Research Article 
A New Bands Selection Algorithm for Hyperspectral Image using Hyperspectral
Derivative on Clifford Manifold
Information Technology Journal, 2012, 11(7), 904-909.
Abstract
In order to select best bands from the hyperspectral image with high computation efficiency, this paper proposed a new bands selection algorithm of hyperspectral image using the hyperspectral derivative on Clifford manifold. It firstly analyze the hyperspectral image in the Clifford algebra which had high efficiency on computation and analysis. Firstly, it discusses the properties of the Clifford algebra and Clifford manifold. Secondly, it gives the definitions of the hyperspectral derivative and the curve change rate on high dimensional space in Clifford algebra for finding the key points. Based on the theories mentioned above, a new bands selection algorithm is given with detail steps. At last, the results of the experiments are shown. Compared with the existing algorithms, the proposed algorithm is highly efficient in respect of computation and time. The study provided a new method for bands selection algorithm of hyperspectral image with the theories of Clifford algorithm.
ASCI-ID: 28-1645
Related Articles
The 7-DOF Automatic Fiber Placement Manipulator’s Self-motion Manifold
Information Technology Journal, 2014, 13(3), 560-565.
Cited By
Dimensionality reduction and derivative spectral feature optimization for hyperspectral target recognition
Optik - International Journal for Light and Electron Optics, 2017, 130(), 1349. DOI: 10.1016/j.ijleo.2016.11.143
A novel unsupervised bands selection algorithm for Hyperspectral imageOptik - International Journal for Light and Electron Optics, 2018, (), . DOI: 10.1016/j.ijleo.2018.01.001