Preprocessing and fusion analysis of GF-2 satellite Remote-sensed spatial data

Abstract

There is no pansharpening method that can be applied to all kinds of images at present due to the principle of fusion processing and the characteristics of sensors that acquire images. In order to explore the suitable fusion method for the " Gaofen-2 " satellite image, PCA, HPF, Gram-Schmidt and NNDiffuse four kinds of fusion methods were selected to merge the panchromatic and multi spectral data of Gaofen-2 satellite images, the fusion results of the image were synthetically compared and evaluated with subjective evaluation and quantitative analysis. The test results show that the NNDiffuse transform method has the best combination effect and is very prominent in the fusion effect of the visible light band; And in the fusion of near-infrared band, Gram-Schmidt method can be considered. The research results of this paper can provide reference for the fusion processing and application of Gaofen-2 satellite image data.