基于光谱特征变量的高寒草甸主要毒草分类方法研究

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

摘要
参考文献
相关文章 (15)

全文: PDF (2614 KB)
输出: BibTeX | EndNote (RIS)

摘要：高寒草甸毒草化是青藏高原草地生态系统面临的主要问题之一。高寒草甸毒草分类技术对草地群落的变化具有及时监测和科学防控的重要意义。近年来，毒草种类及危害面积急剧增加，传统人工实地调查耗时费力、调查结果代表性差。同时毒草在地域分布上具有一定的差异性，依靠人力难以实现大面积调查。高光谱遥感技术凭借分辨率高、波段多、图谱合一等特点，在毒草精细分类中表现出巨大的优势，可满足快速、准确、大尺度获取毒草发生面积的需求。已有学者对草地植物的光谱反射特征开展了研究，证明采用植物光谱反射特征可有效区分其种类。但是，目前尚缺乏针对草地有毒植物光谱特征变量的筛选及构建基于毒草光谱特征的预测分类模型。本研究利用SOC710VP近红外高光谱成像仪，在甘肃省天祝县和玛曲县境内高寒草甸上采集黄花棘豆(Oxytropis ochrocephala)、宽苞棘豆(O latibracteata)、多枝黄芪(Astragalus polycladus)、长毛风毛菊(Saussurea hieracioides)、黄帚橐吾(Ligularia virgaurea)、乳白香青(Anaphalis lactea)、葵花大蓟(Cirsium souliei)、瑞香狼毒(Stellera chamaejasme)、密花香薷(Elsholtzia densa)、露蕊乌头(Aconitum gymnandrum)、碎米蕨叶马先蒿(Pedicularis cheilanrthifolia)11种主要毒草野外光谱数据，采用Savitzky-Golay卷积平滑算法（SG）对原始光谱值进行去噪，使用一阶微分导数（FD）开展光谱特征分析，利用典型判别分析（CDA）对选用的16种光谱特征变量标准化得分系数绝对值进行排序，然后从大到小分别添加到随机森林（RF）、支持向量机-径向核函数（SVM-RBF）、k最邻近分类（KNN）、朴素贝叶斯（NB）、决策树（CART）5种算法中构建分类模型并筛选最佳特征变量，使用混淆矩阵评价分类效果。结果表明：（1）16个光谱特征变量典型判别分析（CDA）总体分类精度为92.34%，R²=0.89；（2）筛选出最佳分类光谱特征变量为绿峰幅值（Mg）、蓝边面积（Ab）、红边幅值（Mre）、红边面积（Are）、红边位置（Lre）、NDVI2、RVI1；（3）将筛选出的7个光谱特征变量用于毒草分类，结果支持向量机-径向核函数（SVM-RBF）分类效果最好，精度达96.45%。

关键词：高寒草甸；毒草；光谱特征；典型判别；分类

Abstract：The extension of poisonous plants in alpine meadows is one of the main problems of the grassland ecosystem in the Qinghai-Tibet Plateau. The classification technology of poisonous plants in alpine meadows is of great significance for timely monitoring, scientific preventing and controlling changes in grassland communities. In recent years, poisonous plants species and harmful areas have increased rapidly. Traditional manual field surveys were time-consuming and laborious, and poorly represented the survey results. At the same time, poisonous plants have certain differences in geographical distribution, so it is not easy to conduct large-scale investigations by the workforce. Hyperspectral remote sensing technology has great advantages in the fine classification of poisonous plants due to its high resolution, multiple bands, integration of maps, and so on, which can meet the needs of fast, accurate, and large-scale acquisition of poisonous plants. Some scholars have carried out studies on the spectral reflectance characteristics of grassland plants, which proved that the spectral reflectance characteristics of plants could effectively distinguish their species. On the contrary, there are few reports on the selection of spectral reflectance characteristics variables of poisonous plants and the construction of a predictive classification model based on the spectral characteristics of poisonous plants. In this study, 11 kinds of main poisonous plants field spectrum data on alpine meadows, including Oxytropis ochrocephala, O latibracteata, Astragalus polycladus, Saussurea hieracioides, Ligularia virgaurea, Anaphalis lactea, Cirsium souliei, Stellera chamaejasme, Elsholtzia Densa, Aconitum gymnandrum, and Pedicularis cheilanrthifolia (in Tianzhu County and Maqu County, Gansu Province) were collect by using the SOC710VP near-infrared hyperspectral imager. The Savitzky-Golay convolution smoothing algorithm (SG) was applied to denoise the original spectral values, the first-order differential derivative (FD) was used to carry out spectral feature analysis, and the canonical discriminant analysis (CDA) was performed to sort the absolute values of the standardized score coefficients of 16 selected spectral feature variables. Then from the size of large to small, they were added to 5 algorithms, namely random forest (RF), support vector machine-radial kernel function (SVM-RBF), k-nearest neighbor classification (KNN), naive bayes (NB), and decision tree (CART) to construct classification models and screen the best feature variables, and the confusion matrix was used to evaluate the classification effects. The results showed that: (1) The overall classification accuracy of canonical discriminant analysis (CDA) for 16 spectral characteristic variables was 92.34%, R²=0.89; (2) The best classification spectral characteristic variables were selected as green peak amplitude (Mg), blue edge area (Ab), red edge amplitude (Mre), red edge area (Are), red edge position (Lre), NDVI2, and RVI1; (3) The selected 7 spectral characteristic variables were used to classify poisonous plants, and then the SVM-RBF has the best classification effects, with an accuracy of 96.45%.

Key words：Alpine meadow; Poisonous plants; Spectral characteristics; Canonical discrimination; Classification

收稿日期: 2021-05-22 修订日期: 2021-08-11

中图分类号:

O433.4

基金资助: 甘肃农业大学公招博士科研启动基金项目（GAU-KYQD-2018-09），国家自然科学基金项目（32001385），甘肃省重点研发计划项目（17YF1NA059），甘肃省科技厅自然科学基金项目(21JR7RA810)，甘肃省教育厅优秀研究生“创新之星”项目(2021CXZX-343)资助

通讯作者: 花立民 E-mail: hualm@gsau.edu.cn

作者简介: 董瑞，1995年生，甘肃农业大学草业学院硕士研究生 e-mail: dongrui_gsau@163.com

引用本文:

董瑞，唐庄生，花蕊，蔡新成，包达尔罕，楚彬，郝媛媛，花立民. 基于光谱特征变量的高寒草甸主要毒草分类方法研究[J]. 光谱学与光谱分析, 2022, 42(04): 1076-1082.
DONG Rui, TANG Zhuang-sheng, HUA Rui, CAI Xin-cheng, BAO Dar-han, CHU Bin, HAO Yuan-yuan, HUA Li-min. Research on Classification Method of Main Poisonous Plants in Alpine Meadow Based on Spectral Characteristic Variables. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1076-1082.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2022)04-1076-07 或 https://www.gpxygpfx.com/CN/Y2022/V42/I04/1076

[1] Yao X X, Wu J P, Gong X Y, et al. Ecological Engineering, 2019, 130: 80.
[2] ZHAO Bao-yu, LIU Zhong-yan, WAN Xue-pan, et al（赵宝玉，刘忠艳，万学攀，等）. Scientia Agriculture Sinica（中国农业科学）, 2008, 41(10): 3094.
[3] HUANG Mei, SHANG Zhan-huan（黄梅，尚占环）. Acta Agrestia Sinica（草地学报）, 2019, 27(5): 1107.
[4] ZHAO Shi-jiao, ZHAO Hong-yang, GAO Dan, et al（赵世姣，赵红阳，高丹，等）. Acta Agrestia Sinica（草地学报）, 2017, 25(6): 1389.
[5] ZHU Ning, WANG Hao, NING Xiao-gang, et al（朱宁，王浩，宁晓刚，等）. Science of Surveying and Mapping（测绘科学）, 2020, 46(5): 66.
[6] YU Lu, WANG Xun, CHAI Sha-tuo, et al（于璐，王迅，柴沙驼，等）. Acta Agrestia Sinica（草地学报）, 2020, 28(2): 547.
[7] Baron J, Hill D J. Remote Sensing of Environment, 2020, 249: 112008.
[8] HAN Wan-qiang, JIN Gui-li, YUE Yong-huan, et al（韩万强，靳瑰丽，岳永寰，等）. Acta Agrestia Sinica（草地学报）, 2020, 28(4): 1153.
[9] MA Wen-qiang, ZHANG Man, LI Yuan, et al（马文强，张漫，李源，等）. Chinese Journal of Analytical Chemistry（分析化学）, 2020, 48(12): 1737.
[10] WEI Ya-hui, ZHAO Bao-yu, WEI Shuo-nan, et al（尉亚辉，赵宝玉，魏朔南，等）. Major Species and Distribution of Poisonous Plants in Natural Grasslands of Western China(中国西部天然草地毒害草主要种类及分布). Beijing: Science Press（北京：科学出版社），2018. 11.
[11] ZENG Shuai, KUANG Run-yuan, XIAO Yang, et al（曾帅，况润元，肖阳，等）. Remote Sensing Information（遥感信息）, 2017, 32(5): 75.
[12] YI Da, LI Yuan-qiao, YANG Yu（易达，李院瞧，杨宇）. Software（软件）, 2020, 41(7): 45.
[13] Shi X, Song J, Wang H, et al. European Journal of Remote Sensing, 2020, (4): 1.