Design of Low-Artifact Interpolation Kernels by Means of Computer Algebra
This work addresses image quality improvement in interpolation for applications like image processing, but it is incremental as it builds on existing linear interpolator methods.
The authors tackled the problem of reducing anisotropic artifacts in image interpolation by designing new piecewise-polynomial kernels optimized using a symbolic computer algebra system, and experimental evaluation with 14 quality assessment methods showed favorable comparisons with existing linear interpolators.
We present a number of new piecewise-polynomial kernels for image interpolation. The kernels are constructed by optimizing a measure of interpolation quality based on the magnitude of anisotropic artifacts. The kernel design process is performed symbolically using Mathematica computer algebra system. Experimental evaluation involving 14 image quality assessment methods demonstrates that our results compare favorably with the existing linear interpolators.