应用小波分析进行油菜籽红外光声光谱去噪

Denoising of Infrared Photoacoustic Spectra of Rapeseeds Using Wavelet Analysis

为探究小波分析对油菜籽红外光声光谱的去噪效果,利用db6小波4尺度分解对其进行去噪研究。对比分析低频系数重构、缺省阈值、Birge-Massart阈值和4种自适应阈值(Rigrsure、Minimaxi、Rigsure和Sqtwolog)模型等模型的去噪效果。同时与Savitzky-Golay卷积平滑和快速傅里叶变换的去噪效果进行比较。研究表明,Birg e-M assart阈值模型的综合小波去噪效果最好,同时小波去噪的方法较Savitzky-Golay卷积平滑和快速傅里叶变换去噪可以更好地捕获光谱的尖峰特征。

In order to explore the denosing effects of wavelet analysis on the infrared photoacoustic spectra of rapseeds,daubechies6 with 4 scale＇s wavelet decomposition was applied to denoise infrared photoacoutic spectra of rapeseed. The performances of different wavelet denoising methods including low-frequency coefficient reconstruction,default threshold,Birge-Massart threshold and self-adaptive threshold were evaluated,and then were compared with the standard Savitzky-Golay and fast Fourier transform denoising results. The results show that Birge-Massart threshold methods is the most effective for denoising, and the wavelet denosing methods are better at capturing characteristic bands in the spectra.