Application of wavelet transform in infrared spectrometry: spectral compression and library search
Application of wavelet transform in infrared spectrometry: spectral compression and library search
复制标题
DOI：
10.1016/s0169-7439(98)00084-7
复制
发表时间：
1998-09-28
影响因子：
3.9
通讯作者：
Shih, TM
中科院分区：
文献类型：
--
作者：
Leung, AKM;Chau, FT;Shih, TM
近年来，与快速傅里叶变换（FFT）相比，小波变换（WT）具有效率高、基函数数量多、数据处理速度快等优点，被广泛应用于分析化学的信号处理中。本文将快速小波变换（FWT）及其衍生物小波包变换（WPT）应用于红外光谱压缩，实现红外光谱的存储和搜索。在WT处理中，要处理的数据数必须为2(P)，其中P为任意整数。本文提出了系数位置保留（CPR）方法来处理长度为奇数的数据。将两种小波变换方法在数据压缩和谱库搜索方面的性能与FFT方法进行了比较。结果表明，本文提出的小波变换方法在红外光谱压缩和光谱库搜索方面优于FFT。(C) 1998 Elsevier Science B.V.版权所有
In recent years, a new mathematical technique called wavelet transform (WT) has been developed and adopted for signal processing in analytical chemistry owing to its efficiency, more number of basis functions available, and higher speed in data treatment compared to fast Fourier transform (FFT). In this paper, the fast wavelet transform (FWT) and its derivative, wavelet packet transform (WPT), were applied to compress infrared (IR) spectrum for storage and spectral searching. In WT treatment, the number of data to be processed has to be 2(P) with P being any integer. In this work, we proposed the coefficient position retaining (CPR) method to handle data with length of odd number. The performance of the two proposed WT methods in data compression and spectral library searching are compared with that of the FFT method. The results indicated that our proposed WT methods works better than FFT in compression of IR spectra and spectral Library searching. (C) 1998 Elsevier Science B.V. All rights reserved.