Abstract

Spectral imaging typically generates a large amount of high-dimensional data that are acquired in different sub-bands for each spatial location of interest. The high dimensionality of spectral data imposes limitations on numerical analysis. As such, there is an emerging demand for robust data compression techniques with loss of less relevant information to manage real spectral data. In this paper, we describe a reduced-order data modeling technique based on local proper orthogonal decomposition (POD) in order to compute low-dimensional models by projecting high-dimensional clusters onto subspaces spanned by local reduced-order bases. We refer to the proposed method as the local-based approach because POD finds locally optimal solutions on each group split by k-means clustering. Experimental results are reported on three public domain databases and an in-house database. Comparisons with three leading spectral recovery techniques, three decomposition techniques used for hyperspectral imaging, and two baseline techniques show that the proposed method leads to promising improvement on spectral and colorimetric accuracy corresponding to the reconstructed spectral reflectance.

© 2015 Optical Society of America

Outlier modeling for spectral data reduction

Farnaz Agahian and Brian Funt
J. Opt. Soc. Am. A 31(7) 1445-1452 (2014)

Hyperspectral tomography based on proper orthogonal decomposition as motivated by imaging diagnostics of unsteady reactive flows

Weiwei Cai and Lin Ma
Appl. Opt. 49(4) 601-610 (2010)

Manifold regularized semi-supervised Gaussian mixture model

Haitao Gan, Nong Sang, and Rui Huang
J. Opt. Soc. Am. A 32(4) 566-575 (2015)

Edge-moment-based color constancy using illumination-coherent regularized regression

Meng Wu, Kai Luo, Jianjun Dang, and Jun Zhou
J. Opt. Soc. Am. A 32(9) 1707-1716 (2015)

Class-based spectral reconstruction based on unmixing of low-resolution spectral information

Yuri Murakami, Masahiro Yamaguchi, and Nagaaki Ohyama
J. Opt. Soc. Am. A 28(7) 1470-1481 (2011)