Feature-specific difference imaging

Shikhar Uttam, Nathan A. Goodman, Mark A Neifeld

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Difference images quantify changes in the object scene over time. In this paper, we use the feature-specific imaging paradigm to present methods for estimating a sequence of difference images from a sequence of compressive measurements of the object scene. Our goal is twofold. First is to design, where possible, the optimal sensing matrix for taking compressive measurements. In scenarios where such sensing matrices are not tractable, we consider plausible candidate sensing matrices that either use the available a priori information or are nonadaptive. Second, we develop closed-form and iterative techniques for estimating the difference images. We specifically look at l 2-and l 1-based methods. We show that l 2-based techniques can directly estimate the difference image from the measurements without first reconstructing the object scene. This direct estimation exploits the spatial and temporal correlations between the object scene at two consecutive time instants. We further develop a method to estimate a generalized difference image from multiple measurements and use it to estimate the sequence of difference images. For l 1-based estimation, we consider modified forms of the total-variation method and basis pursuit denoising. We also look at a third method that directly exploits the sparsity of the difference image. We present results to show the efficacy of these techniques and discuss the advantages of each.

Original languageEnglish (US)
Article number5993541
Pages (from-to)638-652
Number of pages15
JournalIEEE Transactions on Image Processing
Volume21
Issue number2
DOIs
StatePublished - Feb 2012

Fingerprint

Imaging techniques

Keywords

  • Compressive sensing (CS)
  • difference images
  • feature-specific imaging (FSI)
  • l -reconstruction
  • l -reconstruction

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software
  • Medicine(all)

Cite this

Feature-specific difference imaging. / Uttam, Shikhar; Goodman, Nathan A.; Neifeld, Mark A.

In: IEEE Transactions on Image Processing, Vol. 21, No. 2, 5993541, 02.2012, p. 638-652.

Research output: Contribution to journalArticle

Uttam, Shikhar ; Goodman, Nathan A. ; Neifeld, Mark A. / Feature-specific difference imaging. In: IEEE Transactions on Image Processing. 2012 ; Vol. 21, No. 2. pp. 638-652.
@article{577958ff46474f83909ca51bb2f20054,
title = "Feature-specific difference imaging",
abstract = "Difference images quantify changes in the object scene over time. In this paper, we use the feature-specific imaging paradigm to present methods for estimating a sequence of difference images from a sequence of compressive measurements of the object scene. Our goal is twofold. First is to design, where possible, the optimal sensing matrix for taking compressive measurements. In scenarios where such sensing matrices are not tractable, we consider plausible candidate sensing matrices that either use the available a priori information or are nonadaptive. Second, we develop closed-form and iterative techniques for estimating the difference images. We specifically look at l 2-and l 1-based methods. We show that l 2-based techniques can directly estimate the difference image from the measurements without first reconstructing the object scene. This direct estimation exploits the spatial and temporal correlations between the object scene at two consecutive time instants. We further develop a method to estimate a generalized difference image from multiple measurements and use it to estimate the sequence of difference images. For l 1-based estimation, we consider modified forms of the total-variation method and basis pursuit denoising. We also look at a third method that directly exploits the sparsity of the difference image. We present results to show the efficacy of these techniques and discuss the advantages of each.",
keywords = "Compressive sensing (CS), difference images, feature-specific imaging (FSI), l -reconstruction, l -reconstruction",
author = "Shikhar Uttam and Goodman, {Nathan A.} and Neifeld, {Mark A}",
year = "2012",
month = "2",
doi = "10.1109/TIP.2011.2165549",
language = "English (US)",
volume = "21",
pages = "638--652",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Feature-specific difference imaging

AU - Uttam, Shikhar

AU - Goodman, Nathan A.

AU - Neifeld, Mark A

PY - 2012/2

Y1 - 2012/2

N2 - Difference images quantify changes in the object scene over time. In this paper, we use the feature-specific imaging paradigm to present methods for estimating a sequence of difference images from a sequence of compressive measurements of the object scene. Our goal is twofold. First is to design, where possible, the optimal sensing matrix for taking compressive measurements. In scenarios where such sensing matrices are not tractable, we consider plausible candidate sensing matrices that either use the available a priori information or are nonadaptive. Second, we develop closed-form and iterative techniques for estimating the difference images. We specifically look at l 2-and l 1-based methods. We show that l 2-based techniques can directly estimate the difference image from the measurements without first reconstructing the object scene. This direct estimation exploits the spatial and temporal correlations between the object scene at two consecutive time instants. We further develop a method to estimate a generalized difference image from multiple measurements and use it to estimate the sequence of difference images. For l 1-based estimation, we consider modified forms of the total-variation method and basis pursuit denoising. We also look at a third method that directly exploits the sparsity of the difference image. We present results to show the efficacy of these techniques and discuss the advantages of each.

AB - Difference images quantify changes in the object scene over time. In this paper, we use the feature-specific imaging paradigm to present methods for estimating a sequence of difference images from a sequence of compressive measurements of the object scene. Our goal is twofold. First is to design, where possible, the optimal sensing matrix for taking compressive measurements. In scenarios where such sensing matrices are not tractable, we consider plausible candidate sensing matrices that either use the available a priori information or are nonadaptive. Second, we develop closed-form and iterative techniques for estimating the difference images. We specifically look at l 2-and l 1-based methods. We show that l 2-based techniques can directly estimate the difference image from the measurements without first reconstructing the object scene. This direct estimation exploits the spatial and temporal correlations between the object scene at two consecutive time instants. We further develop a method to estimate a generalized difference image from multiple measurements and use it to estimate the sequence of difference images. For l 1-based estimation, we consider modified forms of the total-variation method and basis pursuit denoising. We also look at a third method that directly exploits the sparsity of the difference image. We present results to show the efficacy of these techniques and discuss the advantages of each.

KW - Compressive sensing (CS)

KW - difference images

KW - feature-specific imaging (FSI)

KW - l -reconstruction

KW - l -reconstruction

UR - http://www.scopus.com/inward/record.url?scp=84856276055&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84856276055&partnerID=8YFLogxK

U2 - 10.1109/TIP.2011.2165549

DO - 10.1109/TIP.2011.2165549

M3 - Article

VL - 21

SP - 638

EP - 652

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 2

M1 - 5993541

ER -