木材近红外光谱小波阈值去噪方法

Threshold Denoising of Near Infrared Spectroscopy of Wood Based on Wavelet Transform

采用小波分析对木材近红外光谱进行阈值去噪研究,分别采用硬阈值、固定阈值、最大最小阈值和rigsure阈值对光谱信号进行去噪。结果表明,小波变换可以有效去除光谱噪声,并很好地保留了信号中的尖锐和突变部分,在近红外光谱分析中具有很好的应用前景。

The purpose of the study is to eliminate threshold denoising of near infrared spectrum of wood by wavelet analysis. Several multi-resolution wavelet threshold denoising methods are applied to eliminate threshold denoising, including hard threshold method, sqtwolog threshold method, minimaxi threshold method and Heursure threshold method. Results show that the wavelet transform has a particular advantage on noise eliminating of near infrared and has a wide application foreground.