首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于高光谱成像技术的菜心铜胁迫快速检测
引用本文:黄富荣,周泳欣,郑海平,陈嘉泽,熊征,刘霓红.基于高光谱成像技术的菜心铜胁迫快速检测[J].现代农业装备,2022,43(1):17-23.
作者姓名:黄富荣  周泳欣  郑海平  陈嘉泽  熊征  刘霓红
作者单位:暨南大学理工学院光电工程系,广东 广州 510632,广东省现代农业装备研究所,广东 广州 510630
基金项目:广州市重点领域研发计划;社会发展科技项目;广州市院士工作站项目
摘    要:重金属污染已经成为影响农作物生长的重要环境因素,而铜是我国农业三大重金属污染物之一.农作物受到铜胁迫后,生长发育会受到极大的影响,如何实现对农作物铜胁迫的检测对于农业生产有着极大的意义.传统检测农作物铜胁迫的方法存在分析时间长、准确性不高、易对农作物造成损害等问题.本文提出一种基于高光谱成像技术的农作物铜胁迫的快速检测...

关 键 词:高光谱成像  铜胁迫  偏最小二乘判别分析  支持向量机

Rapid Detection of Copper Stress in Flowering Chinese Cabbage based on Hyperspectral Imaging Technology
Huang Furong,Zhou Yongxin,Zheng Haiping,Chen Jiaze,Xiong Zheng,Liu Nihong.Rapid Detection of Copper Stress in Flowering Chinese Cabbage based on Hyperspectral Imaging Technology[J].Modern Agricultural Equipments,2022,43(1):17-23.
Authors:Huang Furong  Zhou Yongxin  Zheng Haiping  Chen Jiaze  Xiong Zheng  Liu Nihong
Institution:(Opto-electronic Department of Jinan University,Guangzhou 510632,China;Guangdong Institute of Modern Agricultural Equipment,Guangzhou 510630,China)
Abstract:Heavy metal pollution has become an important environmental factor affecting crop growth,and copper is one of the three major heavy metal pollutants in agriculture in China.The growth and development of crops will be greatly affected by copper stress.How to realize the detection of copper stress on crops is of great significance for agricultural production.Traditional methods for detecting copper stress in crops have problems such as long analysis time,low accuracy and easy to cause damage to crops.Therefore,we need to find a fast and non-destructive detection method.This paper presents a rapid detection method of copper stress in crops based on hyperspectral imaging technology.The healthy Chinese cabbage seedlings were divided into 5 groups with 8 strains in each group.Five Hoagland nutrient solutions containing 0.08(normal concentration),8,16,24 and 32 mg/L CuSO4 were used for cultivation respectively.Hyperspectral images were collected at 1,3,5 and 7 days after cultivation.The spectral data were preprocessed by first-order differential(FD),multiple scattering correction(MSC)and standard normal variate(SNV).Partial least squares discriminant analysis(PLSDA)and support vector machine(SVM)were used for modeling analysis.The experimental results show that the SVM model is more effective and the detection accuracy can reach 86.11%when the 4-day total samples are modeled and analyzed.When the samples collected every day were separately modeled and analyzed,PLSDA showed better stability,with an average detection accuracy of 95%,and a better detection effect could be achieved one day after the crops were under copper stress.In summary,the application of hyperspectral imaging technology to achieve rapid detection of copper stress in crops is feasible,which has a certain reference value for the development of precision agriculture in China.
Keywords:hyperspectral imaging  heavy metal stress  PLS-DA  SVM
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号