CS代考 Transformation Matrix – cscodehelp代写

Transformation Matrix
Let’s say our ’data’ is the
̈ ̨ ̈ ̨ 10
Ourbasevectorsare ̋ ‚(ingreen)and ̋ ‚. 01
⃝c -Trenn, King’s College London
2

Transformation Matrix
original transformed
We can stretch along the y-axis and squish along the x-axis with the matrix ̈ ̨
0.5 0 ̋‚
02
⃝c -Trenn, King’s College London 3

Transformation Matrix
original transformed
We can stretch along the y-axis and squish along the x-axis with the matrix ̈ ̨
0.5 0 ̋‚
02
̈ ̨ ̈ ̨ ̈ ̨
0.5 0 x 0.5x Toseethis,calculate ̋ ‚ ̈ ̋ ‚“ ̋
‚
⃝c -Trenn, King’s College London
4
0 2 y 2y

Transformation Matrix
original
We rotate counterclockwise
̈ ̨
0 ́1 ̋‚
10
transformed
⃝c -Trenn, King’s College London
5

Transformation Matrix
original
We rotate counterclockwise
transformed
̈ ̨
0 ́1 ̋‚
10
̈ ̨ ̈ ̨ ̈ ̨
0 ́1 x ́y Toseethis,calculate ̋ ‚ ̈ ̋ ‚“ ̋ ‚
⃝c -Trenn, King’s College London
6
10yx

Transformation Matrix
original transformed
We can also perform a so-called shear mapping
̈ ̨
1m
̋ ‚. Here m « 1.
01
⃝c -Trenn, King’s College London 7

Transformation Matrix
original transformed
We can also perform a so-called shear mapping
̈ ̨
1m
̋ ‚. Here m « 1.
01
̈ ̨ ̈ ̨ ̈ ̨
1m x x`my Toseethis,calculate ̋ ‚ ̈ ̋ ‚“ ̋ ‚
⃝c -Trenn, King’s College London
8
01yy

Transformation Matrix
original
transformed
Notice what happened to the vector ̋ ‚? Nothing. 0
⃝c -Trenn, King’s College London
9
̈ ̨
1

Transformation Matrix
original
transformed
̈ ̨
1
Notice what happened to the vector ̋ ‚? Nothing. 0
When a vector doesn’t change its direction after multiplying with a matrix, then it’s
an eigenvector.
⃝c -Trenn, King’s College London 10

Transformation Matrix
original transformed
In our stretching example from earlier, both vectors were actually eigenvectors. They didn’t change the direction.
However, they change the length.
The value by which is changed the length is called the eigenvalue λ.
⃝c -Trenn, King’s College London 11

Transformation Matrix
original transformed
The largest eigenvalue is λ1 “ 2 and the second largest here is λ2 “ 0.5. ̈ ̨ ̈ ̨
01
The corresponding eigenvectors are v1 “ ̋ ‚and v2 “ ̋ ‚. 10
⃝c -Trenn, King’s College London
12

Transformation Matrix
original
What is an eigenvector for the matrix
̈ ̨
1m ̋ ‚?
01
transformed
⃝c -Trenn, King’s College London
13

Transformation Matrix
original
What is an eigenvector for the matrix
transformed
̈ ̨
1m ̋ ‚?
01
̈ ̨
1
We can see that ̋ ‚must be one. What’s the formula?
0
⃝c -Trenn, King’s College London 14

Formula
A vector v is an eigenvector of the matrix M if M ̈ v “ λv.
λ is the corresponding eigenvalue.
⃝c -Trenn, King’s College London
15

Formula
A vector v is an eigenvector of the matrix M if M ̈ v “ λv.
λ is the corresponding eigenvalue. ̈ ̨
22 5 ́1
1 Wecanverifythatv1 “ ̋ ‚andλ1 “4:
1
̈ ̨ ̈ ̨ ̈ ̨ ̈ ̨
22141
M ̈v1 “ ̋ ‚ ̈ ̋ ‚“ ̋ ‚“4 ̋ ‚“λ1v1
5 ́1 1 4 1
̈ ̨
́2
Consider M “ ̋
‚. ̈ ̨
The second eigenvector is v2 “ ̋ ⃝c -Trenn, King’s College London
‚what’s λ2?
5
16