期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2008
卷号：XXXVII Part B7
页码：357-362
出版社：Copernicus Publications
摘要：The increasing radiometric accuracy and spectral resolution of the new aerospace optical imagers for Earth observation could allow a better characterization of the environment. This is true if accurate radiometric calibrations of the sensor are performed and atmospheric effects on the acquired data are carefully accounted for. To obtain spectral surface reflectance maps from the at-sensor radiance images, an improved atmospheric correction procedure have to be implemented. The availability of data acquired at high spectral resolution allows the detection of different spectral features of many atmospheric constituents. An iterative estimation algorithm based on high resolution data has been developed using the MODTRAN 4 radiative transfer code. The default atmospheric profiles available in that code have been firstly refined through at-ground level measurements of some parameters, like temperature, pressure, humidity, and solar irradiance. Then an iterative procedure has been started tuning H2O, CO2, CO, O3, and aerosol abundances. The MODTRAN 4 code is executed several times with different atmospheric parameters (H2O, CO2, CO, O3, and aerosol abundances) until the calculated ground irradiance matches the in-field measurements and the estimated ground spectral reflectance map is free from the related spectral signatures. To test and validate the method both simulated and acquired at-sensor radiance images have been utilized. The acquired images have been collected on 15th December 2005 during a coastal zone remote sensing campaign by the new airborne sensor HYPER / SIM-GA. The sensor has been operating in the 0.4 – 2.5 μm spectral range with 768 bands and a resolution of 2.4 in the Visible Near Infra-Red (VNIR) and 5.4 nm in the Short Wave Infra-Red (SWIR). First results are presented and discussed taking into account the feasibility of avoiding in-field measurements
