您好，欢迎访问宁夏农林科学院 机构知识库！
宁夏农林科学院机构知识库

全部

  • 全部
  • 标题
  • 学者
  • 机构
  • 关键词

Lycium species and variety recognition technology based on electrochemical sensing of leaf signals

文献类型： 外文期刊

作者： Shi, Xin 1 ; Man, Junjie 2 ; Ye, Weiting 2 ; Zhu, Jiangwei 3 ; Fu, Li 2 ; Zheng, Yuhong 4 ; Yin, Yue 5 ; Niu, Yan 1 ; Wang, Xiaojing 1 ;

作者机构： 1.Ningxia Inst Qual Stand & Testing Technol Agr Prod, Yinchuan 750002, Peoples R China

2.Hangzhou Dianzi Univ, Coll Mat & Environm Engn, Key Lab Novel Mat Sensor Zhejiang Prov, Hangzhou 310018, Peoples R China

3.Nanjing Forestry Univ, Coinnovat Ctr Sustainable Forestry Southern China, Nanjing 210037, Peoples R China

4.Jiangsu Prov & Chinese Acad Sci, Inst Bot, Nanjing Bot Garden Mem Sun Yaseen, Nanjing 210014, Peoples R China

5.Ningxia Acad Agr & Forestry Sci, Natl Wolfberry Engn Res Cente, Yinchuan 750002, Peoples R China

关键词： electrochemical fingerprint; goji berry; species identification; machine learning; phytochemistry

期刊名称：NOTULAE BOTANICAE HORTI AGROBOTANICI CLUJ-NAPOCA （ 影响因子：1.8； 五年影响因子：1.9 ）

ISSN： 0255-965X

年卷期： 2023 年 51 卷 1 期

页码：

收录情况： SCI

摘要： Identification of plant species and variety has important application value in the process of agricultural production. In this work, we try to use electrochemical fingerprinting technology to collect the electrochemical behavior of electrochemically active substances in plant leaf tissues. Twenty Lycium species and varieties were specifically selected to investigate the recognition ability of electrochemical fingerprinting. Two different extraction solvents and electrolytes were used to create different collection environments. The results show that different Lycium spp. can exhibit different electrochemical fingerprints. Different species of the same species exhibit relatively similar electrochemical fingerprints. After the second derivative processing, the electrochemical fingerprint of plants can be used for classification and recognition by different machine learning models. Partial least squares discriminant analysis (PLS-DA), k-nearest neighbor, (KNN), support vector machine (SVM), random forest (RF) and stochastic gradient boosting (SGB) were used to establish recognition model of Lycium spp. The results show that SGB has the best identification accuracy for electrochemical fingerprint after second derivative treatment.

  • 相关文献
作者其他论文 更多>>
置顶
购物车
反馈

© 京ICP备09089781号-29 主办单位：宁夏农林科学院

学术资源： 中国农业科学院 农业专业知识服务系统 农科联盟资源共建共享平台 中国农业科技文献与信息服务平台 中国农业期刊集成服务平台