计算机应用 ›› 2011, Vol. 31 ›› Issue (05): 1209-1213.DOI: 10.3724/SP.J.1087.2011.01209

• 图形图像技术 • 上一篇    下一篇

基于最大后验估计的影像盲超分辨率重建方法

张洪艳1,沈焕锋2,张良培1,李平湘1,袁强强1   

  1. 1.武汉大学 测绘遥感信息工程国家重点实验室,武汉430079
    2.武汉大学 资源与环境科学学院,武汉430079
  • 收稿日期:2010-10-20 修回日期:2010-12-10 发布日期:2011-05-01 出版日期:2011-05-01
  • 通讯作者: 张洪艳
  • 作者简介:张洪艳(1983-),男,河南开封人,讲师,博士,主要研究方向:遥感影像处理、图像工程;沈焕锋(1981-),男,河北河间人,副教授,博士,主要研究方向:为影像质量改善;张良培(1962-),男,湖南安乡人,教授,博士,主要研究方向:遥感影像处理、模式识别;李平湘(1964-),男,湖南萍乡人,教授,博士,主要研究方向:SAR影像处理;袁强强(1986-),男,甘肃兰州人,博士研究生,主要研究方向:影像超分辨率重建。
  • 基金资助:

    国家自然科学基金资助项目(40930532;40971220;40801182)。

Blind super-resolution reconstruction method based on maximum a posterior estimation

ZHANG Hong-yan1, SHEN Huan-feng2, ZHANG Liang-pei1, LI Ping-xiang1, YUAN Qiang-qiang1   

  1. 1. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote sensing, Wuhan University, Wuhan Hubei 430079, China
    2. College of Resource and Environmental Science, Wuhan University, Wuhan Hubei 430079, China
  • Received:2010-10-20 Revised:2010-12-10 Online:2011-05-01 Published:2011-05-01
  • Contact: Zhang Hong-yan

摘要: 为了减小配准误差对盲超分辨率重建的影响,提出了一种影像配准和盲超分辨率重建联合处理的模型与方法。将配准参数、模糊函数和高分辨率影像建立在统一的最大后验估计模型框架内,并利用循环坐标下降最优化策略对模型进行求解,从而实现了配准参数、模糊函数和高分辨率影像的联合求解。实验结果证明:与传统盲超分辨率重建算法相比,该算法能够有效减少重建影像中的伪痕,在视觉评估上和定量评价上均能得到更好的结果。

关键词: 超分辨率, 配准误差, 最大后验概率, 影像配准, 模糊函数辨识

Abstract: In this paper, a new joint Maximum A Posterior (MAP) formulation was proposed to integrate image registration into blind image Super-Resolution (SR) reconstruction to reduce image registration errors. The formulation was built upon the MAP framework, which judiciously combined image registration, blur identification and SR. A cyclic coordinate descent optimization procedure was developed to solve the MAP formulation, in which the registration parameters, blurring function and High Resolution (HR) image were estimated in an alternative manner given to the two others, respectively. The experimental results indicate that the proposed algorithm has considerable effectiveness in terms of both quantitative measurement and visual evaluation.

Key words: Super-Resolution (SR), registration error, Maximum A Posterior (MAP), image registration, blurring function identification