Dimension reduction using PCA, PCA whitening and ZCA whitening
16 views (last 30 days)
Show older comments
Hi all, I briefly understand the concept of PCA, PCA whitening and ZAC whitening which begins
1. Center the data
2. Calculate the covariance matrix
3. Solve the eigenvalue, eigenvectors of the covariance matrix
4. rotated data = eigenvector' * data
5. dimension reduced data = selected eigenvector' * data
6. PCA whitening = rotated data * eigenvalues^-0.5
7. ZCA = PCA whitening * eigenvector
So what I want to ask is that PCA whitening, ZCA whitening are calculated with all eigenvectors retained so does this mean PCA whitening and ZCA whitening do not reduce the dimension ?
0 Comments
Answers (1)
Kenta
on 14 Apr 2020
ZCA whitening for image classification is available here. You can easily try this.
I hope it helps you.
0 Comments
See Also
Categories
Find more on Dimensionality Reduction and Feature Extraction in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!