摘要
研究了混合噪声即高斯白噪声加椒盐噪声所降质的超分辨率图像重建模型,提出了基于全变分正则化方法和两步方法的超分辨率图像重建方法,并应用Chambolle投影算法对模型进行求解;通过实验仿真,实验结果表明本文提出的方法比基于传统的全变分正则化方法无论是在运行速度,MSE与SNR评价指标上还是在视觉效果方面都具有明显的优越性。
Super-resolution image reconstruction models degraded by mixed Gaussian plus salt pepper noises are studied in the paper.The method is proposed combing total variation regularization method and two-phase method.The proposed method is used to reconstruct super-resolution image,which was solved by employing Chambolle＇s projection algorithm.Compared with the traditional super-resolution methods based on least mean square and total variation by using simulation experiments,the experimental results confirm the effectiveness of the proposed method and demonstrate its superiority from fidelity,processing time,MSE and SNR.
出处
《红外技术》
CSCD
北大核心
2010年第10期604-607,共4页
Infrared Technology
基金
华南农业校长基金
编号:2009K203
2008K011
广东省自然科学基金
编号:9251064201000009
国家青年基金项目
编号:60702030
国家自然科学基金项目
编号:50775079
关键词
超分辨率
去混合噪声
全变分
正则化
两步方法
图像重建
super-resolution
mixed noises removal
total variation
regularization
two-phase method
image reconstruction
