一种多光谱图像和全色图像融合算法

文章提出了一种新的基于独立成分分析的多光谱图像和全色图像的融合算法:对于两幅图像的重叠区域,使用独立成分分析去除遥感图像中高阶数据冗余问题,通过独立成分分析使得所分离的各个分量进行叠加从而得到最终的融合结果,从而提高了融合后图像的信息量和信噪比,同时使融合图像兼具全色图像的高分辨率和多光谱图像中特有的对目标某些独特特征的精细描述,使融合图像包含更丰富更有效的信息.

A Fusion Algorithm of Multi-Spectral Images and Panchromatic

In this paper, a new approach using the Independent Component Analysis (ICA) for remote image fusion between panchromatic and multi-spectral images is presented: First, we perform ICA to eliminate high-order image data redundancy for two overlapping images from different channels, then the respective independent components are fused. Image quality and signal-to-noise ratio are improved. At the same time, the fused image keeps the both property between panchromatic and multi-spectral images.