Decomposing Images into Harmonic Components Using Fourier Transform

[ramdas]

I am beginer in image processing. Any signal whether it is 1D,2D or any multidimensional signal can be represented using combination of number of sine and cosine wavesforms (harmonics).Similerly any image can be termed as a function of sinusoidal signals.I want to see individual pattern for the number of waveforms/harmonics present in an image .for example the following link contains few of them http://www.cs.toronto.edu/~rfm/factored/filters_out.png .So are there any Matlab/ OpenCV /C code or results to understand easily these components(sine and cosine functions(harmonics)) present in any image?

 

[FactChecker]

Matlab fftn says it does multidimensional Fourier transform. I don't have any experience with it.

 

[ramdas]

@FactChecker Sir fftn function gives Fourier transform of whole image.I want result as per link given in the question

 

[Fooality]

@FactChecker Sir fftn function gives Fourier transform of whole image.I want result as per link given in the question

Could you be a little more clear on exactly what you are looking for? Typically you take a 2D FFT of an image, and then do something with it and transform it back. For instance, if you only wanted to see certain frequencies in an image, you would take the 2D FFT, and multiply the frequencies you don't want by 0, and then transform it back.

The image you linked to is from this page:
http://www.cs.toronto.edu/~rfm/factored/
This page describes those as learned filters from a machine learning model that looks at different images. This professor's work is much deeper than any simple thing that can be done with FFT.

 

[ramdas]

@Fooality what i want is that I want to decompose an image into its bases function/components using Fourier transform/series. I want to see waveforms/harmonics when added together forms the original image.
Sir,Do u get me or not?
@FactChecker Sir fftn function gives Fourier transform of whole image.I want result as per link given in the question

Could you be a little more clear on exactly what you are looking for? Typically you take a 2D FFT of an image, and then do something with it and transform it back. For instance, if you only wanted to see certain frequencies in an image, you would take the 2D FFT, and multiply the frequencies you don't want by 0, and then transform it back.

The image you linked to is from this page:
http://www.cs.toronto.edu/~rfm/factored/
This page describes those as learned filters from a machine learning model that looks at different images. This professor's work is much deeper than any simple thing that can be done with FFT.