Abstract-This paper presents a new interpolation method for edge preserving when a low-resolution image is converted into a high-resolution image. The algorithm is based on both geometric closeness and photometric similarity, unlike the conventional interpolation algorithm which only uses geometric closeness. The proposed interpolation algorithm can preserve high frequencies in edge regions and retain similar conventional interpolation in smooth regions. The results of the proposed algorithm have higher peak signal-to-noise ratios (PSNR) and show a visual improvement with sharp edges compared to the conventional interpolation algorithm.
展开▼
