小波分析在X荧光光谱去噪声中的应用

核辐射探测器中固有统计涨落、电子学噪声,使X射线荧光光谱中带有噪声,主要表现为出现假峰或丢失弱峰.论文探讨使用Mallat算法对X射线荧光光谱信号进行小波分解,采用阈值滤波算法在小波域内将噪声抑制或消除,最后重构除去噪声后的能谱信号.通过与多项式最小二乘拟合光滑除去噪声的光谱做定性和定量分析的比较,表明小波分析能够较好除去X射线荧光光谱中的噪声.

Wavelet Analysis in the Application of X Fluoresence Spectrum Denoising

The inherent statistical fluctuation and electronic noise in the nuclear radiation detectors result in noise in the X-ray fluorescence spectrum. The noise presents mainly in the form of fake pulse and fee-ble pulse. The paper probes into adopting the Mallat algorithm to carry out the wavelet decomposition of the X-ray fluorescence spectrum signal, and using threshold filtering algorithm to bate or eliminate the noise in Wavelet domain, finally reconstructing the spectrum signal in which the noise has been elimina-ted. By means of qualitative and quantitative analysis of the least-squares polynomial fitting method, it shows that Wavelet analysis will be a better way to remove the noise in the X-ray fluorescence spectrum.