Retrieval of Atmospheric Parameters and Land Surface Rflectance from Airborn Hyperspectral Data

摘要

Surface reflectance spectrum is an important information in many applications. However, accurate retrieval of reflectance from remotely sensed data is difficult because of the existence of atmosphere. This paper described an improved physically-based algorithm to retrieve the aerosol optical thickness@550nm, columnar water vapor and the surface reflectance from airborne optical hyperspectral data. The improved method introduced were tested on synthetic and airborne collected optical hyperspectral data. The RMSE between retrieved and measured land surface reflectance were less than 0.0299 and 0.0158 for tested artificial targets and agriculture crops, respectively. The results indicate that this method is reliable for atmospheric and surface reflectance retrieval from airborne hyperspectral remote sensing data.