28.02.2013 Views

Principal Component Analysis (PCA)

Principal Component Analysis (PCA)

Principal Component Analysis (PCA)

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

• Let x 1 , x 2 , …, x p be the p sample images (n-dimensional).<br />

• Find the mean image, m, and subtract it from each<br />

image: z i = x i – m<br />

• Let A be the matrix whose columns are the<br />

mean-subtracted sample images.<br />

⎡ ↑ ↑ ↑ ⎤<br />

⎢<br />

⎥<br />

A = ⎢z1<br />

z 2 L z p ⎥ n ×<br />

⎢<br />

⎥<br />

⎣ ↓ ↓ ↓ ⎦<br />

• Estimate the covariance matrix:<br />

Σ<br />

=<br />

1<br />

Cov( A)<br />

= A A<br />

p −1<br />

What are the dimensions of Σ ?<br />

matrix<br />

Jochen Triesch, UC San Diego, http://cogsci.ucsd.edu/~triesch 22<br />

T<br />

p

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!