Modified Laplacian Filter and Intensity Correction Technique for Image Resolution Enhancement

Abstract

By analyzing the deterministic relationship between the lower-resolution and the corresponding higher resolution images, we propose two core techniques namely MLF (modified Laplacian filter) and IC (intensity correction) for image resolution enhancement, by which the image size can be increased revealing better details of the image contents. The simple 3times3 MLF is designed for properly restoring the frequency components attenuated in the averaging and down-sampling degradation process. The IC process iteratively refines the image quality for any resolution enhanced (enlarged) image. Experiments show that the proposed techniques can effectively improve the image qualities than bilinear or bicubic interpolation alone. It outperforms other recently developed algorithms both in perceptual quality (especially in the texture areas) and in objective quality in terms of PSNR. Both MLF and IC are simple in computations, which is quite desirable in many time sensitive applications