联合运动估计与基于模式的超分辨率图像重构  

Joint Image Registration and Pattern-Based High-Resolution Image Reconstruction

作　　者：张地[1] 彭宏[1] 

机构地区：[1]华南理工大学计算机科学与工程学院

出　　处：《电子学报》2008年第1期180-183,共4页Acta Electronica Sinica

基　　金：国家自然科学基金(No.60772117);中国博士后科学基金(No.20060400218);广东省科技攻关项目(No.2005B10101033)

摘　　要：超分辨率图像重构是利用关于同一场景的多帧低分辨率图像重构出一幅具有更高分辨率图像的过程.已有的超分辨率图像重构算法对于人工模拟所得到的低分辨率图像序列具有很好的效果,但对于拍摄到的真实低分辨率图像序列而言,重构后的图像往往比较模糊,有时甚至仍然无法分辨.为此,本文提出了一个联合运动估计与基于模式的超分辨率图像重构算法.实验结果表明,该算法能够得到优于常规算法的高分辨率图像.Super-resolution image reconstruction is the process of producing a high-resolution image from a set of low-resolution images of the same scene. Previously published techniques usually perform well on simulated low-resolution image sequences, but the quality of the reconstructed images from real low-resolution videos, is always blurring or even unrecognizable. A joint image registration and pattern-based high-resolution image reconstruction algorithm was thus proposed in this paper. Experimental results demonstrated that the proposed algorithm could produce superior results against normal algorithms.

关 键 词：超分辨率 模式识别 图像序列 图像重构 

分 类 号：TN911.73[电子电信—通信与信息系统]