遥感影像融合及高保真算法比较分析研究

The Research on Comparing and Analyzing The Remote Sensing Image Fusion and Algorithm with High Spectral and Spatial Quality

随着传感器性能的提高,现代遥感技术提供了各种光谱分辨率、空间分辨率的遥感数据,如何充分、有效的利用多源、多尺度遥感影像数据成为遥感应用技术的一个挑战。本文系统阐述了遥感影像融合的理论和方法,详细分析了像素级、特征级、决策级遥感影像融合的方法和优缺点,深入探讨了Gram-Schmidt和SFIM两种高保真的融合方法,给出了融合结果的评价体系,结果表明SFIM融合方法是最优方法。

With the capability of the sensor promoting,modern remote sensing technology provides various spectral and spatial resolution remote sensing data.It becomes a challenge for remote sensing application to how to use multi-sources and multi-scales remote sensing image adequately and effective.This paper expatiates systemically the theory and method of the remote sensing image fusion,analyzes detailedly the method and character of the remote sensing image fusion of the pixel level,feature level,decision level.We discuss two remote sensing image fusion algorithm with high spectral and spatial quality —Gram-Schmidt and SFIM.We present the evaluating system of the fusion result.The research results show that the SFIM is the optimum fusion method.