Image interpolation based on the pixel value corresponding to the smallest absolute difference

This paper proposes a novel algorithm for image interpolation. The motivation is based on relentless image interpolation artefacts and computational complexity. The solution proposed is founded on reprocessing one of the four pixels surrounding the unknown location and calculating the mean between that pixel and the value introduced by the bilinear interpolation. Subsequently, the mean is multiplied by the control factor k whose value is selected according to experimental analysis. Experimental results are further presented to show the effectiveness and performance of the proposed algorithm.