基于GF-1号与ZY-3号遥感影像同源异源融合研究

Remote Sensing Image Fusion of the Same Group and Different Groups Based on GF-1 and ZY-3

影像融合是遥感影像处理方向的研究热点之一,目前图像融合研究大多面向同源影像或异源自身进行,同时将二者结合起来的研究并不多见。以GF-1与ZY-3遥感影像作为融合数据,采用Brovey、HSV、Gram Schmidt、NNDiffuse Pan Sharpening、PC Spectral锐化等常见融合方法对同源、异源两组影像进行融合实验与对比,研究在同源影像融合中取得较好效果的算法是否也能在异源传感器的融合中行之有效。实验结果表明:同源影像融合中Gram Schmidt方法表现最好,PC Spectral锐化表现次之。异源影像融合中,PC Spectral锐化表现最好,Gram Schmidt方法表现次之。经典融合算法在同源和异源影像融合中均有优异表现。

Image fusion is a research hotspot in remote sensing image processing.Most of the present image fusion research deals with the same group of images or different groups,and the combination of the two is rare.In this paper,GF-1 and ZY-3 remote sensing images are used as fusion data,and fusion methods such as Brovey,HSV,Gram Schmidt,NNDiffuse Pan Sharpening and PC Spectral sharpening are used for fusion in the same group and different groups.Through contrast,it is possible to detect whether the algorithm that achieves good results in the same group of image fusion can be effective in the fusion of different sensors.The experiment results show that the Gram Schmidt method performs the best in image fusion of the same group,followed by the PC Spectral sharpening performance.PC Spectral sharpening performed best in different sets of image fusion,and the second is the Gram Schmidt method.The classical fusion algorithm performs well both in the same and different sets of image fusion.