Dimension reduction using PCA, PCA whitening and ZCA whitening

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 ?

Answer 1

[Kenta]

ZCA whitening for image classification is available here. You can easily try this.

I hope it helps you.

https://jp.mathworks.com/matlabcentral/fileexchange/74990-zca-whitening-for-image-classification-zca