Color image interpolation with edge-enhancement

We propose the use of an edge-retaining asymptotic projections onto convex sets (ERAPOCS) algorithm for the interpolation of color images. Our algorithm differs from POCS in that the projections sequence in one (or more) of the convex sets can enter a desired region rapidly. Like POCS, our algorithm is guaranteed to converge. In addition, our algorithm is more computationally efficient than the POCS algorithm since it converges more quickly and has computational complexity for each iteration. The algorithm alleviates edge blurring by properly amplifying the wavelet transform (WT) of the image. The amplification in the WT domain biases the projection sequence to the subset of interpolated images that has less edge-blurring. In this way, we can obtain an interpolated image that retains a significant amount of edge information. We use our algorithm to interpolate a color image in the red, green and blue planes.

[1]  D. Youla,et al.  Image Restoration by the Method of Convex Projections: Part 1ߞTheory , 1982, IEEE Transactions on Medical Imaging.

[2]  Martin Vetterli,et al.  Discrete-time wavelet extrema representation: design and consistent reconstruction , 1995, IEEE Trans. Signal Process..

[3]  Martin Vetterli,et al.  Resolution enhancement of images using wavelet transform extrema extrapolation , 1995, 1995 International Conference on Acoustics, Speech, and Signal Processing.

[4]  Victor E. DeBrunner,et al.  Edge-retaining asymptotic projections onto convex sets for image interpolation , 2000, 4th IEEE Southwest Symposium on Image Analysis and Interpretation.

[5]  M. Sezan,et al.  Image Restoration by the Method of Convex Projections: Part 2-Applications and Numerical Results , 1982, IEEE Transactions on Medical Imaging.

[6]  Stéphane Mallat,et al.  Characterization of Signals from Multiscale Edges , 2011, IEEE Trans. Pattern Anal. Mach. Intell..