Estimation of adjacency effect in the remotely sensed Hyperspectral data over rugged scenes

Abstract

Adjacency Effect heavily affects the remotely sensed Hyperspectral data by adding surface surrounding scattering signals at sensor. Estimation of adjacency effect remains difficult because of the complex phenomena induced by rugged terrain. Indeed, existing method is not valid for a rough, heterogeneous surface like rugged terrain. In this paper we describe a method to estimate the adjacency effect in Hyperspectral data of rugged terrain. We compute the scattering and reflecting of light from adjacent surfaces into the field-of-view (FOV) based on the theory of radiation transfer. Our computation takes molecular/aerosol scattering phase functions and topographic features into account. We also discuss the dependence of the adjacency effect on the aerosol optical thickness and the relief of terrain.