人眼视觉感知模型指导的有理函数图像插值

Rational function for image interpolation based on human contrast sensitivity

基于有理函数模型提出了一种新的自适应图像插值算法.此类有理函数具有简单显性数学表达式,且含有可调参数.当两个参数都等于1时,有理函数变为双三次插值函数.基于有理函数构造图像插值曲面,原始图像通过等值线分析自适应地分解为平滑和边缘两部分.其中自适应阈值根据有理函数构造原理来确定,思路是将人眼视觉感知与图像结构相融合,对于像素结构简单视觉关注度低的平滑区域,采用双三次插值.对于像素结构复杂视觉关注度高的边缘区域,采用有理插值处理,参数由人眼对比敏感特性来确定.实验表明,该算法细节信息保持优于当前经典的图像插值算法,具有较好的视觉效果.

A new image interpolation algorithm based on the rational function model with both smoothing and surface constraints is proposed. The rational function has a simple and explicit expression with two parameters. The image edge regions can be detected adaptively by using contour analysis. Basically, the detective threshold is selected based on the rational function construction. Selecting different parameters in the rational function model, the edge regions and smooth regions are interpolated by bicubic interpolation and rational interpolation, respectively. The optimal rational interpolation parameters can be obtained with a set of exact solutions by solving a maximization problem based on the contrast sensitivity enhancement of human eyes. Experimental results demonstrate that the proposed method achieves competitive performance with the state-of-the-art interpolation algorithms, especially in image details features.