Super-resolution sharpening-demosaicking method for removing image blurs caused by an optical low-pass filter
Takashi E. Komatsu, Takahiro Saito · 2005
We previously presented a demosaicking method that simultaneously removes blurs caused by an optical low-pass filter used in a digital camera. For the sharpening-demosaicking approach, the RGB color filter array is not necessarily proper, and we studied another color-filter array, namely the WRB filter array, where the W-filtering means that all visible light passes through it. Our prototypal method employed the iterative algorithm, and restored only spatial frequency components lower than the Nyquist frequency corresponding to the mosaicking pattern. First, this paper formulates the same recovery problem in the DFT domain more explicitly. Moreover, to suppress ringing artifacts often produced by our prototypal method near sharp color edges, we introduce the total-variation-based super-resolution into the sharpening-demosaicking approach. This super-resolution approach can demosaic and sharpen images while preserving image structures that intensity values are almost constant along the edge, without producing ringing artifacts.