Optimization of Artifact Free Image Upscaling Using Genetic Algorithm

Abstract

Creating an upscaled image from a low-resolution (LR) image to high resolution (HR) is very interesting. The nature and texture of an image should be maintained during upscaling of an image. This upscaling is also termed as “super-resolution”. It is related to both, statistical relationships between low-resolution and high-resolution. Several methods have been proposed to obtain better results, involving simple heuristics, edge modeling or statistical learning. The most powerful ones, however, present a high computational complexity, while fast methods, even if edge-adaptive, are not able to provide artifacts free images. In this paper, a simple and effective scheme for image upscaling is proposed. To balance visual quality and complexity, an optimal weight allocation using genetic algorithm is proposed which interpolates local pixel values along the direction where second order image derivative is lower. The scheme keeps the images as sharp as the original while avoiding magnifying noise and artifacts in the image. The high quality of the images enlarged with the new method is demonstrated with objective and subjective tests, while the computation time is reduced.

Keywords: Super-resolution, image upscaling, low resolution, high resolution, genetic algorithm

Cite this Article: Anishma MK, Shamna AR, Asmin MK, Asmin MK. Optimization of artifact free image upscaling using genetic algorithm. Journal of Web Engineering & Technology. 2020; 7(1): 21–29.