Abstract
This paper proved the cubic polynomial kernel design has capability to replace the famous cubic spline interpolation kernel. The main contribution this work to proven the cubic polynomial has good performance with less aliasing and blurring rather than applied cubic spline interpolation.
The cubic spline kernel provided good response better than nearest neighbor and bilinear function and it was widely use in digital images signal processing application especially in super resolution techniques. Nevertheless the finding better method is never end. It has two ways to validate
the performance of both techniques. An analysis is done with comparison both techniques kernel with use frequency response and measuring the highest PSNR as benchmark of this study. This study proved that the cubic polynomial produced more smooth function in frequency response and highest
gain PSNR images. The propose techniques has less computational complexity and good response thus it could gain efficiency in digital image application.