In-orbit absolute radiometric calibration method for micro-nano hyperspectral satellite constellation

Abstract

The invention discloses an in-orbit absolute radiometric calibration method for a micro-nano hyperspectral satellite constellation. Comprising a radiometric calibration model design part scrupulous hyperspectral satellite multi-integral series imaging design, a hyperspectral data integral series normalization processing part, a ground actual measurement data apparent radiance calculation part scrupulous hyperspectral satellite spectral response function and a discrete point least square optimal fitting parameter solving part. The method has the following advantages that a satellite hyperspectral sensor in-orbit radiometric calibration model with different sites and different integral series being scrupulous is constructed, based on a global autonomous radiometric calibration network (RadCalNet), long-time-sequence ground automatic observation data of four calibration fields are utilized, and through time-space element normalization processing, the consistency of long-sequence observation data in time and space dimensions is realized, and the accuracy of the long-sequence observation data in time and space dimensions is improved. Absolute radiometric calibration parameter discrete point solving is carried out based on a radiation transfer model, and finally optimization of a calibration result is achieved through least square fitting.

Description

technical field [0001] The invention is a hyperspectral radiation calibration method, especially for multi-integral series dynamic imaging design of multi-CMOS stitching micro-nano sensor on-orbit absolute radiation calibration, and proposes a multi-integration method that meets the scruples of commercial micro-nano satellite constellations Imaging mode and multi-CMOS splicing design sensor global radiometric calibration network integrated calibration method. Background technique [0002] Absolute radiometric calibration is the premise of quantitative inversion of hyperspectral remote sensing data. The reliability and quantitative application level of data largely depend on the accuracy and stability of on-orbit radiometric calibration of hyperspectral sensors. [0003] For physical spectroscopic hyperspectral satellites designed with three-chip CMOS character-shaped splicing, such as figure 1 As shown, each CMOS corresponds to its own gradient spectral filter, and the resp...

AI Technical Summary

Problems solved by technology

At the same time, the physical spectroscopic sensor design of micro-nano hyperspectral satellite multi-CMOS splicing has differences in radiation consistency between chips, and independent calibration increases the difficulty and complexity of calibration. Micro-nano hyperspectral satellites have relatively insufficient in-orbit radiation stability. Achieving rapid periodic calibration of satellite constellations is a major challenge for commercial remote sensing services

Images