首页> 外文会议>Informationization, Automation and Electrification in Agriculture >Discrimination of Cocoon Sex Using Middle-Infrared Spectroscopy with the Aid of Chemometrics
【24h】

Discrimination of Cocoon Sex Using Middle-Infrared Spectroscopy with the Aid of Chemometrics

机译:借助化学计量学,使用中红外光谱区分茧性

获取原文
获取原文并翻译 | 示例

摘要

Study on method of discriminating cocoon sex on basis of middle-infrared spectroscopy.Cocoon was cut to identify cocoon sex through examining its pupa, and male and female sex was set to values of 1 and 0 respectively. It were acquired that the attenuated total reflectance spectra of cocoon samples from wavenumber 4 000 cm-1 to 600 cm-1. After one odd spectrum was rejected through principal component cluster analysis, twenty-seven samples were selected randomly as calibration set and the other 13 were as test set. It was analyzed that the influence of Savitzky-Golay convolution smooth (window points was 3, 5, 7, 9, 11 and 13) and different principle components (1-10) on prediction accuracy. Establish PLS model and artificial neural networks model. Results indicated that 100 percent prediction accuracy for both calibration and prediction sets could be achieved when 3 principle components and 3 window points for Savitzky-Golay convolution smooth were adopted in PLS model, or 2 principle components and 5 window points for Savitzky-Golay convolution smooth. It indicated that MIR spectroscopy can be used for prediction of cocoon sex.
机译:基于中红外光谱法鉴别茧性的方法的研究。将茧切去以通过检查其来鉴定茧性,将男女分别设置为1和0。测得茧样品从波数4000 cm-1到600 cm-1的衰减全反射光谱。通过主成分聚类分析剔除一个奇数光谱后,随机选择27个样品作为校准集,其他13个样品作为测试集。分析了Savitzky-Golay卷积平滑度(窗口点为3、5、7、9、11和13)和不同的主成分(1-10)对预测精度的影响。建立PLS模型和人工神经网络模型。结果表明,在PLS模型中采用Savitzky-Golay卷积平滑的3个主成分和3个窗点,或对于Savitzky-Golay卷积平滑采用2个主成分和5个窗点,则校准和预测集的预测精度都可以达到100% 。这表明MIR光谱可用于预测茧性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号