改进小波阈值方法对γ能谱去噪的研究  

Study of γ Energy Spectrum Denoising Based on Improved Wavelet Threshold Method

作　　者：李伦辉[1] 何剑锋[1,2] 王芹[1] 王远[1] 

机构地区：[1]东华理工大学信息工程学院,江西南昌330013 [2]东华理工大学软件学院,江西南昌330013

出　　处：《原子能科学技术》2016年第7期1279-1283,共5页Atomic Energy Science and Technology

基　　金：国家自然科学基金资助项目(11365001);江西省教育厅科学技术研究项目资助(GJJ13464);江西省研究生创新专项资助项目(YC2014-S335)

摘　　要：在γ能谱测量过程中,如何有效地对谱数据进行平滑处理,关系到峰位道址的确定和峰面积的计算是否准确。本文基于小波分析,针对影响去噪结果的关键要素——阈值选择方法和阈值函数,分析两者的缺陷并对其进行改进。研究结果表明:改进后小波阈值方法的平滑度指标和去噪前后能谱能量比值分别为0.985 3和0.999 8,其去噪效果明显优于其他方法。In the process of γ energy spectrum measurement,how to smooth spectrum data effectively concerns the accuracies of determination of the peak channel address and the calculation of the peak area.Based on the wavelet analysis,the imperfections of key factors that influence the results of denoising（threshold selection method and threshold function）were analyzed and improved.The results show that the smoothness index and the ratio of energy of denoised energy spectrum and original energy spectrum for the improved wavelet threshold method are 0.985 3 and 0.999 8,respectively,and the denoising effect of the improved wavelet threshold method is better than that of other methods.

关 键 词：小波去噪 阈值函数 平滑 Γ能谱 

分 类 号：TL812[核科学技术—核技术及应用] TP391.9[自动化与计算机技术—计算机应用技术]