【摘要】 国内对转基因作物的监管非常严格,但是对转基因作物的检测缺乏快速准确的计量方法。太赫兹时域光谱结合机器学习分类算法可以实现对转基因作物快速有效地检测识别。通过太赫兹时域光谱技术提取了2种转基因油菜种子和一种非转基因油菜种子的太赫兹吸收谱,朴素贝叶斯算法、基于朴素贝叶斯的自适应提升算法、主成分分析结合随机森林算法、主成分分析结合支持向量计算法被应用于转基因油菜种子的太赫兹吸收谱的分类识别。通过实验对比,基于朴素贝叶斯的自适应提升算法获得了高达96.6%的检测准确率。该研究为运用太赫兹光谱技术手段开展转基因作物的快速检测提供方法参考。更多还原

【Abstract】 Regulation of genetically modified crops is very strict,but the detection of genetically modified crops lacks fast and accurate measurement methods.Terahertz time-domain spectroscopy combined with machine learning classification algorithm can achieve rapid and effective detection of genetically modified crops.The terahertz absorption spectra of two transgenic rapeseeds and one non-transgenic rapeseed are extracted by terahertz time-domain spectroscopy.Naive Bayes algorithm,adaptive boosting based on Naive Bayes algorithm,principal component analysis combined with random forest algorithm,principal component analysis and support vector machine are applied to classification and recognition of the terahertz absorption spectra of transgenic rapeseeds.Experimental comparison shows that the adaptive boosting based on Naive Bayes algorithm achieves a detection accuracy of up to 96.6%.This study provides a reference for the rapid detection of genetically modified crops using terahertz spectroscopy.更多还原