铜污染下玉米叶片重金属含量反演模型  

Inversion Model of Heavy Metal Content in Maize Leaves under Cu^(2+)Stress

作　　者：郭辉[1] 石海 张全旺 GUO Hui;SHI Hai;ZHANG Quanwang(School of Geomatics,Anhui University of Science and Technology,Huainan Anhui 232001,China)

机构地区：[1]安徽理工大学空间信息与测绘工程学院,安徽淮南232001

出　　处：《安徽理工大学学报（自然科学版）》2023年第3期40-46,共7页Journal of Anhui University of Science and Technology:Natural Science

基　　金：安徽省高校自然科学基金资助项目(KJ2018A0070)。

摘　　要：针对不同浓度Cu^(2+)污染胁迫下玉米叶片光谱差异微弱、难以有效区分其受Cu^(2+)污染程度的问题,通过采集盆栽玉米叶片光谱反射率与叶片Cu^(2+)含量数据,构建了叶片光谱共振峰峰值(FPV)与共振峰最大系数(FMC)两个特征参数,分析了特征参数与叶片Cu^(2+)含量的相关性,建立了FPV、FMC与叶片Cu^(2+)含量的回归模型。结果表明,共振峰体现了叶片光谱不同波段与细胞内部结构的变化,包络谱反映出信号能量的大小;定义的两个共振峰特征参数均与玉米叶片Cu^(2+)含量呈正相关;基于FPV与FMC建立的反演模型中,FMC反演模型验证精度最高,R^(2)为0.93,RMSE为3.54μg/g。将FMC作为监测玉米Cu^(2+)污染程度的最优指标,可为作物重金属污染监测提供借鉴和参考。Under different concentrations of Cu^(2+)pollution stress,the spectral difference of maize leaves is weak and difficult to effectively distinguish the degree of Cu^(2+)pollution.By collecting leaf spectral reflectance and leaf Cu^(2+)content data of potted maize,two characteristic parameters of leaf spectral formant peak value(FPV)and formant maximum coefficient(FMC)were constructed.The correlation between characteristic parameters and leaf Cu^(2+)content was analyzed,and a regression model of FPV,FMC and leaf Cu^(2+)content was established.The results showed that the formant reflected the change of different spectral bands of leaves and the internal structure of cells,and the envelope spectrum reflected the signal energy.The two characteristic parameters of formants were positively correlated with the content of Cu^(2+)in maize leaves.Among the inversion models established based on FPV and FMC,the validation accuracy of FMC inversion model was the highest,with R^(2)=0.93,RMSE=3.54μg/g.Therefore,FMC is able to be used as the optimal index to monitor the Cu^(2+)pollution degree of maize,providing a reference for monitoring the heavy metal pollution of crops.

关 键 词：高光谱 盆栽玉米 Cu^(2+)胁迫 希尔伯特-黄变换 共振峰特征 

分 类 号：X87[环境科学与工程—环境工程]