Interference rejection-based Radial Basis Function Neural Network for hyperspectral image classification

A new application for Radial Basis Function Neural Networks (RBFNN) in nonlinear mixed pixel classification for hyperspectral imaging is presented. It is a three-layer neural network with the input layer specified by spectral signatures of a mixed pixel vector, the hidden layer used for nonlinear mixing functions and the output layer used to produce classification results of the mixed pixel vector. A noise estimation method in conjunction with noise subspace projection is developed to reliably estimate the number of mixing materials plus interference signatures that can be used as the number of hidden nodes as well as the number of input nodes. In order to implement the RBFNN the Least-Mean-Square (LMS) learning algorithm is applied to adjust parameters used in the hidden layer and weights of the output layers adaptively and simultaneously so as to achieve best possible performance. The performance is evaluated through a series of experiments via AVIRIS data. A comparative analysis is also conducted among various methods: a linear mixture model-based Orthogonal Subspace Projection (OSP) approach, Gaussian mixture-based Expectation-Maximization (EM) algorithm, Perceptron-based Linear Discriminant Analysis (PLDA) and Back-Propagation Neural Network (BPNN).