A HYPERSPECTRAL REMOTE SENSING FUSION TECHNOLOGY BASED ON SPECTRAL NORMALIZATION OF GF AND ZY SERIES SATELLITES

Abstract. Globalized surface coverage, environmental monitoring and other earth system science and high-quality global surface coverage monitoring applications urgently need basic hyperspectral remote sensing reflectance data to support. Spectral reflectance is the core data product of hyperspectral satellite remote sensing data, which is directly related to the application efficiency and quality of hyperspectral satellite. Due to the differences in spectral range, resolution width, central wavelength and other factors between multispectral and hyperspectral remote sensing images, the fusion processing of hyperspectral and multispectral remote sensing images is extremely difficult. In this study, the idea of normalization of physical parameters is adopted. Based on the remote sensing data of multi-source satellites such as GF5, GF1WFV and ZY3, the physical parameters of GF series satellites are fused to improve the spatial resolution of hyperspectral remote sensing images while the physical parameters of surface parameters are maintained. According to the fusion quality evaluation results of noise, information entropy, clarity, spectral angle cosine, correlation coefficient and root mean square error, the fusion results are consistent with the original numerical performance. The noise level, information entropy and clarity index of GF5 and ZY3 fusion are better. The spectral angle cosine, correlation coefficient and root mean square error index of GF5 and GF1 fusion are better. This study provides a new way and method for high-precision and quantitative processing of hyperspectral remote sensing data.