How can a normal (raster) image:


Be converted in the Wolfram Language to a waveform, for example in greyscale, being the sum of all the harmonics in both dimensions that create numbers that when viewed as an image result in light and dark areas to match the original image. I don't know the correct terminology but I have seen high pass and low pass filters used on images in the Wolfram Language.

After which this complicated waveform can be adjusted subtly with processes that are more usual for audio signals, not just the *pass filters, for example a reverberation or chorus effects could be used.

Before converting the waveform back into an image.

This could result in new types of visual adjustments/operations for images.

How can this be done in the Wolfram Language?

  • $\begingroup$ Fourier transform ? $\endgroup$ – Sektor Apr 22 '15 at 10:35
  • $\begingroup$ Yes, that function doesn't seem to take image though. Maybe could look inside the HighPass[Image] function and see how to works. Image -> (to Waveform) -> Filter -> (to Image) $\endgroup$ – alan2here Apr 22 '15 at 11:14
  • 1
    $\begingroup$ You can use ImageData to solve the issue. Just go here and you will see Fourier in action. That's arguably the most popular transform there is and extensive work has been done. $\endgroup$ – Sektor Apr 22 '15 at 11:41
  • 1
    $\begingroup$ I think your question is somewhat ill-formed. Could you describe the operations you want to perform on the image data in more detail? For example, what is their mathematical description, and how should we understand a time-domain (one-dimensional) filter if applied to the (two-dimensional) spatial domain? What is a "greyscale waveform"? And so on. If you do not know the appropriate terminology, try to give examples. Because at the moment, the question is so vague that it is difficult to know what would constitute an acceptable answer. If you want an artistic effect only, please say so. $\endgroup$ – Oleksandr R. Apr 22 '15 at 14:33