基于太赫兹光谱和支持向量机快速鉴别咖啡豆产地

发布时间：2018-03-10 09:13

本文选题：光谱学　切入点：模型　出处：《农业工程学报》2017年09期 　论文类型：期刊论文

【摘要】：结合太赫兹时域光谱技术和支持向量机对3种典型产地的咖啡豆进行了鉴别。选取埃塞俄比亚(Ethiopia)、哥斯达黎加(Costa Rica)以及印度尼西亚(Indonesia)3个产地咖啡豆样品进行压片处理,采用太赫兹透射模式获取样品的时域和频域光谱信号,并用主成分分析法对太赫兹频域光谱信号进行分析;构造了基于粒子群(partical swarm optimization,PSO)参数寻优的支持向量机(support vector machine,SVM)鉴别模型,模型对不同产地咖啡豆样品的综合识别正确率达到95%。试验结果表明,太赫兹作为新型的检测手段结合模式识别方法可用于咖啡豆的产地鉴别。该文为一类在太赫兹波段下没有明显特征吸收峰的农产品/食品安全检测和产地追溯研究提供了一种快速、准确的方法。
[Abstract]:In combination with terahertz time-domain spectroscopy (THz) and support vector machine (SVM), the coffee beans from three typical areas were identified. The samples from Ethiopia (Ethiopia), Costa Rica (Costa Costa) and Indonesia (Indonesia) were treated by pressing. The time-domain and frequency-domain spectral signals of samples were obtained by terahertz transmission mode, and the spectral signals of terahertz frequency-domain were analyzed by principal component analysis (PCA). A support vector machine (SVM) discriminant model based on particle swarm optimization (PSO) parameters optimization was constructed. The comprehensive recognition accuracy of the model for coffee bean samples from different areas is 95%. The experimental results show that, Terahertz as a new detection method combined with pattern recognition can be used to identify the origin of coffee beans. This paper is a kind of agricultural products / food safety detection and origin traceability research without obvious absorption peak in terahertz band. The research provides a kind of rapidity, An accurate method.
【作者单位】：合肥工业大学计算机与信息学院;合肥学院机器视觉与智能控制实验室;合肥工业大学食品科学与工程学院;
【基金】：国家重点研发计划项目(2016YFD0401104)
【分类号】：S571.2;TP18

【相似文献】