一种基于Contourlet变换的遥感影像超分辨率重建改进方法  

An improved super-resolution reconstruction method of remote sensing image using Contourlet transform

作　　者：周靖鸿 周璀[1,2] 朱建军[1] 章浙涛 樊东昊 

机构地区：[1]中南大学地球科学与信息物理学院,湖南长沙410083 [2]湖南交通职业技术学院交通信息学院,湖南长沙410132

出　　处：《测绘工程》2016年第5期59-63,共5页Engineering of Surveying and Mapping

基　　金：国家973计划资助项目(2013CB733303);国家863资助项目(2012AA121301);国家自然科学基金资助项目(41274010;40974007)

摘　　要：针对原有方法轮廓波(Contourlet)分解高频信息处理的不足,提出用零填充重采样代替双三次插值用于高频信息处理的新方法,该方法可使Contourlet分解后的高频信息重采样过程不引入噪声能量的同时获得理想的插值结果。改进方法利用Contourlet多分辨率分析的特征,以及零填充重采样方法较传统插值方法优越的性能,此两者相结合可改进原有方法的不足,提高单帧影像的超分辨率重建效果。经理论分析与试验验证,改进后方法的超分辨率重建结果无论在定性分析与定量对比上都要优于传统方法。An improved method that uses zero-filled resampling in Fourier domain instead of the bi-cubic interpolation is proposed to solve the problem of high frequency information process using traditional Contourlet transform. This algorithm will not make the resampiing process of high frequency to introduce noise energy and also can get the expected interpolation results. The improved method takes the advantage of Contourlet transform multi-resolution analysis and zero-filled resampling, which can improve the traditional method＇ insufficient and also enhance the super-resolution reconstruction＇ effect of single frame image. Theoretical analysis and experimental verification approve the improved method both in qualitative analysis and quantitative comparison is better than traditional methods.

关 键 词：超分辨率 图像重建 轮廓波变换 傅里叶变换 零填充重采样 小波变换 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]