Image Zooming Using Non-Linear Partial Differential Equation

The main issue in any image zooming techniques is to preserve the structure of the zoomed image. the zoomed image may suffer from the discontinuities in the soft regions and edges; it may contain artifacts, such as image blurring and blocky, and staircase effects. this paper presents a novel image zooming technique using partial differential equations (pdes). it combines a non-linear fourth-order pde method with the locally adaptive zooming (laz) algorithm. the proposed method uses high-resolution image obtained from laz algorithm to construct zoomed image by fourth-order pde. this proposed method preserves edges and minimizes blurring and staircase effects in the zoomed image. in order to evaluate image quality obtained from the proposed method, this paper focuses on both subjective and objective assessments. the results of these measures on a variety of images show that the proposed method is superior over the other image zooming methods.