一种新的NSCT超分辨率图像复原技术

考虑到非下采样Contourlet变换具有多尺度性和多方向性,提出了一种基于非下采样Contourlet变换的图像复原算法.首先对给出的序列低分辨率图像进行图像配准,对配准图像进行非下采样Contourlet变换,分解成低频及高频部分,对低频和高频分别采取不同的融合策略进行融合,再对融合后的高频和低频进行非下采样Contourlet变换逆变换得到融合图像.对融合后的图像再进行非下采样Contourlet变换,分别对高频细节和低频进行插值,再经非下采样Contourlet逆变换复原成高分辨率的图像.实验表明,该算法在主观效果和客观评价上均优于传统复原算法,是一种可行的图像复原新算法.

New image super-resolution restoration method by NSCT

Based on the multi-scaled and multi-directional properties of Nonsubsampled Contourlet Transform(NSCT),an image super-resolution restoration algorithm is proposed.The algorithm includes image registration and image restoration.First sub-pixel image registration is applied.Then a fused image can be obtained from the registered images by performing NSCT and the inverse NSCT.After the low frequency and high frequency of the fused image are interpolated,a high-resolution image can be obtained by performing the inverse NSCT.Experimental results show that the proposed algorithm outperforms the traditional algorithms in terms of both visual quality and objective evaluation criteria.