.2 1 −1
E=[ ], F=[ ]
0 1 1 1
Show the effect of each of these matrices on L shape in example 1 and describe
each of the transformation in words.
Solution: As, ⃗ ) = A𝒙
T (𝒙 ⃗
1 2 0 1
T[ ]=[ ][ ]=[ ]
0 0 2 0 4
0 2 0 0 3
Original
T[ ]=[ ][ ]=[ ]
2 0 2 2 2
0
For T (⃗⃗⃗𝒙) = C𝒙
⃗ []
2 1
1 −1 0 1 0
T[ ]=[ ][ ]=[ ] 0 1 2 3 4
0 0 1 0
0 −1 0 0 1
T[ ]=[ ][ ]=[ ] []
0
2 0 1 2
.
.
.
Types of linear Transformations
There are two types of linear transformations (defining from R2 to R2):
1- Euclidean Transformation
2- Affine Transformation
1 – Euclidean Transformations
A Euclidean Transformation is a transformation T: 𝑅 2 → 𝑅 2 defined by
⃗ +𝒂
T(x) = A𝒙 ⃗ ∈ 𝑅2
⃗ , ∀𝒙
⃗ ∈ 𝑅 2 . These types of transformations
Where A is an orthogonal 2 x 2 matrix and 𝒂
always preserve distance/shape.
An orthogonal matrix holds the property AAT= 1 or AT= A-1
2 – Affine Transformations
An affine transformation is a transformation T: R2 to R2 define by
⃗ + ⃗𝒂 ,
T(x) =A𝒙 ∀ ⃗𝒙 ∈ 𝑅 2
⃗ ∈ 𝑅2 .
where A is a 2x2 invertible matrix and 𝒂
Remarks:
1. Every orthogonal matrix is invertible but an invertible matrix may or may not be
orthogonal.
1- Euclidean geometry is a subset of affine geometry or Affine transformations are
the generalization of Euclidean transformation.
Types of Euclidean transformation:
1- Translation,
2- Reflection,
3- Rotation.
Types of Affine transformation:
1- Scaling
2- Stretching
3- Shearing
1- Translation
Translation is a transformation from (R2 to R2) or (R3 to R3) defined as:
⃗)=𝒙
T (𝒙 ⃗ +𝒂
⃗ , ⃗ ∈ 𝑅2
∀𝒙
where matrix A is the identity matrix.
Example 2: (Translation of a triangle)
Let A = (-2, -2), B = (2, -2), C = (0, 2) form a triangle. Find the translated triangle
5
with vector 𝒂⃗ = [ ].
8
Solution: As the transformation of translation is
⃗)=𝒙
T (𝒙 ⃗ +𝒂
⃗ .
−2 5 3
For point A: D = T (A) =[ ]+[ ]=[ ]
−2 8 6
2 5 7
For point B: E = T (B) = [ ] + [ ] = [ ]
−2 8 6
0 5 5
For point C: F = T (C) = [ ] + [ ] = [ ]
2 8 10
10
F (5,10)
9
8 Translated
7 triangle
6
D(3,6) E (7,6)
5
4
3
2 C (0,2)
1
0
-5 -4 -3 -2 -1 -1 0 1 2 3 4 5 6 7 8 9 10
-2
A (-2,2) B (2, -2)
-3
-4
-5
Example -3: (Translation of a line)
⃗ = (2, 3).
For a line 3x - 4y = 2, find the equation of line translated through vector 𝒂
Solution: The transformation of translation is:
⃗)=𝒙
T (𝒙 ⃗ +𝒂
⃗
𝒙′ 𝒙 𝟐
So [ ]=[𝒚] + [ ]
𝒚′ 𝟑
𝒙′ 𝒙 + 𝟐
Or [ ]= [ ]
𝒚′ 𝒚 + 𝟑
Which implies 𝑥′= x + 2 and 𝑦′= y + 3.
Then, x = 𝑥′ - 2 and y = 𝑦′ - 3.
Put these in our given equation of line that is
3(𝑥′– 2) – 4(𝑦′– 3) = 2
3𝑥′ – 4𝑦′ = -4 is the required translated line.
To draw original line 3x – 4y = 2
put x = 0 implies y = -1/2, so A (0, -1/2) is a point on this line.
Similarly x =1 implies y = ¼ and B = (1, 1/4) is another point on it.
In the same manner to draw the Translated line 3𝑥′ - 4𝑦′= -4
Putting x = 0 gives 𝑦′= 1 and C = (0, 1).
Putting x =1 provides 𝑦′= 7/4 and D= (1, 7/4).
4
3 Translated line
2
Original line
1
0
-4 -3 -2 -1 0 1 2 3 4
-1
-2
-3
-4
Example -4: (Translation of circle)
Let (x – 4) 2 + (y - 3)2 = 9 be a circle. Find the equation of the translated circle
using vector (2, 3).
Note: As equation of circle: (x – a) 2 + (y – b) 2 = 𝑟 2 with Centre = (a, b) and
Radius = r. While x2 + y2 = r2 is circle with Center = (0, 0) and Radius = r .
Solution: The transformation of translation is
⃗)=𝒙
T (𝒙 ⃗ +𝒂
⃗
𝒙′ 𝒙 𝟐 𝒙 + 𝟐
[ ]=[𝒚] + [ ] = [ ]
𝒚′ 𝟑 𝒚 + 𝟑
𝑥′= x + 2 then x = 𝑥′ - 2 and 𝑦′ = y + 3 then y = 𝑦′ - 3
Putting these equations in the equation of circle
(x - 4)2 + (y - 3)2 = 9
(𝒙′ -2 - 4)2 + (𝒚′- 3 - 3)2 = 9
(𝒙′- 6)2 + (𝒚′– 6)2 = 9
Hence, Original circle is (x - 4)2 + (y - 3)2 = 9 with Center = (4, 3), Radius = 3.
While Translated circle is (𝒙′- 6)2 + (𝒚′- 6)2 = 9 with Center = (6, 6), Radius = 3
12
11
10
9
8
7
6 (6,6)
5 Translated
circle
4
3 (4,3)
2 Original
1
circle
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Note: As we said earlier that Euclidean transformations are distance/shape
preserving. So in all above examples we can see that translation transformation
being a Euclidean transformation preserves the shape of each object and just
translated or moved the object.
Work to do:
Q1. Let A = (3, 4), B = (3, 2), C = (6, 2) and D = (6, 4) form a rectangle. Find its
translation thorough vector (3, 5) and verify your translated rectangle from the
figure below.
Translated rectangle
10
5
A (3,4) D( 6,4)
4
3 Original rectangle
2
B (3,2) C (6,2)
1
0
0 1 2 3 4 5 6 7 8 9 10
