同态滤波法消除γ能谱混合噪声  

Eliminating the Mixed Noise in γ Spectra by the Homomorphic Filtering Method

作　　者：王崇杰[1] 范成杰 吉鹏 徐杨 田源 杜文艳 孙翱 

机构地区：[1]辽宁师范大学物理与电子技术学院,辽宁大连116029

出　　处：《核电子学与探测技术》2017年第3期323-328,共6页Nuclear Electronics & Detection Technology

基　　金：辽宁教育厅科研基金(L2013410);国家自然科学基金委与中国科学院大科学装置联合基金(U1332201)资助

摘　　要：针对γ能谱传统消噪方法的不足,提出了γ能谱的混合噪声模型和基于小波变换的同态滤波消噪法。本文对γ能谱存在乘性噪声进行了论证,并详细介绍了消噪算法。通过Matlab软件对^(60)Co γ能谱消噪处理结果表明,与传统消噪方法相比,以混合噪声模型为基础的同态滤波法较大幅度地提高了γ能谱的信噪比,同时还在一定程度上消除了谱线的"展宽效应"。研究表明,混合噪声模型和同态滤波消噪法是合理有效的,而且消噪结果优于传统消噪方法。Aiming at the deficiencies of the traditional de - noising methods, a mixed noise model of γ spectra and the homomorphic filtering de - noising method based on wavelet transform are proposed. The existence of multiplicative noise in γ spectra is demonstrated, and the de - noising algorithm of homomorphic filtering is introduced in detail. The de - noising of 60 Co γ spectra with different signal - to - ratios is conducted by Matlab software. The results show that the homomorphic filtering de - noising method based on the mixed noise model can not only improve the SNR more greatly than the traditional additive noise model, but also eliminate the＂ broadening effect＂ of γ spectra to a certain extent. The research shows that the mixed noise model and homomorphic filtering method are reasonable and effective, and the de - noising results are better than the traditional de - noising methods.

关 键 词：γ能谱分析 随机过程 非线性 时变性 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]