Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (12): 153-154.DOI: 10.3778/j.issn.1002-8331.2010.12.045

• 图形、图像、模式识别 • Previous Articles     Next Articles

Image super-resolution reconstruction based on ICM algorithm

TANG Li-huan,CHEN Hui,LV Xiao-qian,KONG Fan-hui   

  1. School of Information Science and Technology,Shandong University,Jinan 250100,China
  • Received:2008-10-15 Revised:2008-12-25 Online:2010-04-21 Published:2010-04-21
  • Contact: TANG Li-huan

基于迭代条件模型算法的图像超分辨率重建

唐丽焕,陈 辉,吕小倩,孔凡慧   

  1. 山东大学 信息科学与工程学院,济南 250100
  • 通讯作者: 唐丽焕

Abstract: Super-resolution(SR) image reconstruction is an algorithm to reconstruct high quality and high-resolution(HR)images from a sequence of degraded low-resolution(LR) images.After studying maximum a posteriori probability,this paper proposes a procedure for super-resolution image reconstruction based on Markov Random Fields(MRF) and reconstructs high-resolution images through the Iterated Conditional Modes(ICM).The experimental results show that the algorithm can be extremely efficient in computing the Maximum A Posteriori probability(MAP).A Potts-Strauss model is assumed for the priori probability density function of image.After 5 or 6 iterations,a sharp image will be obtained.The algorithm has some practical value.

摘要: 图像超分辨重建是从一系列降质的低分辨率图像中获取高分辨率的图像。在最大后验概率算法基础上提出了一种基于马尔可夫随机场的超分辨率重建算法,并通过迭代条件模型实现超分辨率图像重建。实验结果表明,与传统的超分辨率重建算法相比,该算法是一种快速的计算最大后验概率的方法,采用Potts-Strauss模型作为图像的先验概率密度函数,经过五、六次的迭代就能达到理想的迭代效果,解决了最大后验概率算法计算量大的缺点,是一种高效的超分辨率重建算法,具有一定的实用价值。

CLC Number: