摘要
食源性致病菌的检测分类一直是食品安全领域的重要研究对象,与传统的病原菌分类方法相比,基于拉曼光谱的分类识别方法具有更高的灵活性和准确性。实验以常见食源性致病菌的拉曼光谱为对象,利用11种病原菌的132条光谱数据,提出一种基于主成成分分析(PCA)和线性判别分析(LDA)的Adaboost集成分类识别模型。实验结果表明,该集成模型不仅优于传统的病原菌分类方法,而且分类准确率比决策树、支持向量机和logistic回归等单一算法模型更高,可以有效地对食源性致病菌进行分类,且分类准确率达到99.23%。
The detection and classification of food borne pathogens has always been an important research object in the field of food safety.Compared with traditional pathogen classification methods,the classification and recognition method based on Raman spectroscopy has higher flexibility and accuracy.Based on the Raman spectra of common food borne pathogens and 132 spectral data of 11 pathogens,an Adaboost integrated classification and recognition model based on principal component analysis(PCA)and linear discriminant analysis(LDA)is proposed.The experimental results show that the integrated model is not only superior to the traditional pathogen classification methods,but also has higher classification accuracy than single algorithm models such as decision tree,support vector machine and logistic regression.It can effectively classify food borne pathogens,and classification accuracy reaches 99.23%.
作者
黄杰伦
曾万聃
杨瑞君
吴敏
薛庆水
夏志平
HUANG Jielun;ZENG Wandan;YANG Ruijun;WU Min;XUE Qingshui;XIA Zhiping(School of Computer Science and Information Engineering,Shanghai Institute of Technology,Shanghai 201418,China;Military Veterinary Institute,Changchun 130022,China)
出处
《激光杂志》
CAS
北大核心
2022年第2期205-209,共5页
Laser Journal
基金
国家重点研发计划(No.2016YFC1201605)。
关键词
拉曼光谱
食源性致病菌
ADABOOST算法
判别分析
主成分分析
机器学习
raman spectroscopy
food borne pathogens
Aaboost algorithm
discriminant analysis
principal component analysis
machine learning