影像局部直方图匹配滤波技术用于遥感影像数据融合

本文提出了一种基于局部直方图匹配波技术的影像融合新方法。该方法分析了局部影像统计特性，应用均值或均值－方差匹配正态函数，对要融合的两幅影像局部直方图进行匹配。文中给出了３种不同的融合算法（高通滤波法，局部均值匹配法和局部均值与方差匹配法）用于ＳＰＯＴ全色影像和ＴＭ多光谱影像融合的结果，并进行了比较，质量评价结果证明了提出的方法在提高了原ＴＭ多光谱影像空间分辨率的同时，较好地保留了多光谱影像频道的光

A New Fusion Approach Based on Local Histogram Matching Filtering Techniques

This paper presents a new image fusion scheme based on local histogram matching filtering techniques for fusing high resolution panchromatic and multi spectral images. The proposed method analyses local image statistics and then matches the local histograms of two images to be fused by applying mean or mean variance matching normalization functions. Three different fusion algorithms (HPF, LMM,LMVM) were applied to merge SPOT panchromatic and TM multi spectral images. The quality evaluation results show that the amount of spectral information can be efficiently preserved while the spatial resolution has been improved.