Universal sharpening-demosaicing for various types of color-filter array
Takashi Komatsu, Takahiro Saito · European Signal Processing Conference · 2010
A monochrome-image iterative deblurring method with the classic soft-shrinkage in the shift-invariant Haar wavelet transform domain was recently proposed by R. H. Chan et al. Extending this deblurring method, we present a new iterative sharpening-demosaicing method with the shift-invariant Haar wavelet transform and our color shrinkage utilizing redundant color transform. Our new sharpening-demosaicing method is originally constructed for the Bayer's primary color-filter array (CFA), but its minor modification renders it applicable to various CFA's other than the Bayer's CFA: the complementary CFA, the random arrangement CFA, and so on. Simulation results demonstrate that our new sharpening-demosaicing method in the shift-invariant Haar wavelet transform domain works much more efficiently than our previously proposed sharpening-demosaicing method with the totalvariation regularization in the spatial image-domain.