FAST ALGORITHMS FOR ESTIMATING AEROSOL OPTICAL DEPTH AND CORRECTING THEMATIC MAPPER IMAGERY

Remotely sensed images collected by satellites are usually contaminate d by the effects of atmospheric particles through the absorption and s cattering of radiation from the earth's surface. The objective of atmo spheric correction is to retrieve the surface reflectance from remotel y sensed imagery by removing the atmospheric effects, which is usually performed in two steps. First, the optical characteristics of the atm osphere are estimated and then the remotely sensed imagery is correcte d by inversion procedures that derive the surface reflectance. In this paper we introduce an efficient algorithm to estimate the optical cha racteristics of the Thematic Mapper imagery and to remove the atmosphe ric effects from it. Our algorithm introduces a set of techniques to s ignificantly improve the quality of the retrieved images. We pay parti cular attention to the computational efficiency of the algorithm, ther eby allowing us to correct large TM images quickly. We also provide a parallel implementation of our algorithm and show its portability and scalability on three parallel machines.