摘要
为建立一种快速无损分类塑料瓶盖的方法,采用高光谱成像技术对48个塑料瓶盖样品进行检验。首先对原始光谱进行预处理,再分别采用主成分分析法、偏最小二乘-判别分析法和竞争自适应重加权采样法构建高光谱数据集,并对数据集分别使用支持向量机、多层感知机模型和卷积神经网络进行训练。结果表明:利用竞争自适应重加权特征提取构建的塑料瓶盖高光谱图像,在卷积神经网络中的测试集准确率达到了100%。该方法方便快捷,对样品无损且用量少,为塑料瓶盖的分类提供了有力的支持。
In order to establish a fast and non-destructive analytical method for plastic bottle cap inspection,a hyperspectral imaging system was used to inspect 48 plastic bottle cap samples.Firstly,the original spectra were preprocessed,and then principal component analysis,partial least squares-discriminant analysis,and competitive adaptive reweighted sampling were used to construct hyperspectral datasets.Support vector machines,multi-layer perceptron models,and convolutional neural networks were used to train the datasets.The results show that the hyperspectral images of plastic bottle caps constructed using competitive adaptive reweighting sampling extraction achieved an accuracy of 100%in the test set of convolutional neural networks.This method is convenient,fast,non-destructive,and requires minimal usage,providing strong support for the classification of plastic bottle caps.
作者
周飞翔
姜红
钟方昊
周贯旭
刘业林
ZHOU Feixiang;JIANG Hong;ZHONG Fanghao;ZHOU Guanxu;LIU Yelin(Collage of Investigation,People’s Public Security University of China,Beijing 100038,China;Judicial Appraisal Center of Wanzijian Testing Technology Co.,Ltd.,Beijing 100141,China;Jiangsu Dualix Spectral Imaging Co.,Ltd.,Wuxi 214000,Jiangsu,China)
出处
《上海塑料》
CAS
2024年第2期54-59,共6页
Shanghai Plastics
基金
食品药品安全防范山西省重点实验室开放课题(202204010931006)。
关键词
高光谱技术
塑料瓶盖
偏最小二乘-判别分析
竞争自适应重加权采样
卷积神经网络
hyperspectral technique
plastic bottle cap
partial least squares-discriminant analysis
competitive adaptive reweighted sampling
convolutional neural network
