一种基于非采样Contourlet变换的遥感图像融合算法

为使融合后的多光谱图像尽可能保持原多光谱图像光谱特性的同时提高空间质量，提出了一种基于非采样Contourlet变换（NSCT）和多尺度边缘检测的融合算法。介绍了非采样Contourlet变换和多尺度边缘检测；设计了基于多尺度边缘检测、直接替代的高频、低频子带融合规则；用QuickBird卫星高分辨率遥感图像进行仿真实验。实验结果表明该算法能够在保持光谱信息的同时注入更丰富的空间细节信息，优于传统的Wavelet变换法和Contourlet变换法。

Remote Sensing Images Fusion Algorithm Based on the Nonsubsampled Contourlet Transform

In order to make the fused multispectral (MS) image keep the spectrum information of the original multispectral image as much as possible and improve the spatial resolution effectively, a novel image fusion algorithm based on the nonsubsampled contourlet transform (NSCT) and the multiresolution edge detection is proposed. The advantages of NSCT and multiresolution edge detection are presented; The fusion rules of the high and low frequency subbands respectively based on multiresolution edge detection and direct substitution are devised; The QuickBird images have been fused with this new method. The simulation experimental results demonstrate that the method proposed could keep the spectral information and improve spatial resolution simultaneously, and work better than traditional wavelet and contourlet transform.