Bayesian PIXE background subtraction

J. Padayachee, V. Prozesky, W. von der Linden, M. S. Nkwinika, V. Dose

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The subtraction of the X-ray background in a PIXE spectrum has been the subject of many investigations and different techniques have been developed. These techniques vary from filtering to fitting polynomial functions. The promising Bayesian Statistics technique has been used in this study to eliminate the background from the spectrum in a rigorous and self-consistent manner. We compare the results of the Bayesian background subtraction method to that obtained by stripping and the “rolling ball”; method.
Original languageEnglish
Pages (from-to)129-135
Number of pages7
JournalNuclear instruments & methods in physics research / B
Volume150
Issue number1–4
DOIs
Publication statusPublished - 1 Apr 1999

Fingerprint

subtraction
Statistics
Polynomials
Bayes theorem
X rays
stripping
balls
polynomials
x rays

Keywords

  • Background subtraction, Bayesian data analysis

Cite this

Bayesian PIXE background subtraction. / Padayachee, J.; Prozesky, V.; von der Linden, W.; Nkwinika, M. S.; Dose, V.

In: Nuclear instruments & methods in physics research / B, Vol. 150, No. 1–4, 01.04.1999, p. 129-135.

Research output: Contribution to journalArticleResearchpeer-review

Padayachee, J. ; Prozesky, V. ; von der Linden, W. ; Nkwinika, M. S. ; Dose, V. / Bayesian PIXE background subtraction. In: Nuclear instruments & methods in physics research / B. 1999 ; Vol. 150, No. 1–4. pp. 129-135.
@article{c41d2683302345b3832ed0605e5717a4,
title = "Bayesian PIXE background subtraction",
abstract = "The subtraction of the X-ray background in a PIXE spectrum has been the subject of many investigations and different techniques have been developed. These techniques vary from filtering to fitting polynomial functions. The promising Bayesian Statistics technique has been used in this study to eliminate the background from the spectrum in a rigorous and self-consistent manner. We compare the results of the Bayesian background subtraction method to that obtained by stripping and the “rolling ball”; method.",
keywords = "Background subtraction, Bayesian data analysis",
author = "J. Padayachee and V. Prozesky and {von der Linden}, W. and Nkwinika, {M. S.} and V. Dose",
year = "1999",
month = "4",
day = "1",
doi = "10.1016/S0168-583X(98)01063-5",
language = "English",
volume = "150",
pages = "129--135",
journal = "Nuclear instruments & methods in physics research / B",
issn = "0168-583X",
publisher = "Elsevier B.V.",
number = "1–4",

}

TY - JOUR

T1 - Bayesian PIXE background subtraction

AU - Padayachee, J.

AU - Prozesky, V.

AU - von der Linden, W.

AU - Nkwinika, M. S.

AU - Dose, V.

PY - 1999/4/1

Y1 - 1999/4/1

N2 - The subtraction of the X-ray background in a PIXE spectrum has been the subject of many investigations and different techniques have been developed. These techniques vary from filtering to fitting polynomial functions. The promising Bayesian Statistics technique has been used in this study to eliminate the background from the spectrum in a rigorous and self-consistent manner. We compare the results of the Bayesian background subtraction method to that obtained by stripping and the “rolling ball”; method.

AB - The subtraction of the X-ray background in a PIXE spectrum has been the subject of many investigations and different techniques have been developed. These techniques vary from filtering to fitting polynomial functions. The promising Bayesian Statistics technique has been used in this study to eliminate the background from the spectrum in a rigorous and self-consistent manner. We compare the results of the Bayesian background subtraction method to that obtained by stripping and the “rolling ball”; method.

KW - Background subtraction, Bayesian data analysis

U2 - 10.1016/S0168-583X(98)01063-5

DO - 10.1016/S0168-583X(98)01063-5

M3 - Article

VL - 150

SP - 129

EP - 135

JO - Nuclear instruments & methods in physics research / B

JF - Nuclear instruments & methods in physics research / B

SN - 0168-583X

IS - 1–4

ER -