基于小波变换和高斯拟合的在线谱图综合处理方法

An Integrated On-Line Processing Method for Spectrometric Data Based on Wavelet Transform and Gaussian Fitting

微小型移动式现场在线检测技术是分析仪器发展的新领域。针对复杂工作环境中谱图存在强噪声干扰、谱峰重叠、不规则峰形等严重影响仪器的定性和定量准确度的瓶颈技术,提出了一种基于小波变换和高斯拟合相结合的谱图在线综合处理方法,用自研的仪器对甲苯和全氟三丁胺两种典型化合物的谱图进行了处理,并与实验室分析仪器普遍应用的算法进行了对比分析。结果表明,综合方法能够有效解决强噪声干扰、谱峰重叠、不规则峰形问题,提高仪器的定性和定量准确性,同时能够实现数据压缩,满足仪器的在线实时检测要求。综合方法处理甲苯特征峰的平均信噪比(SNR)较移动平滑方法提高了1.3倍,峰位误差ΔM降低了3.6倍,处理全氟三丁胺谱图的数据压缩比为197∶1。

Miniature mobile field spectrometry is pivotal equipment for qualitative and quantitative in-situ analysis of chemical substances.To solve the problem of spectrum signal interfered by complicated noise,overlapped and irregular peak shape recognition,and quick monitoring,an integrated on-line processing method for spectrometric data based on wavelet transform and Gaussian fitting was developed.In this way,toluene and perfluorotributylamine were processed,and the results shows that the integrated method can powerfully and effectively eliminate the noise,retain the original feature,and correct the overlapped and asymmetrical peaks,which can improve the analysis accuracy of instrument,and also achieve data compression.In addition,the method satisfies the requirement of on-site analysis for mobile field spectrometry.For the processing of mass spectra of toluene,at the characteristic peaks of 91 and 92,the SNR increased 1.3 times compared to that of moving average smoothing method,while the error between original peaks and theoretic peaks decreased 3.6 times.In addition,Gaussian fitting described the multipoint mass spectra data by three Gaussian parameters,and achieved data compression.For the processing of mass spectrogram of perfluorotributylamine,the ratio of compression was 197∶1.