基于支持向量机的云烟87烟叶成熟度高光谱遥感识别

利用烟叶光谱反射率数据研究烟叶的成熟度是将高光谱遥感探测技术应用于烟叶成熟采收的前提。对生理成熟期内云烟87的中部叶成熟度进行了研究,研究是从提取和分析烟叶高光谱信息的角度上进行的。基于提取到的反映烟叶成熟度的光谱特征和支持向量机这一基于数据的小样本学习方法,建立了识别烟叶成熟度的数学模型,这个数学模型是由支持向量机生成的2个最优决策超平面,一个将初熟类烟叶和适熟类烟叶分开,另一个将适熟类烟叶和过熟类烟叶分开,且分类精度均在90%以上。...

Full description

Saved in:
Bibliographic Details
Published in西南农业学报 Vol. 26; no. 3; pp. 957 - 962
Main Author 梁寅 张云伟 李军营
Format Journal Article
LanguageChinese
Published 昆明理工大学现代农业工程学院,云南昆明,650500%云南省烟草农业科学研究院,云南昆明,650500 2013
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:利用烟叶光谱反射率数据研究烟叶的成熟度是将高光谱遥感探测技术应用于烟叶成熟采收的前提。对生理成熟期内云烟87的中部叶成熟度进行了研究,研究是从提取和分析烟叶高光谱信息的角度上进行的。基于提取到的反映烟叶成熟度的光谱特征和支持向量机这一基于数据的小样本学习方法,建立了识别烟叶成熟度的数学模型,这个数学模型是由支持向量机生成的2个最优决策超平面,一个将初熟类烟叶和适熟类烟叶分开,另一个将适熟类烟叶和过熟类烟叶分开,且分类精度均在90%以上。
Bibliography:LIANG Yin1,ZHANG Yun-wei1,LI Jun-ying2(1.Faculty of Modern Agricultural Engineering,Kunming University of Science and Technology,Yunnan Kunming 650500,China;2.Yunnan Academy of Tobacco Agricultural Sciences,Yunnan Kunming 650500,China)
Using spectral reflectance data to research the maturity of tobacco leaf was the premise of the application of hyperspectral remote sensing detection technology on tobacco harvest.The perspectives of extraction and analysis of spectral information was researched on the maturity of tobacco in this paper.Based on the spectral feature and support vector machine,the mathematical model for recognizing the maturity of tobacco had been built.This model contained two optimal decision hyperplanes.One was for initial maturity category and proper maturity category,the other was for proper maturity category and over maturity category.The accuracy of the classifier was more than 90 %.
51-1213/S
Maturity; Spectral feature; Support vector machine; Optimal decision hyperplane; Classification accu
ISSN:1001-4829