Iterative design of concentration factors for jump detection

Adityavikram Viswanathan, Anne Gelb, Douglas Cochran

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The concentration method of edge detection was developed to compute the locations and values of jump discontinuities in a piecewise-Analytic function from its first few Fourier series coefficients. The accuracy and characteristic features of the resulting jump approximation depend on Fourier space filter factors known as concentration factors. In this paper, we provide a flexible, iterative framework for the design of these factors. Previously devised concentration factors are shown to be the solutions of specific problem formulations within this new framework. We also provide sample formulations of the procedure applicable to the design of concentration factors for data with missing spectral bands. Several illustrative examples are used to demonstrate the capabilities of the method.

Original languageEnglish (US)
Pages (from-to)631-649
Number of pages19
JournalJournal of Scientific Computing
Volume51
Issue number3
DOIs
StatePublished - 2012

Fingerprint

Jump
Fourier series
Edge detection
Formulation
Edge Detection
Discontinuity
Analytic function
Design
Filter
Coefficient
Approximation
Demonstrate
Framework

Keywords

  • Concentration edge detection
  • Convex optimization
  • Jump detection

ASJC Scopus subject areas

  • Software
  • Computational Theory and Mathematics
  • Theoretical Computer Science
  • Engineering(all)

Cite this

Iterative design of concentration factors for jump detection. / Viswanathan, Adityavikram; Gelb, Anne; Cochran, Douglas.

In: Journal of Scientific Computing, Vol. 51, No. 3, 2012, p. 631-649.

Research output: Contribution to journalArticle

Viswanathan, Adityavikram ; Gelb, Anne ; Cochran, Douglas. / Iterative design of concentration factors for jump detection. In: Journal of Scientific Computing. 2012 ; Vol. 51, No. 3. pp. 631-649.
@article{e58efb98145245aab97ed4c56c69960f,
title = "Iterative design of concentration factors for jump detection",
abstract = "The concentration method of edge detection was developed to compute the locations and values of jump discontinuities in a piecewise-Analytic function from its first few Fourier series coefficients. The accuracy and characteristic features of the resulting jump approximation depend on Fourier space filter factors known as concentration factors. In this paper, we provide a flexible, iterative framework for the design of these factors. Previously devised concentration factors are shown to be the solutions of specific problem formulations within this new framework. We also provide sample formulations of the procedure applicable to the design of concentration factors for data with missing spectral bands. Several illustrative examples are used to demonstrate the capabilities of the method.",
keywords = "Concentration edge detection, Convex optimization, Jump detection",
author = "Adityavikram Viswanathan and Anne Gelb and Douglas Cochran",
year = "2012",
doi = "10.1007/s10915-011-9524-0",
language = "English (US)",
volume = "51",
pages = "631--649",
journal = "Journal of Scientific Computing",
issn = "0885-7474",
publisher = "Springer New York",
number = "3",

}

TY - JOUR

T1 - Iterative design of concentration factors for jump detection

AU - Viswanathan, Adityavikram

AU - Gelb, Anne

AU - Cochran, Douglas

PY - 2012

Y1 - 2012

N2 - The concentration method of edge detection was developed to compute the locations and values of jump discontinuities in a piecewise-Analytic function from its first few Fourier series coefficients. The accuracy and characteristic features of the resulting jump approximation depend on Fourier space filter factors known as concentration factors. In this paper, we provide a flexible, iterative framework for the design of these factors. Previously devised concentration factors are shown to be the solutions of specific problem formulations within this new framework. We also provide sample formulations of the procedure applicable to the design of concentration factors for data with missing spectral bands. Several illustrative examples are used to demonstrate the capabilities of the method.

AB - The concentration method of edge detection was developed to compute the locations and values of jump discontinuities in a piecewise-Analytic function from its first few Fourier series coefficients. The accuracy and characteristic features of the resulting jump approximation depend on Fourier space filter factors known as concentration factors. In this paper, we provide a flexible, iterative framework for the design of these factors. Previously devised concentration factors are shown to be the solutions of specific problem formulations within this new framework. We also provide sample formulations of the procedure applicable to the design of concentration factors for data with missing spectral bands. Several illustrative examples are used to demonstrate the capabilities of the method.

KW - Concentration edge detection

KW - Convex optimization

KW - Jump detection

UR - http://www.scopus.com/inward/record.url?scp=84905866839&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905866839&partnerID=8YFLogxK

U2 - 10.1007/s10915-011-9524-0

DO - 10.1007/s10915-011-9524-0

M3 - Article

VL - 51

SP - 631

EP - 649

JO - Journal of Scientific Computing

JF - Journal of Scientific Computing

SN - 0885-7474

IS - 3

ER -