Mineral information recognition based on simulated high spatial resolution full spectrum remote sensing images

Minerals are natural compounds with certain chemical composition, which have stable phase interfaces and crystallization habits, and are of great significance for inversion of diagenetic and metallogenic geochemical characteristics and exploration. The use of remote sensing information to identify mineral types has achieved significant application results in the field of geology and mineral resources. In this paper, CASI-SASI-TASI airborne hyperspectral data and USGS standard spectrum library are used to establish a remote sensing image simulation method based on the combination of statistical model and Gaussian function, and the full spectrum remote sensing image with a spectral range of 425nm~12050nm and a spatial resolution of 2.25m is simulated. The simulated full spectral data were used to identify and extract 8 mineral information of limonite, hornblende, calcite/dolomite, high alumina sericite, medium alumina sericite, low alumina sericite, chlorite/epidote and quartz in Liuyuan area, Gansu Province, compared with the recognition results of airborne hyperspectral data, it was found that the two have strong consistency, this indicates that the simulated full spectrum remote sensing data in this article has strong practicality in identifying typical mineral information, and can provide important reference for the future development of spaceborne full spectrum high-resolution sensors and common key technologies.