Thanks a lot, guys.
I believe the k-l transform is the PCA. But it seems that all the PCA kind of research is base on the vectors...
I've found how to do k-l transform on a block of an image,but it is not what I want...
Basically, they first partition the image into blocks such as 8x8 (it is a matrix), then they
rewrite these matrices as the vector (row by row for instance). Finally, they do the 1D K-L transform.
What I want is use the 8x8 matrix directly (not change it into a vector) and then do a 2D K-L transform (just like a 2D DCT used in the image coding and video coding standards). So can you guys give me a hint of this?
Post by chang
do you guys know how to do 2 dimensional K-L transform? I know how to do a 1-D one. But I really have no idea about how 2-D transform works. And I have google a lot and find nothing on 2-D case...
Thanks in advance!