中文摘要：

以 Landsat8 0 LI卫星遥感影像为数据源, 厦门岛为研究区, 从融合影像的光谱保真度和高 频信息融入度两方面, 评价了包括 SFIM、 MLT、 HPF、 MB （ modified brovey）、 HCS 和 GS （gram-schmidt）等 6 种遥感影像融合算法. 结果显示 除了 MB和 MLT方法外, 其他方法较好地保留了影像的光谱信息;除 MLT方法外, 其他方法均具有好的高频信息融入度, 能够显著提高影像的空间分辨率; 其中, G S法光 谱保真度最佳, HPF法高频信息融入度最好, HCS法则兼具良好的光谱保真度和高频信息融入度; 综合排 名显示 GS、 HPF和 HCS法融合效果不分伯仲.

英文摘要：

Six image fusion algorithms of SFIM, MLT , HPF , MB （ Modified Brovey）, HCS and GS （ Gram- Schmidt） were applied to Landsat 8 OLI images and their performances were evaluated based on the spectral fidelity and spatial frequency information gain , taking Xiamen Island as the study area. The results show four of the algorithms maintain the spectral information of images excluding MB and MLT , five have higher spatial frequency information gain than original image excluding MLT. GS has the best spectrum fidelity, HPF has the best high-frequency information gain and HCS has both outstanding spectrum fidelity and high-frequency information gain. The three produce similar fusion effects.