基于变量选择的蚕茧茧层量可见-近红外光谱无损检测

以蚕茧茧层量为研究对象,研究了基于可见-近红外光谱技术的蚕茧茧层量无损检测方法。采用最小二乘支持向量机(least square-support vector machine,LS-SVM)建立可见-近红外光谱模型。采用无信息变量消除算法(uninformative variable elimination,UVE)与连续投影算法(successive projections algorithm,SPA)相结合选取光谱有效波长。结果表明,基于UVE-SPA法进行变量选择,最终将原始光谱的600个光谱变量减少到了8个(673,937,963,982,989,992,995和1008nm)。基于此...

Full description

Saved in:
Bibliographic Details
Published inNong ye gong cheng xue bao Vol. 26; no. 2; pp. 231 - 236
Main Author 黄凌霞 吴迪 金航峰 赵丽华 何勇 金佩华 楼程富
Format Journal Article
LanguageChinese
Published 浙江大学动物科学学院,杭州,310029%浙江大学生物系统工程与食品科学学院,杭州,310029%浙江省湖州市农业科学研究院,湖州,313000 2010
Subjects
Online AccessGet full text
ISSN1002-6819
DOI10.3969/j.issn.1002-6819.2010.02.040

Cover

More Information
Summary:以蚕茧茧层量为研究对象,研究了基于可见-近红外光谱技术的蚕茧茧层量无损检测方法。采用最小二乘支持向量机(least square-support vector machine,LS-SVM)建立可见-近红外光谱模型。采用无信息变量消除算法(uninformative variable elimination,UVE)与连续投影算法(successive projections algorithm,SPA)相结合选取光谱有效波长。结果表明,基于UVE-SPA法进行变量选择,最终将原始光谱的600个光谱变量减少到了8个(673,937,963,982,989,992,995和1008nm)。基于此8个变量建立的LS-SVM模型得到了预测集的确定系数(Rp^2)为0.5354,误差均方根(RMSEP)为0.0373的预测结果。表明可见-近红外光谱可以用于对蚕茧的茧层量进行无损检测,同时UVE-SPA是一种有效的光谱变量选择方法。
Bibliography:nondestructive examination
model analysis
11-2047/S
S886.3
near infrared spectroscopy
cocoon
shell weight
uninformative variable elimination(UVE)
near infrared spectroscopy; nondestructive examination; model analysis; cocoon; shell weight; uninformative variable elimination(UVE); successive projections algorithm(SPA)
successive projections algorithm(SPA)
O657.3
ISSN:1002-6819
DOI:10.3969/j.issn.1002-6819.2010.02.040