A RGB-Guided Low-Rank Method for Compressive Hyperspectral Image Reconstruction

Limin CHEN; Jing XU; Peter Xiaoping LIU; Hui YU

doi:10.1587/transfun.E101.A.481

A RGB-Guided Low-Rank Method for Compressive Hyperspectral Image Reconstruction

Limin CHEN, Jing XU, Peter Xiaoping LIU, Hui YU

Full Text Views

0

Share
Cite this

Summary :

Compressive spectral imaging (CSI) systems capture the 3D spatiospectral data by measuring the 2D compressed focal plane array (FPA) coded projection with the help of reconstruction algorithms exploiting the sparsity of signals. However, the contradiction between the multi-dimension of the scenes and the limited dimension of the sensors has limited improvement of recovery performance. In order to solve the problem, a novel CSI system based on a coded aperture snapshot spectral imager, RGB-CASSI, is proposed, which has two branches, one for CASSI, another for RGB images. In addition, considering that conventional reconstruction algorithms lead to oversmoothing, a RGB-guided low-rank (RGBLR) method for compressive hyperspectral image reconstruction based on compressed sensing and coded aperture spectral imaging system is presented, in which the available additional RGB information is used to guide the reconstruction and a low-rank regularization for compressive sensing and a non-convex surrogate of the rank is also used instead of nuclear norm for seeking a preferable solution. Experiments show that the proposed algorithm performs better in both PSNR and subjective effects compared with other state-of-art methods.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E101-A No.2 pp.481-487

Publication Date: 2018/02/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E101.A.481

Type of Manuscript: PAPER

Category: Image

Authors

Limin CHEN
  Nanchang University
Jing XU
  Nanchang University
Peter Xiaoping LIU
  Nanchang University,Carleton University
Hui YU
  Nanchang University

Keyword

hyperspectral image reconstruction, RGB-guide, low-rank, auxiliary information, compressive sensing

Cite this

Copy

Limin CHEN, Jing XU, Peter Xiaoping LIU, Hui YU, "A RGB-Guided Low-Rank Method for Compressive Hyperspectral Image Reconstruction" in IEICE TRANSACTIONS on Fundamentals, vol. E101-A, no. 2, pp. 481-487, February 2018, doi: 10.1587/transfun.E101.A.481.
Abstract: Compressive spectral imaging (CSI) systems capture the 3D spatiospectral data by measuring the 2D compressed focal plane array (FPA) coded projection with the help of reconstruction algorithms exploiting the sparsity of signals. However, the contradiction between the multi-dimension of the scenes and the limited dimension of the sensors has limited improvement of recovery performance. In order to solve the problem, a novel CSI system based on a coded aperture snapshot spectral imager, RGB-CASSI, is proposed, which has two branches, one for CASSI, another for RGB images. In addition, considering that conventional reconstruction algorithms lead to oversmoothing, a RGB-guided low-rank (RGBLR) method for compressive hyperspectral image reconstruction based on compressed sensing and coded aperture spectral imaging system is presented, in which the available additional RGB information is used to guide the reconstruction and a low-rank regularization for compressive sensing and a non-convex surrogate of the rank is also used instead of nuclear norm for seeking a preferable solution. Experiments show that the proposed algorithm performs better in both PSNR and subjective effects compared with other state-of-art methods.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/transfun.E101.A.481/_p

Copy

@ARTICLE{e101-a_2_481,
author={Limin CHEN, Jing XU, Peter Xiaoping LIU, Hui YU, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={A RGB-Guided Low-Rank Method for Compressive Hyperspectral Image Reconstruction},
year={2018},
volume={E101-A},
number={2},
pages={481-487},
abstract={Compressive spectral imaging (CSI) systems capture the 3D spatiospectral data by measuring the 2D compressed focal plane array (FPA) coded projection with the help of reconstruction algorithms exploiting the sparsity of signals. However, the contradiction between the multi-dimension of the scenes and the limited dimension of the sensors has limited improvement of recovery performance. In order to solve the problem, a novel CSI system based on a coded aperture snapshot spectral imager, RGB-CASSI, is proposed, which has two branches, one for CASSI, another for RGB images. In addition, considering that conventional reconstruction algorithms lead to oversmoothing, a RGB-guided low-rank (RGBLR) method for compressive hyperspectral image reconstruction based on compressed sensing and coded aperture spectral imaging system is presented, in which the available additional RGB information is used to guide the reconstruction and a low-rank regularization for compressive sensing and a non-convex surrogate of the rank is also used instead of nuclear norm for seeking a preferable solution. Experiments show that the proposed algorithm performs better in both PSNR and subjective effects compared with other state-of-art methods.},
keywords={},
doi={10.1587/transfun.E101.A.481},
ISSN={1745-1337},
month={February},}

Copy

TY - JOUR
TI - A RGB-Guided Low-Rank Method for Compressive Hyperspectral Image Reconstruction
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 481
EP - 487
AU - Limin CHEN
AU - Jing XU
AU - Peter Xiaoping LIU
AU - Hui YU
PY - 2018
DO - 10.1587/transfun.E101.A.481
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E101-A
IS - 2
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - February 2018
AB - Compressive spectral imaging (CSI) systems capture the 3D spatiospectral data by measuring the 2D compressed focal plane array (FPA) coded projection with the help of reconstruction algorithms exploiting the sparsity of signals. However, the contradiction between the multi-dimension of the scenes and the limited dimension of the sensors has limited improvement of recovery performance. In order to solve the problem, a novel CSI system based on a coded aperture snapshot spectral imager, RGB-CASSI, is proposed, which has two branches, one for CASSI, another for RGB images. In addition, considering that conventional reconstruction algorithms lead to oversmoothing, a RGB-guided low-rank (RGBLR) method for compressive hyperspectral image reconstruction based on compressed sensing and coded aperture spectral imaging system is presented, in which the available additional RGB information is used to guide the reconstruction and a low-rank regularization for compressive sensing and a non-convex surrogate of the rank is also used instead of nuclear norm for seeking a preferable solution. Experiments show that the proposed algorithm performs better in both PSNR and subjective effects compared with other state-of-art methods.
ER -