高分一号卫星影像融合方法及质量评价  

Image Fusion and Quality Assessment of GF-1

作　　者：董倩[1] 岳彩荣[1] DONG Qian YUE Cairong(College of Forest, Southwest Forestry University, Kunming 650224, Chin)

机构地区：[1]西南林业大学林学院,云南昆明650224

出　　处：《林业调查规划》2016年第5期1-5,10,共6页Forest Inventory and Planning

基　　金：国家自然科学基金(31260156);云南省林学一流学科建设经费资助

摘　　要：影像融合是提高影像利用的有效方法之一。利用Brovey变换融合、PC主成分变换、GS融合和NND融合4种方法对高分一号的多光谱影像和全色影像进行融合试验。通过目视评价和定量评价方法对影像的融合效果进行评价。结果表明,Brovey变换融合方法在空间细节上表现好,PC主成份变换的光谱畸变小,但空间信息损失大,GS融合和NND融合在空间细节和光谱保真2方面表现都不错。Image fusion is one of the effective methods to improve the utilization of image information. This paper conducted a GF-1 fuse test of multi-spectral images and panchromatic images by using four different fusion algorithm： Brovey, PC, GS and NND. Based on visual interpretation and quantitative evaluation method, the results showed that Brovey worked well on spatial detail, which enhanced the spatial information; PC kept the spectrum best, but lost amount of spatial information; GS and NND had a good performance both on detail of spatial information and spectral fidelity.

关 键 词：高分一号卫星 影像融合 融合效果评价 光谱畸变 空间细节 光谱保真 定量评价 

分 类 号：S758.4[农业科学—森林经理学]