Wavelet-based hyperspectral and multispectral image fusion

被引：95

作者：

Gomez, RB ^{[1
]}

Jazaeri, A ^{[1
]}

Kafatos, M ^{[1
]}

机构：

[1] George Mason Univ, Ctr Earth Observing & Space Res, Sch Computat Sci, Fairfax, VA 22030 USA

来源：

GEO-SPATIAL IMAGE AND DATA EXPLOITATION II | 2001年 / 4383卷

关键词：

hyperspectral; multispectral; wavelet; image fusion; data fusion;

D O I：

10.1117/12.428249

中图分类号：

TP7 [遥感技术];

学科分类号：

081102 ; 0816 ; 081602 ; 083002 ; 1404 ;

摘要：

Different research groups have recently studied the concept of wavelet image fusion between panchromatic and multispectral images using different approaches. In this paper, a new approach using the wavelet based method for data fusion between hyperspectral and multispectral images is presented. Using this wavelet concept of hyperspectral and multispectral data fusion, we performed image fusion between two spectral levels of a hyperspectral. image and one band of multispectral image. The reconstructed image has a root mean square error (RMSE) of 2.8 per pixel and a signal-to-noise ratio (SNR) of 36 dB. We achieved our goal of creating a composite image that has the same spectral resolution as the hyperspectral image and the same spatial resolution as the multispectral image with minimum artifacts.

引用

页码：36 / 42

页数：7