基于总变分的超分辨率图像重建  

Super resolution image reconstruction based on total variation

作　　者：周卫峰[1] 李成军[1] 朱重光[2] 

机构地区：[1]中国科学院研究生院,北京100049 [2]中国科学院遥感应用研究所,北京100101

出　　处：《计算机工程与应用》2007年第36期91-93,共3页Computer Engineering and Applications

摘　　要：超分辨率图像重建是指从一组降晰的低分辨率图像重建出一帧清晰的高分辨率图像的过程。建立了超分辨率图像重建的数学模型,估计出场景在观测图像中的运动参数,选择总变分规整化克服问题的病态性得到重建结果。运用算法对模拟和实际图像序列进行重建,分别从主观效果和客观衡量标准两方面与基于Tikhonov规整化的超分辨率重建结果进行比较,结果表明该算法具有更好的处理效果。Super resolution image reconstruction is the process of reconstructing a high-resolution image from a set of degraded low-resolution images.The image formation model is introduced,then a motion estimation algorithm is proposed,and total variation is chosen to regularize the ill-posed problem.Simulated and practically taken image sequences are used in the experiments,and the results show that the proposed algorithm has better subjective vision effect and objective quality than that by Tikhonov regularization based reconstruction.

关 键 词：超分辨率 运动估计 总变分规整化 共轭梯度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]