基于太赫兹时域光谱技术与PCA-SVM的转基因大豆油鉴别研究

发布时间：2018-02-11 03:10

本文关键词： 转基因大豆油太赫兹主成分分析支持向量机无损检测　出处：《红外与激光工程》2017年11期 　论文类型：期刊论文

【摘要】：太赫兹时域光谱(THz-TDS)技术是基于飞秒超快激光技术的THz波段光谱测量新技术,具有较强的光谱分辨本领以及良好的透视性和安全性,在物质检测方面具有广泛的应用价值。文中在采用太赫兹时域光谱技术对转基因大豆油光谱检测的基础上结合主成分分析方法(PCA)及支持向量机(SVM),构建PCA-SVM模型对转基因大豆油进行鉴别。首先,从样品在太赫兹波段测得的时域光谱中得到其吸光度光谱;然后,将其作为输入源导入PCA-SVM模型中,剔除冗余数据、降低数据维数并鉴别。实验结果表明,所建立的PCA-SVM模型能准确识别校验集,可以准确地对转基因大豆油进行鉴别。研究结果表明:太赫兹时域光谱技术可以实现转基因大豆油的快速、无损检测,在食品安全领域有广泛的应用前景。
[Abstract]:Terahertz time-domain spectroscopy (THz) is a new THz spectral measurement technique based on femtosecond ultra-fast laser technology. It has strong spectral resolution and good perspective and security. It has wide application value in substance detection. In this paper, the PCA-SVM model is constructed by using terahertz time-domain spectroscopy (THz) and principal component analysis (PCA) and support vector machine (SVM) to detect the spectrum of genetically modified soybean oil in combination with the principal component analysis method (PCA) and support vector machine (SVM). Genetic soybean oil identification. First, The absorbance spectra are obtained from the time-domain spectra of the samples measured in terahertz band, and then introduced into the PCA-SVM model as input sources, eliminating redundant data, reducing the dimension of data and discriminating. The experimental results show that, The established PCA-SVM model can accurately identify the calibration set and identify the transgenic soybean oil accurately. The results show that terahertz time-domain spectroscopy can be used for fast and nondestructive detection of transgenic soybean oil. It has a wide application prospect in the field of food safety.
【作者单位】：桂林电子科技大学电子工程与自动化学院;广西高校光电信息处理重点实验室;
【基金】：国家自然科学基金(61565004) 广西自然科学基金(2014GXNSFGA118003) 桂林市科学研究与技术开发课题(20140127-1,20150133-3)
【分类号】：O433.4

【相似文献】