基于GA-BP算法的太赫兹波鉴定苜蓿草品种  

Identification of alfalfa varieties by terahertz wave based on GA-BP method

作　　者：王芳[1] 张玉 张春红 夏红岩 WANG Fang;ZHANG Yu;ZHANG Chunhong;XIA Hongyan(College of Science,China University of Petroleum,Beijing 102249,China;Inner Mongolia Grassland Station,Huhehot 010020,China)

机构地区：[1]中国石油大学(北京)理学院,北京102249 [2]内蒙古自治区草原工作站,内蒙古自治区呼和浩特010020

出　　处：《河北大学学报（自然科学版）》2021年第6期666-671,共6页Journal of Hebei University(Natural Science Edition)

基　　金：内蒙古自治区科技计划项目(201602054)。

摘　　要：基于太赫兹时域光谱技术(THz-TDS)测得8类苜蓿草种的折射率光谱数据,利用遗传算法优化BP神经网络(GA-BP)模型对苜蓿草种的分类进行了研究.结果表明GA-BP模型对8个苜蓿品种的平均分类准确率达到94%,且对单个品种的分类准确率最高能达到94.6%.本研究为牧草品种的分类鉴别提供了一种新方法,对种质资源的鉴别也具有重要的参考价值.Based on the refractive index data of alfalfa species measured by the terahertz time domain spectroscopy(THz-TDS),the genetic algorithm optimized BP neural network(GA-BP)model was used to identificate alfalfa species.The results show that the average classification accuracy rate of 8 varieties is 94%,and the maxial classification accuracy rate of a single variety is 94.6%.A new method is provided for the classification and identification of alfalfa species,which provides important reference value for the identification of germplasm resources.

关 键 词：THZ-TDS GA-BP 苜蓿草种 分类 

分 类 号：O433.4[机械工程—光学工程]