摘要
本文针对水果内部品质评定分类存在主观性强和一致性差的特点,提出了一种新的鸭梨褐变识别分类方法,该方法是利用近红外光谱仪获取正常鸭梨和褐变鸭梨近红外光谱并对其进行分析的基础上,应用支持向量机(SVM)算法的识别原理建立正常鸭梨和褐变鸭梨的分类识别模型。在多项式核函数下对试验样品的识别准确率为95%。研究结果表明NIR-SVM用于对于鸭梨褐变病果的无损检测识别是可行的。
Aiming at the deficiencies of fruit classification by internal quality evaluation such as result subjectivityandpoor consistency,a new method of yali with brown identification was proposed which is base on pattern recongnition theory of suppport vector machine(SVM)and uses near infrared spectra analyzer to acquire and analysis of spectroscopic curves of 2 kinds of yali .The accuracy rate of the indentification of yali with brown heart is up to 95%.The research result shows the estability of establishing ...
出处
《农业网络信息》
2008年第3期133-135,共3页
Agriculture Network Information
关键词
近红外光谱
支持向量机
鸭梨
褐变
Near infrared spectra
Support vector machine
YALI
Brown heart
