全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

可见/近红外光谱预测杨梅汁酸度的方法研究

Keywords: 可见/近红外光谱偏最小二乘杨梅汁酸度人工神经网络

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对可见/近红外光与杨梅汁酸度存在非线性相关的特点,提出了应用偏最小二乘(PLS)法预测线性部分和人工神经网络(ANN)预测非线性部分,结合两种方法综合预测杨梅汁酸度值,通过比较r,RMSEP,Bias的值来检验该方法.其中PLS模型用于寻找与杨梅汁酸度值有关的敏感波段,预测杨梅汁酸度的线性部分,将这些敏感波段对应的光谱吸光度值作为人工神经网络的输入,并将杨梅汁酸度的实际测量值减去PLS模型校正值,获得的差额部分作为神经网络的输出,建立一个差额神经网络预测杨梅汁酸度的非线性部分.46个样本用于建模,30个样本用于预测.结果表明该方法对样本的预测相关系数r=0.939,RMSEP=0.218,Bias=-0.121,好于只使用PLS模型的相关系数r=0.921,RMSEP=0.228,Bias=-0.132.

References

[1]  HE Yong,LI Xiao-Li,SHAO Yong-Ni.Quantitative analysis of the varieties of apple using near infrared spectroscopy by principal component analysis and BP model[J].Lecture Notes in Artificial Intelligence,2005,3809:1053-1056.
[2]  Bennedsen B S,Peterson D L.Performance of a system for apple surface defect identification in near-infrared images[J].Biosystems Engineering,2005,90(4):419-431.
[3]  HE Yong,LI Xiao-Li.Discrimination of varieties of waxberry using near infrared spectra[J].J.Infrared Millim.Waves(何勇,李晓丽.近红外光谱杨梅品种鉴别方法的研究.红外与毫米波学报),2006,25(3):192-194.
[4]  LIN San-Hu,ZHU Hong,ZHAO Yi-Gong.Model for sea clutter based on neural network[J].J.Infrared Millim.Waves(林三虎,朱红,赵亦工.基于神经网络的海杂波模型.红外与毫米波学报),2004,23(1):55-58.
[5]  HE Yong,SONG Hai-Yan,Pereira A G,et al.A new approach to predict N,P,K and OM content in a loamy mixed soil by using near infrared reflectance spectroscopy[J].Lecture Notes in Computer Science,2005,3644:859-867.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133