Regularization for Inverse Problems using the Singular Value Decomposition (SVD)

Question

I am reading these lecture notes on Optimisation and Inverse Problems in Imaging, and I have difficulties understanding how figures on page 20 (Figure 3.2) or page 21 (Figure 3.3).

Precisely, I don't understand what numbers on horizontal and vertical axes mean. I would appreciate if you could explain me this. Here is the code for Figure (3.3).

The question attains more attention if you include all the pertaining data in the question body. Please avoid references to external sites, if possible. The amount of effort you put in when writing your question is the amount of useful answers you end up with. — V.V.T, Sep 09 '20 at 15:02
Yes, but I don't understand how to show you figures. Anyway, thanks! — Novak Djokovic, Sep 09 '20 at 15:06
@NovakDjokovic there's a button with a label that looks like a picture in the question editor. — Marcus Müller, Sep 09 '20 at 16:06

score 5 · Accepted Answer · answered Sep 10 '20 at 06:23

Similar to The Concepts Behind SVD Based Image Processing the horizontal axis are the samples index of the SVD basis.

The idea in the chapter you linked is generalizing the Wiener Filter.
While the Wiener Filter uses the Fourier Transform as a basis the SVD uses the data adaptive basis.

Regularization for Inverse Problems using the Singular Value Decomposition (SVD)

1 Answers1