A de-noising algorithm to improve SNR of segmented gamma scanner for spectrum analysis

被引：10

作者：

Li, Huailiang ^{[1
]}

Tuo, Xianguo ^{[1
,2
]}

Shi, Rui ^{[2
]}

Zhang, Jinzhao ^{[1
]}

Henderson, Mark Julian ^{[1
]}

Courtois, Jeremie ^{[3
]}

Yan, Minhao ^{[3
]}

机构：

[1] Southwest Univ Sci & Technol, Fundamental Sci Nucl Wastes & Environm Safety Lab, Mianyang 621010, Peoples R China

[2] Chengdu Univ Technol, State Key Lab Geohazard Prevent & Geoenvironm Pro, Chengdu 610059, Peoples R China

[3] Southwest Univ Sci & Technol, State Key Lab Cultivat Base Nonmet Composites & F, Mianyang 621010, Peoples R China

来源：

NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT | 2016年 / 818卷

基金：

中国国家自然科学基金;

关键词：

gamma-spectrum analysis; Shift-invariant wavelet transform; De-noising; TRANSFORM; SYSTEM;

D O I：

10.1016/j.nima.2016.02.047

中图分类号：

TH7 [仪器、仪表];

学科分类号：

0804 ; 080401 ; 081102 ;

摘要：

An improved threshold shift-invariant wavelet transform de-noising algorithm for high-resolution gamma-ray spectroscopy is proposed to optimize the threshold function of wavelet transforms and reduce signal resulting from pseudo-Gibbs artificial fluctuations. This algorithm was applied to a segmented gamma scanning system with large samples in which high continuum levels caused by Compton scattering are routinely encountered. De-noising data from the gamma ray spectrum measured by segmented gamma scanning system with improved, shift-invariant and traditional wavelet transform algorithms were all evaluated. The improved wavelet transform method generated significantly enhanced performance of the figure of merit, the root mean square error, the peak area, and the sample attenuation correction in the segmented gamma scanning system assays. We also found that the gamma energy spectrum can be viewed as a low frequency signal as well as high frequency noise superposition by the spectrum analysis. Moreover, a smoothed spectrum can be appropriate for straightforward automated quantitative analysis. (C) 2016 Elsevier B.V. All rights reserved.

引用

页码：68 / 75

页数：8

共 23 条

[1] The near shift-invariance of the dual-tree complex wavelet transform revisited [J].