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

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