In
the learning based single image super-resolution restoration, the high
frequency information is enhanced by retrieving the high-frequency information
from the high resolution training samples. Therefore how to reveal the
underlying relations between the HR and the LR patch spaces is the key issue. In this work, we propose to cluster the pre-collected HR
example patches to generate subdictionary and select the proper subdictionary
for any image patch according to the frequency spectrum feature in Fourier
domain, because the Fourier spectrogram can reflect the feature complexity,
local directionality and the texture periodicity of the image patch simultaneously.
