摘要
采用小波变换对小麦近红外漫反射光谱进行阈值去噪研究,在软阈值函数下,分别采用固定阈值、启发式阈值、自适应阈值、最大最小阈值对光谱信号进行去噪,并用偏最小二乘建模的方法对去噪效果进行评价。结果表明,小波阈值降噪方法可以有效地降低原始光谱噪声,并很好地保留了信号中的尖锐和突变部分,而在阈值降噪过程中,最大最小阈值去噪效果最好,其建模后的绝对系数(R2)是0.8935,均方根误差(RMSE)是0.2886。
Threshold de-noising study of wheat NIR diffuse reflectance spectra by Wavelet transform,in soft threshold function,the spectral signals was de-noised by Fixed threshold,Heuristic threshold,Adaptive threshold,Minimax threshold Respectively,with the method Partial least-square modeling to evaluate the effect of the de-noising.The results show that the wavelet threshold method can effectively reduce the noise of the original spectrum,and retain the signal part of the sharp and abrupt well,in the threshold de-noising process,the best de-noising effect is the maximum and minimum threshold of the soft threshold function.Its absolute coefficient(R2) is 0.8935,and the mean square error(RMSE) is 0.2886 after modeling.
出处
《长春理工大学学报(自然科学版)》
2011年第1期27-29,共3页
Journal of Changchun University of Science and Technology(Natural Science Edition)
关键词
小波变换
阈值
降噪
近红外光谱
wavelet transform
threshold
de-noising
NIR spectroscopy