光谱学与光谱分析, 2019, 39 (10): 3040, 网络出版: 2019-11-05  

一种针对作物生育期光谱迁移的修正植被指数

A Modified Vegetationindex for Spectral Migration During Crop Growth
作者单位
1 中国农业大学现代精细农业系统集成研究教育部重点实验室, 北京 100083
2 北京工商大学, 食品安全大数据技术北京市重点实验室, 北京 100048
摘要
针对基于固定特征波长的植被指数不能适用于多个生育期叶绿素含量的诊断这一问题, 研究优化提出一种基于双波长计算光谱覆盖面积的叶绿素诊断植被指数, 用于稳健地诊断多生育期的营养。 以拔节期、 孕穗期和扬花期的冬小麦为研究对象, 采集其325~1 075 nm范围的冠层反射光谱, 测定采样样本的叶绿素含量。 采用小波去噪和多元散射校正算法对光谱数据进行预处理。 通过相关性分析, 确定生育期特征波长的迁移范围, 进而提出了基于光谱覆盖面积的冬小麦叶绿素含量光谱诊断参数(modified normalized area over reflectance curve, MNAOC)。 以信噪比(SNR)和平滑度指标(S)进行综合评价, 小波去噪函数的最佳参数为(“sqtwolog”, “mln”, “3”, “db5”)。 相关性分析结果表明, 生育期特征波段的迁移范围为(700 nm, 723 nm)。 在分析MNAOC指数对叶绿素含量诊断分辨率的基础上, 以0.5 mg·L-1的分辨率建立一元线性回归模型的结果为: 拔节期R2c=0.840 1, R2v=0.823 7; 孕穗期R2c=0.865 5, R2v=0.817 4; 扬花期R2c=0.833 8, R2v=0.807 6。 与ratio vegetation index(RVI)等5种双波长植被指数对比表明, 由于700和723 nm计算的光谱面积包含了由于生育期导致的光谱动态迁移特征, 使得MNAOC指数在模型精度上和多个生育期的普适性上, 都优于其他双波长代数运算植被指数, 为大田环境冬小麦生育期叶绿素含量诊断提供支持。
Abstract
The vegetation indices based on fixed characteristic wavelengths cannot be applied to the diagnosis of chlorophyll content across multiple growth stages. To solve this issue, this study proposed a diagnostic parameter based on spectral coverage area, which can be applied in multiple growth stages. The canopy reflectance spectra of 325~1 075 nm and leaf samples were collected at jointing stage, booting stage and flowering stage. The spectral were pretreated by wavelet denoising and multiple scattering correction (MSC) method and the chlorophyll content was measured by spectrophotometry. The migration range of characteristic wavelengths across different growth stages was determined by correlation analysis and a spectral parameter, named Modified Normalized Area Over reflectance Curve (MNAOC), was proposed based on the migration range coverage area. Firstly, the orthogonal experiment of wavelet parameters was designed for selecting the optimal parameters combination of wavelet basis function, decomposition layer number, threshold selection rule and threshold adjustment scheme. By the comprehensive evaluation of the SNR and S, the best parameter set was (“sqtwolog”, “mln”, “3”, “db5”). Then, correlation analysis showed that the migration range was (700 nm, 723 nm) within the characteristic wavelengths across different growth stages. After the resolution analysis, linear regression models were established for chlorophyll content diagnosis by the MNAOC with the concentration of 0.5 mg·L-1. Among them, R2c of the models were 0.840 1, 0.865 5 and 0.833 8 for each stage respectively, and R2v of the models were 0.823 7, 0.817 4 and 0.807 6 for each stage respectively. Finally, compared with the dual-wavelength based vegetation indices, the applicability advantage of MNAOC across multiple growth stages was verified. The comparison showed that MNAOC calculated by 700 and 723 nm, which contained the spectral dynamic migration characteristics, was superior to other dual-wavelength based vegetation indices, such as Ratio Vegetation Index (RVI) and Normalized Difference Vegetation Index (NDVI), in terms of model accuracy and universality in multiple growth periods. The results provided support for diagnosing chlorophyll content during the growth of winter wheat in field environment.
参考文献

[1] Shardendu Kumar Singh, James H Houx, Michael J W, et al. Field Crops Research, 2017, 209: 47.

[2] Li Dong, Cheng Tao, Zhou Kai, et al. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 129: 103.

[3] Tan Kezhu, Wang Shuwen, Song Yuzhu, et al. Chemometrics and Intelligent Laboratory Systems, 2018, 172: 68.

[4] Cao Q, Miao Y, Li F, et al. Precision Agriculture, 2016, 18(1): 1.

[5] Zhao B, Duan A W, Syed T A, et al. European Journal of Agronomy, 2018, 94: 113.

[6] Li Fei, Bodo Mistele, Hu Yuncai, et al. European Journal of Agronomy, 2014, 52: 198.

[7] Li F, Miao Y X, Feng G H, et al. Field Crops Research, 2014, 157: 111.

[8] Gnyp M L, Miao Y X, Yuan F, et al. Field Crops Research, 2014, 155: 42.

[9] Asa G, Jan M, Veronika K M, et al. Forests, 2016, 7(10): 226.

[10] Jesús Delegido, Luis Alonso, Gonzalo González, et al. International Journal of Applied Earth Observation and Geoinformation, 2010, 12(3): 165.

[11] XU Xin-gang, ZHAO Chun-jiang, WANG Ji-hua, et al(徐新刚, 赵春江, 王纪华, 等). Journal of Infrared and Millimeter Waves(红外与毫米波学报), 2013, 32(4): 351.

[12] Wang Junjie, Wang Tiejun, Shi Tiezhu, et al. Remote Sensing, 2015, 7(11): 15340.

刘豪杰, 李民赞, 张俊逸, 高德华, 孙红, 吴静珠. 一种针对作物生育期光谱迁移的修正植被指数[J]. 光谱学与光谱分析, 2019, 39(10): 3040. LIU Hao-jie, LI Min-zan, ZHANG Jun-yi, GAO De-hua, SUN Hong, WU Jing-zhu. A Modified Vegetationindex for Spectral Migration During Crop Growth[J]. Spectroscopy and Spectral Analysis, 2019, 39(10): 3040.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!