استخراج ویژگی در تصاویر ابرطیفی به کمک برازش منحنی با توابع گویا

In this paper, with due respect to the original data and based on the extraction of new features by smaller dimensions, a new feature reduction technique is proposed for Hyper-Spectral data classification. For each pixel of a Hyper-Spectral image, a specific rational function approximation is developed to fit its own spectral response curve (SRC) and the coefficients of the numerator and denominator polynomials of this function are considered as new extracted features. The method focuses on geometrical nature of SRCs and relies on the fact that the sequence discipline - ordinance of reflectance coefficients in spectral response curve - contains some information which has not been addressed by many other existing methods based on the statistical analysis of data.  Maximum likelihood classification results demonstrate that our method provides better classification accuracies in comparison with many competing feature extraction algorithms. In addition, the proposed algorithm has the possibility  of being  applied to all pixels of image individually and simultaneously as well.