Atmospheric Correction of Satellite Imagery Using Atmospheric Profiles GPS Radio Occultation

A fuzzy predictive model is The atmospheric gases, aerosols and clouds scatter and absorb solar radiation and can modulate the reflected radiation from the earth. Therefore, the atmosphere contributes to the at-sensor radiance. So, one of the requirements for extracting reliable spectral information is to first compensate for atmospheric effects. Therefore, Atmospheric correction is a most important preprocessing step required in many remote sensing applications. There is a plethora of methods available, from simple to complex, which addresses correction and calibration issues. Some of the most accurate and severe methods are physically-based models, which require atmospheric data coincident with remote sensing data acquisition. Preparation of these data is difficult and expensive. In this study, a physical based algorithm proposed which provide partial information about atmosphere from GPS radio occultation profile. The method is provided to two global steps: in first step, atmospheric parameters are calculated using radiative transfer model. Then, in second step, calculated parameters are used for atmospheric correction. The results of the atmospheric correction process are evaluated by checking the level of spectral information using entropy criterion and comparing the computed at-surface reflectance with in-situ surface reflectance measurements. The results show the improvement of atmospheric corrected surface reflectance based on GPS radio occultation. Also, the information content criterion shows higher spectral content in images resulting from the proposed method.