SPA算法与机器学习的黄河源土壤水分反演

doi:10.3964/j.issn.1000-0593(2023)06-1961-07

摘要
参考文献
相关文章 (15)

全文: PDF (7310 KB)
输出: BibTeX | EndNote (RIS)

摘要：土壤水分在一定程度上决定着地区的生态承载力和土壤理化性质。准确、快速的获取土壤水分含量，对生态环境监测、土壤退化恢复等具有重要意义。高光谱遥感在土壤参数反演方面应用广泛，但对高寒草甸土壤的高光谱特征与参数反演研究有待深入。为建立适用于高寒脆弱生态系统的高寒草甸土壤含水量高光谱反演模型，以黄河源区的河南县为例，利用多元线性逐步回归(MLSR)、偏最小二乘回归(PLSR)和BP神经网络(BPNN)对土壤样本含水量与土壤光谱及其数学变换的特征波段进行建模，由决定系数(R²)、均方根误差(RMSE)和预测残差比(RPD)对模型精度进行验证。结果表明: (1) 在可见光-近红外波段，土壤样本的光谱反射率在710、780和950 nm附近存在强弱不同的水分吸收区间，且随着含水量的增加，反射率呈现先迅速降低，后缓慢增加的趋势。(2) 通过连续投影算法(SPA)筛选的光谱特征波段作为自变量，含水量作为因变量，分别通过MLSR和PLSR建立反演模型，其中一阶微分(FD)、对数一阶微分(FDL)变换对应的PLSR模型可实现高寒草甸土壤水分的粗略反演，且FD变换对应的PLSR模型精度较高。(3)BPNN反演模型中，除去包络(CR)建模外，其余模型R²均大于0.9，RMSE在0.048~0.074之间。其中FD、FDL、LG变换光谱对应的BPNN模型反演精度较高，验证结果的R²均大于0.8，RPD均大于2.5，精度最高的为对数(LG)变换对应的BPNN模型，R²、RMSE和RPD分别达到0.967、0.038和5.039。因此，BPNN模型能较精确的实现黄河源区高寒草甸土壤含水量的高光谱反演，为该地区乃至其他高寒地区生态环境监测与土壤恢复提供技术基础和数据支撑。

关键词：高光谱反演；高寒草甸；机器学习；SPA算法；黄河源

Abstract：Soil moisture determines a region’s ecological carrying capacity and soil physical and chemical properties to a certain extent. It is significant to obtain soil moisture content accurately and quickly for ecological environment monitoring and soil degradation restoration. Hyperspectral remote sensing is widely used in soil parameter inversion, but the research on hyperspectral characteristics and parameter inversion of alpine meadow soil needs further study. Consequently, to develop a hyperspectral inversion model of soil moisture content in alpine meadows applicable to fragile alpine ecosystems, 102 soil samples were collected from Henan County in the Yellow River source area. Multiple linear stepwise regression (MLSR), partial least squares regression (PLSR) and back propagation neural network (BPNN) methods were used to model the soil moisture content with the original spectrum and its mathematically transformed characteristic bands, and the inversion accuracy was verified by the coefficient of determination (R²), root mean square error (RMSE) and the residual ratio of prediction (RPD). The major findings are as follows: (1) In the visible-near infrared band, the spectral reflectance of soil samples has water absorption interval near 710, 780 and 950 nm, and the absorption intensity is different. The reflectance tends to decrease rapidly and increase slowly with increasing soil moisture content. (2) SPA algorithm was used to select the spectrum’s characteristic bands after S-G smoothing, four transformations as independent variables and water content as dependent variables. Then MLSR and PLSR were used to establish the inversion model. The PLSR model corresponding to the first-order differential (FD) and first-order logarithmic differential (FDL) transformations can achieve a rough inversion of soil moisture in alpine meadows, and the PLSR model corresponding to the FD transformation is accurate. (3) In the BPNN inversion models, except for the model corresponding to continue to remove (CR), the R² of other models is greater than 0.9, and RMSE is between 0.048 and 0.074. In all the models, the BPNN model corresponding to FD, FDL and LG transform is highly accurate, with R² and RPD greater than 0.8 and 2.5 respectively. The BPNN model corresponding to the LG transform has the highest accuracy, with R², RMSE and RPD up to 0.967, 0.038 and 5.039, respectively. Therefore, the BPNN model can achieve relatively accurate hyperspectral inversion of soil moisture content of alpine meadow in the source region of the Yellow River, which can provide the technical basis and data support for ecological environment monitoring and soil restoration in this region and even other alpine regions.

Key words：Hyperspectral inversion; Alpine Meadow; Machine Learning; SPA; Yellow River Source

收稿日期: 2022-03-22 修订日期: 2022-06-02

中图分类号:

TP79

基金资助: 国家自然科学基金项目(42161068)，青海省科技厅基金项目(2021-ZJ-743)，甘肃省祁连山生态环境研究中心开放基金项目(QLS202007)资助

通讯作者: 赵健赟 E-mail: zhaojianyun1981@163.com

作者简介: 姜传礼，1996年生，青海大学地质工程系硕士研究生 e-mail: jcl18755853214@163.com

引用本文:

姜传礼，赵健赟，丁圆圆，赵沁浩，马宏嫣. SPA算法与机器学习的黄河源土壤水分反演[J]. 光谱学与光谱分析, 2023, 43(06): 1961-1967.
JIANG Chuan-li, ZHAO Jian-yun, DING Yuan-yuan, ZHAO Qin-hao, MA Hong-yan. Study on Soil Water Retrieval Technology of Yellow River Source Based on SPA Algorithm and Machine Learning. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(06): 1961-1967.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2023)06-1961-07 或 https://www.gpxygpfx.com/CN/Y2023/V43/I06/1961

[1] HUANG Qian，DING Ming-jun，CHEN Li-wen, et al(黄倩，丁明军，陈利文, 等). Journal of Soil and Water Conservation(水土保持学报), 2022, 36(1): 189.
[2] Wanders N, Bierkens M F P, Jong S M D, et al. Water Resources Research, 2015, 50(8): 6874.
[3] Guillod B P, Orlowsky B, Miralles D G, et al. Nature Communications, 2015, 6: e6443.
[4] ZHAO Jian-yun，DING Yuan-yuan，DU Mei, et al(赵健赟，丁圆圆，杜梅, 等). Science Technology and Engineering(科学技术与工程), 2021, 21(24): 10209.
[5] HU Guang-yin，DONG Zhi-bao，LU Jun-feng, et al(胡光印，董治宝，逯军峰, 等). Acta Ecologica Sinica(生态学报), 2011, 31(14): 3872.
[6] LI Xin-xing，LIANG Bu-wen，BAI Xue-bing, et al(李鑫星，梁步稳，白雪冰, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2020, 40(12): 3705.
[7] Peng J, Loew A, Merlin O, et al. Reviews of Geophysics, 2017, 55(2): 341.
[8] LIU Huan-jun，ZHANG Bai，SONG Kai-shan, et al(刘焕军，张柏，宋开山, 等). Journal of University of Chinese Academy of Sciences(中国科学院大学学报), 2008，(4): 503.
[9] Hummel J W, Sudduth K A, Hollinger S E. Computers and Electronics in Agriculture, 2001, 32(2): 149.
[10] Wang W, Gao M, Wang J. Hyperspectral Parameters and Prediction Model of Soil Moisture in Apple Orchards: IOP Conference Series: Earth and Environmental Science, 2021, 687: 012085.
[11] TIAN Mei-ling，GE Xiang-yu，DING Jian-li, et al(田美玲，葛翔宇，丁建丽, 等). Laser & Optoelectronics Progress(激光与光电子学进展), 2020, 57(9): 232.
[12] ZHANG Jun-yong，DING Jian-li，TAN Jiao(张钧泳，丁建丽，谭娇). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2019, 50(3): 221.
[13] SHANG Tian-hao，JIA Ping-ping，SUN Yuan, et al(尚天浩, 贾萍萍, 孙媛, 等). Bulletin of Soil and Water Conservation(水土保持通报), 2020, 40(4): 183.
[14] LIU Huan-jun，WANG Xiang，ZHANG Xiao-kang, et al(刘焕军，王翔，张小康, 等). Chinese Journal of Soil Science(土壤通报), 2018, 49(1): 38.
[15] Bregman L M. Dokl. Akad. Nauk SSSR, 1965, 162: 487.
[16] NIU Fang-peng，LI Xin-guo，MAMATTURSUN Eziz, et al(牛芳鹏，李新国，麦麦提吐尔逊·艾则孜, 等). Journal of Zhejiang University(Agriculture and Life Sciences)[浙江大学学报(农业与生命科学版)], 2021, 47(5): 673.
[17] CAI Liang-hong，DING Jian-li(蔡亮红，丁建丽). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2017, 33(16): 144.
[18] TIAN An-hong，FU Cheng-biao，XIONG Hei-gang, et al(田安红，付承彪，熊黑钢, 等). Chinese Journal of Eco-Agriculture(中国生态农业学报), 2020, 28(2): 296.
[19] Li H, Jia S, Le Z. Sensors, 2019, 19(20): 4355.
[20] Žíala D, Zádorová T, Kapička J. Remote Sensing, 2017, 9(1): 28.