Importance of Discrete Cosine Transform over High Pass filter?

Question

I have recently studied few work where the low and high frequency components of an image is accessed using the discrete cosine transform (one such example is presented here).

Low and High frequency components of an image can be easily accessed by applying a high pass filter in transformed domain.

My question is that what extra importance does DCT have over the simple high pass filter in accessing high and low frequency components of an image. In other words instead of applying DCT can't we directly access these components using High Pass filter?

Answer 1

Yes, if you apply a filter, say a high Pass, you get the low frequency components of the image, but still in the image domain. You need a Transformation, like the DCT or the FFT to get to the frequency domain, to see the separate frequency components of the image.