A New Sparse Dimension Reduction Method for Hyperspectral Images

JING-ZUN ZHANG, LI-PING LIAO, GUANG-MEI XU, RUI-ZHE ZHANG, NING HE

Abstract


Motivated by the development of sparse representation, many sparsity-based methods have been successfully applied in Hyperspectral images (HSI) domain. These methods are beneficial to find the intrinsic structure from the high dimension data. Sparse dimension reduction is becoming one of the research hotspots, more efforts are needed to be further. In this paper, we present a new unsupervised dimension reduction (DR) method, which is called unsupervised double sparse learning method (UDSDL). UDSDL inherits the merits of sparse representation, it can support the possibility of data compressing while preserving more discriminative features. Experiments on a real HSI data set collected by the Airborne visible/Infrared Imaging Spectrometer are performed to demonstrate the effectiveness of the proposed UDSDL method

Keywords


Hyperspectral image, Dimension reduction, Double sparse& unsupervised.Text


DOI
10.12783/dteees/peems2019/34003

Full Text:

PDF

Refbacks

  • There are currently no refbacks.