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?