Spectroscopy and Spectral Analysis, Volume. 42, Issue 4, 1076(2022)

Research on Classification Method of Main Poisonous Plants in Alpine Meadow Based on Spectral Characteristic Variables

Rui DONG... Zhuang-sheng TANG, Rui HUA, Xin-cheng CAI, Dar-han BAO, Bin CHU, Yuan-yuan HAO and Li-min HUA*; |Show fewer author(s)
Author Affiliations
  • Grassland College of Gansu Agricultural University, Key Laboratory of Grassland Ecosystem Ministry of Education, Engineering and Technology Research Center for Alpine Rodent Pest Control, National Forestry and Grassland Administration, Lanzhou 730070, China
  • show less
    Figures & Tables(8)
    Average spectral curves of 11 different alpine meadow poisonous plants
    The average first-order differential spectrum curve of 11 different alpine meadow poisonous plants
    Kendall correlation matrixNote: The graph is divided into two parts, the upper triangle is the significance test, the lower triangle is the correlation coefficient, the asterisk in the figure represents the significance test, * means the difference is significant, no * means the difference is not significant, *p≤ 0.01
    Sorting of 16 spectral feature variables(a): The cumulative percentage of typical variables; (b): The standardized score coefficient
    Classification accuracy of 16 spectral variables in 5 algorithms
    • Table 1. 11 poisonous plants and their characteristics

      View table
      View in Article

      Table 1. 11 poisonous plants and their characteristics

      毒草名毒草化学成分部位类型代表
      黄花棘豆(Oxytropis ochrocephala)吲哚兹定生物碱全株有毒OO
      宽苞棘豆(O latibracteata)吲哚兹定生物碱全株有毒OL
      多枝黄芪(Astragalus polycladus)吲哚兹定生物碱全株有毒AP
      长毛风毛菊(Saussurea hieracioides)倍半萜全草有毒SH
      黄帚橐吾(Ligularia virgaurea)倍半萜全草有毒LV
      乳白香青(Anaphalis lactea)倍半萜全草有毒AL
      葵花大蓟(Cirsium souliei)倍半萜全草有毒CS
      瑞香狼毒(Stellera chamaejasme)黄酮类、 萜类、 本质素类全株有毒SC
      密花香薷(Elsholtzia densa)黄酮类化合物全草有毒ED
      露蕊乌头(Aconitum gymnandrum)生物碱全草有毒, 块根剧毒AG
      碎米蕨叶马先蒿(Pedicularis cheilanrthifolia)甾体或三萜类、 生物碱、 黄酮、 强心苷全草有毒PC
    • Table 2. Definition of spectral characteristic variables

      View table
      View in Article

      Table 2. Definition of spectral characteristic variables

      参数类型参数符号定义
      位置变量红边幅值Mre在红边680~760 nm内一阶微分最大值
      红边位置Lre红边幅值对应波长
      红谷幅值Mr红光范围内640~680 nm最大反射率
      红谷位置Lr红光范围内640~680 nm红谷对应波长
      绿峰幅值Mg绿光范围内510~560 nm最大反射率
      绿峰位置Lg绿光范围内510~560 nm绿峰对应的波长
      蓝边幅值Mb在蓝边490~530 nm内一阶微分最大值
      蓝边位置Lb蓝边幅值对应的波长
      黄边幅值My在黄边560~640 nm内一阶微分最大值
      黄边位置Ly黄边幅值对应的波长
      面积变量红边面积Are红边范围内一阶微分值总和
      蓝边面积Ab蓝边范围内一阶微分值总和
      植被指数变量Mg/MrRVI1绿峰与红谷幅值比值
      Are/AbRVI2红边与蓝边面积比值
      (Mg-Mr)/(Mg+Mr)NDVI1绿峰与红谷幅值归一化比值
      (Are-Ab)/(Are+Ab)NDVI2红边与蓝边面积归一化比值
    • Table 3. Classification accuracy based on typical discriminant analysis

      View table
      View in Article

      Table 3. Classification accuracy based on typical discriminant analysis

      毒草种类APOOLVOLSCAGEDALPCSHCS样本生产者精度/%
      AP5050100
      OO3513115070
      LV438625076
      OL4465092
      SC5050100
      AG41455090
      ED21454893.75
      AL5050100
      PC4915098
      SH1495098
      CS1495098
      分类总和5639396150475150515149
      用户精度/%89.2889.7497.4375.4110095.7488.2410096.0896.08100
    Tools

    Get Citation

    Copy Citation Text

    Rui DONG, Zhuang-sheng TANG, Rui HUA, Xin-cheng CAI, Dar-han BAO, Bin CHU, Yuan-yuan HAO, Li-min HUA. Research on Classification Method of Main Poisonous Plants in Alpine Meadow Based on Spectral Characteristic Variables[J]. Spectroscopy and Spectral Analysis, 2022, 42(4): 1076

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Research Articles

    Received: May. 22, 2021

    Accepted: --

    Published Online: Jul. 25, 2023

    The Author Email:

    DOI:10.3964/j.issn.1000-0593(2022)04-1076-07

    Topics