Illumination-Invariant Recognition of 3D Hyperspectral TexturesUsing Spectral/Spatial Gabor Filters

摘要

We develop a method for the recognition of textures with three-dimensional structure in hyperspectral images. Propertiesof a texture are captured by a feature vector that is generated by a bank of spectral/spatial Gabor filters. Variation in theillumination and atmospheric conditions is modeled using a subspace of the feature vectors. Since a large bank of filtersis used, we develop methods for reducing the dimension of the feature vector that is used to represent a texture. The goalof the dimension-reduction process is to optimize the discriminability of a set of textures. We demonstrate the utility ofthe approach using experiments with hyperspectral textures of three-dimensional objects that are generated by DIRSIGover a range of conditions.