摘要
为建立一种快速、准确、无损的眼药水塑料瓶的分类方法,利用X射线荧光光谱,在电流为200μA、电压为50 kV、功率为10 W、测试时间为60 s的条件下,对33个不同厂家、不同品牌、不同批次的眼药水塑料瓶样品进行检验,根据人工分类与系统聚类的方法将33个样品分为4类后,分别采用一对一法构建多元分类器和K-fold交叉验证2种方法构建支持向量机模型,前者预测准确率为66.67%,后者预测准确率为84.8%。采用K-fold交叉验证的方法构建支持向量机,能够较为准确地鉴别未知眼药水塑料瓶的类别,该方法可为公安办案提供参考。
In order to establish a fast,accurate and nondestructive classification method for eye drops plastic bottles,X-ray fluorescence spectroscopy was used to inspect 33 samples of eye drops plastic bottles from different manufacturers,brands and batches under the conditions of 200μA current,50 kV voltage,10 W power and 60 s test time.After 33 samples were divided into four categories according to manual classification and systematic clustering,the support vector machine models were constructed by one to one multivariate classifier and K-fold cross validation respectively.The prediction accuracy of the former was 66.67%,and the prediction accuracy of the latter was 84.8%.The support vector machine constructed by K-fold cross validation method is more effective in identifying the unknown types of eye drops plastic bottles,which can provide reference for the police to handle cases.
作者
李若琳
陈丽萍
姜红
杨俊
满吉
LI Ruolin;CHEN Liping;JIANG Hong;YANG Jun;MAN Ji(Investigation Institute,People’s Public Security University of China,Beijing 100038,China;Beijing Huayi Hongsheng Technology Co.,Ltd.,Beijing 100123,China)
出处
《上海塑料》
2022年第6期56-63,共8页
Shanghai Plastics
基金
中国人民公安大学2021年度基科费重点项目(2021JKF212)
国家重点研发计划项目(2019YFF0303405)。
关键词
眼药水塑料瓶
X射线荧光光谱
系统聚类
支持向量机
K-fold交叉验证
eye drops plastic bottle
X-ray fluorescence spectrum
systematic clustering
support vector machine
K-fold cross validation
