一种基于小波变换的近红外化学指纹图谱分析方法

提出了一种快速无损的化学指纹图谱分析新方法. 根据小波变换多分辨分析的特点, 对近红外(NIR)光谱进行小波分解, 从中提取被测样品的化学特征信息, 并作数字信息可视化处理, 构建形成可直观识别样品模式特征的NIR指纹图谱. 将该方法用于中药材丹参的质量检测, 检测结果与色谱指纹图谱检测结果相符, 能快速有效地识别丹参质量模式间的差异, 有望发展成为一种快速分析检测天然产物质量的方法.

Near Infrared Chemical Fingerprinting Based on Wavelet Transform

Based on wavelet transform and data visualization, a type of NIR fingerprinting was proposed and applied to assessing the quality of Danshen(Radix salviae miltirrhizae). First, wavelet transform was performed to extract the feature information from the raw NIR spectra. And then, the extracted information was represented by the data visualization. The integral pattern difference of the Danshen samples could be identified intuitively with the proposed method. HPLC fingerprints were also collected, and they were used as the reference. It was shown that the results obtained from NIR fingerprint was consistent with that of the corresponding HPLC fingerprint. As a fast and non-destructive method for quality assessment of natural products, the potential of NIR fingerprinting is exemplarily shown in this paper.