基于小波变换和神经网络的可见-近红外光谱对烟草品种的鉴别  

Discrimination of tobacco varieties using vis/near infrared spectra by wavelet transform and back propagation neural network model

作　　者：王遵义[1] 邵咏妮[2] 

机构地区：[1]浙江万里学院,浙江宁波315100 [2]浙江大学生物系统工程与食品科学学院,浙江杭州310029

出　　处：《浙江大学学报（农业与生命科学版）》2009年第6期655-658,共4页Journal of Zhejiang University:Agriculture and Life Sciences

基　　金：国家自然科学基金资助项目(30671213);浙江省宁波市重大科技攻关资助项目(2007C10034);浙江省宁波市自然科学基金资助项目(2007A610080)

摘　　要：为了实现烟草不同品种的快速光谱鉴别,采用主成分分析法对光谱数据进行聚类分析,并将小波变换用于对大量光谱数据的压缩,同时结合神经网络建立烟草品种鉴别模型.该模型将压缩后的数据作为神经网络的输入,加速了神经网络的训练速度.通过对4个品种的80个烟草样本建立训练模型,并用每个品种5个样本,共20个烟草样本进行预测.结果表明,用该方法对本研究4个品种的烟草鉴别正确率达100%.说明该方法具有很好的分类和鉴别作用,为烟草品种的快速鉴别提供了一种新方法.In order to discriminate quickly varieties of tobacco samples based on vis/near infrared spectroscopy, principal component analysis （PCA） was used to analyze qualitatively the clustering of tobacco samples, and the data compressed by wavelet transform （WT） combined with artificial neural network （ANN） was used to build discrimination model for tobacco varieties. This model regarded the compressed data as the input of ANN and accelerated the training speed of ANN. Eighty samples from four varieties were selected randomly to build training model, and the left 20 samples were used as predict set. The discrimination rate of 100% was achieved for these four varieties. It was indicated that WT combined with ANN is an available method for variety discrimination based on the vis/near infrared reflectance spectroscopy technology. It offered a new approach to the fast discrimination of tobacco varieties for future.

关 键 词：可见-近红外光谱 烟草 主成分分析 小波变换 人工神经网络 品种鉴别 

分 类 号：S572[农业科学—烟草工业]