I am doing my final year M.Tech (Computer Science) project in Image Processing. I came across a method to remove non-linearity in the images by using surrogate images.
In order to generate surrogate image I am required to use Iterative Amplitude Adjusted Fourier Transform (IAAFT), the one described in the book NONLINEAR TIME SERIES ANALYSIS by Holger Kantz and Thomas Schreiber, 2004.
This is used to preserve probability density function and amplitude of the original image in the surrogate image generated. I know what PDF and amplitude mean but I wonder how it this method preserves them.
I have read it but unable to get a complete picture of what IAAFT is exactly doing or how does it exactly work. Can anyone help me to understand this concept?