作　　者： ; ; ; ; ;

机构地区： 浙江大学

出　　处： 《传感技术学报》 2011年第7期941-945,共5页

摘　　要： 采用电子鼻对三种不同年份的番茄种子进行分析。结果表明:利用电子鼻可以很好的区分不同年份的番茄种子;利用主成分分析方法(PCA)基本上可以辨别出不同掺杂比例的种子,但是当掺杂比例为37.5%和50%时,较难利用电子鼻进行辨别区分;利用线性判别分析方法(LDA)可以很好的辨别出不同掺杂比例的番茄种子,并且每个混合种类的区域集中性都很好,所以相对于PCA方法,利用LDA方法得到的数据区域集中性要明显优于PCA方法得到的结果。在主成分分析和线性分析的基础上,利用BP神经网络和支持向量机两种模式对相互掺杂的种子进行分类识别,结果表明:相对于支持向量机模式识别,BP神经网络预测系统的误差较小,具有很好的预测性能。 An investigation was made to evaluate the capacity of an electronic nose to classify the tomato seeds with different storage time.The tomato seeds with different storage time were blended each other at different proportion.The data were processed by Principal Component Analysis（PCA）and Linear Discrimination Analysis（LDA）.The result shows that the electronic nose can distinguish the different storage time tomato seeds without blending easily.The tomato seeds blended each other at different proportion are discriminated completely by LDA;however,samples with proportion of 37.5% and 50% are overlapped by PCA.The e-nose can distinguish the different time tomato seeds which blended each at different proportion by using BPNN and SVM.A better predicted rate was obtained by BPNN than SVM.

关 键 词： 番茄种子 电子鼻 神经网络 支持向量机

领　　域： [自动化与计算机技术] [自动化与计算机技术]