The Novel Bilateral Quadratic Interpolation Image Super-resolution Algorithm

Abstract

As a method of processing images, image interpolation has been widely applied to image processing. This paper proposed a new method of image super-resolution algorithm based on bilateral quadratic interpolation. We translate interpolation areas of the pixels to the specified area to construct the bilateral quadratic interpolation surfaces. The constructed surfaces are used to estimate the pixel values of the compensating pixel areas. By replacing each pixel with the corresponding areas, the image is amplified. The amplified images of the algorithm have more details remained than the results of the common algorithms. And this novel algorithm has a better improvement in the fidelity of the images. Moreover, it has a better performance in running speed and the quality of the images such as PSNR and SSIM. It can be used on the amplification of the color images, which can provide better quality amplified images for people. And it makes it convenient for people to study carefully on partial information of images.