An Adaptive Neuro Fuzzy Interference System for Feature Extraction of Hyperspectral Image

Abstract

In this paper, a novel feature extraction method based on proposed for hyperspectral image classification. Hyperspectral images contain a large amount of data. Techniques are presented in this paper for visualizing important features contained in a hyperspectral data set. The major cause is that the size of training data set does not correspond to the increase of dimensionality of hyperspectral data. Actually, the problem of the “Finding minerals in hyper spectral images is too tough” emerges when a statistic-based classification method is applied to the hyperspectral data.  It was discovered that the resulting image is heavily influenced by the choice of focus bands used for display. When averaging hyper spectral signatures, choosing the correct pixels makes a difference, and desirable results are not always obtained. It was discovered that a procedure for visualizing hyper spectral image data that uses the peaks of the spectral signatures of pixels of interest provides a promising method for visualization.  Using wavelet coefficients and data from the hyperspectral bands produces noticeably different results, which suggests that wavelet analysis could provide a superior means for visualization in some instances when using bands does not provide acceptable results. The proposed Anfis (Adaptive neuro fuzzy interference system) method proves exceptional performance in terms of classification accuracy and computational efficiency.