Information optimal compressive sensing: Static measurement design

Amit Ashok, Liang Chih Huang, Mark A Neifeld

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

The compressive sensing paradigm exploits the inherent sparsity/ compressibility of signals to reduce the number of measurements required for reliable reconstruction/recovery. In many applications additional prior information beyond signal sparsity, such as structure in sparsity, is available, and current efforts are mainly limited to exploiting that information exclusively in the signal reconstruction problem. In this work, we describe an information-theoretic framework that incorporates the additional prior information as well as appropriate measurement constraints in the design of compressive measurements. Using a Gaussian binomial mixture prior we design and analyze the performance of optimized projections relative to random projections under two specific design constraints and different operating measurement signal-to-noise ratio (SNR) regimes. We find that the information-optimized designs yield significant, in some cases nearly an order of magnitude, improvements in the reconstruction performance with respect to the random projections. These improvements are especially notable in the low measurement SNR regime where the energy-efficient design of optimized projections is most advantageous. In such cases, the optimized projection design departs significantly from random projections in terms of their incoherence with the representation basis. In fact, we find that the maximizing incoherence of projections with the representation basis is not necessarily optimal in the presence of additional prior information and finite measurement noise/error. We also apply the information-optimized projections to the compressive image formation problem for natural scenes, and the improved visual quality of reconstructed images with respect to random projections and other compressive measurement design affirms the overall effectiveness of the information- theoretic design framework.

Original languageEnglish (US)
Pages (from-to)831-853
Number of pages23
JournalJournal of the Optical Society of America A: Optics and Image Science, and Vision
Volume30
Issue number5
DOIs
StatePublished - 2013

Fingerprint

Signal-To-Noise Ratio
projection
Noise
incoherence
Signal to noise ratio
signal to noise ratios
Signal reconstruction
Compressibility
noise measurement
Image processing
compressibility
recovery
Recovery

Keywords

  • (100.7410) Wavelets
  • (110.1758) Computational imaging
  • (110.3055) Information theoretical analysis

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials

Cite this

@article{042982b709b14f54aee92fa27c3a2521,
title = "Information optimal compressive sensing: Static measurement design",
abstract = "The compressive sensing paradigm exploits the inherent sparsity/ compressibility of signals to reduce the number of measurements required for reliable reconstruction/recovery. In many applications additional prior information beyond signal sparsity, such as structure in sparsity, is available, and current efforts are mainly limited to exploiting that information exclusively in the signal reconstruction problem. In this work, we describe an information-theoretic framework that incorporates the additional prior information as well as appropriate measurement constraints in the design of compressive measurements. Using a Gaussian binomial mixture prior we design and analyze the performance of optimized projections relative to random projections under two specific design constraints and different operating measurement signal-to-noise ratio (SNR) regimes. We find that the information-optimized designs yield significant, in some cases nearly an order of magnitude, improvements in the reconstruction performance with respect to the random projections. These improvements are especially notable in the low measurement SNR regime where the energy-efficient design of optimized projections is most advantageous. In such cases, the optimized projection design departs significantly from random projections in terms of their incoherence with the representation basis. In fact, we find that the maximizing incoherence of projections with the representation basis is not necessarily optimal in the presence of additional prior information and finite measurement noise/error. We also apply the information-optimized projections to the compressive image formation problem for natural scenes, and the improved visual quality of reconstructed images with respect to random projections and other compressive measurement design affirms the overall effectiveness of the information- theoretic design framework.",
keywords = "(100.7410) Wavelets, (110.1758) Computational imaging, (110.3055) Information theoretical analysis",
author = "Amit Ashok and Huang, {Liang Chih} and Neifeld, {Mark A}",
year = "2013",
doi = "10.1364/JOSAA.30.000831",
language = "English (US)",
volume = "30",
pages = "831--853",
journal = "Journal of the Optical Society of America A: Optics and Image Science, and Vision",
issn = "1084-7529",
publisher = "The Optical Society",
number = "5",

}

TY - JOUR

T1 - Information optimal compressive sensing

T2 - Static measurement design

AU - Ashok, Amit

AU - Huang, Liang Chih

AU - Neifeld, Mark A

PY - 2013

Y1 - 2013

N2 - The compressive sensing paradigm exploits the inherent sparsity/ compressibility of signals to reduce the number of measurements required for reliable reconstruction/recovery. In many applications additional prior information beyond signal sparsity, such as structure in sparsity, is available, and current efforts are mainly limited to exploiting that information exclusively in the signal reconstruction problem. In this work, we describe an information-theoretic framework that incorporates the additional prior information as well as appropriate measurement constraints in the design of compressive measurements. Using a Gaussian binomial mixture prior we design and analyze the performance of optimized projections relative to random projections under two specific design constraints and different operating measurement signal-to-noise ratio (SNR) regimes. We find that the information-optimized designs yield significant, in some cases nearly an order of magnitude, improvements in the reconstruction performance with respect to the random projections. These improvements are especially notable in the low measurement SNR regime where the energy-efficient design of optimized projections is most advantageous. In such cases, the optimized projection design departs significantly from random projections in terms of their incoherence with the representation basis. In fact, we find that the maximizing incoherence of projections with the representation basis is not necessarily optimal in the presence of additional prior information and finite measurement noise/error. We also apply the information-optimized projections to the compressive image formation problem for natural scenes, and the improved visual quality of reconstructed images with respect to random projections and other compressive measurement design affirms the overall effectiveness of the information- theoretic design framework.

AB - The compressive sensing paradigm exploits the inherent sparsity/ compressibility of signals to reduce the number of measurements required for reliable reconstruction/recovery. In many applications additional prior information beyond signal sparsity, such as structure in sparsity, is available, and current efforts are mainly limited to exploiting that information exclusively in the signal reconstruction problem. In this work, we describe an information-theoretic framework that incorporates the additional prior information as well as appropriate measurement constraints in the design of compressive measurements. Using a Gaussian binomial mixture prior we design and analyze the performance of optimized projections relative to random projections under two specific design constraints and different operating measurement signal-to-noise ratio (SNR) regimes. We find that the information-optimized designs yield significant, in some cases nearly an order of magnitude, improvements in the reconstruction performance with respect to the random projections. These improvements are especially notable in the low measurement SNR regime where the energy-efficient design of optimized projections is most advantageous. In such cases, the optimized projection design departs significantly from random projections in terms of their incoherence with the representation basis. In fact, we find that the maximizing incoherence of projections with the representation basis is not necessarily optimal in the presence of additional prior information and finite measurement noise/error. We also apply the information-optimized projections to the compressive image formation problem for natural scenes, and the improved visual quality of reconstructed images with respect to random projections and other compressive measurement design affirms the overall effectiveness of the information- theoretic design framework.

KW - (100.7410) Wavelets

KW - (110.1758) Computational imaging

KW - (110.3055) Information theoretical analysis

UR - http://www.scopus.com/inward/record.url?scp=84877597690&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84877597690&partnerID=8YFLogxK

U2 - 10.1364/JOSAA.30.000831

DO - 10.1364/JOSAA.30.000831

M3 - Article

C2 - 23695314

AN - SCOPUS:84877597690

VL - 30

SP - 831

EP - 853

JO - Journal of the Optical Society of America A: Optics and Image Science, and Vision

JF - Journal of the Optical Society of America A: Optics and Image Science, and Vision

SN - 1084-7529

IS - 5

ER -