融合孤立森林和深度学习的GNSS-IR土壤湿度反演

doi:10.13474/j.cnki.11-2246.2024.0716

摘要/Abstract

摘要： 针对GNSS反射信号遥感中单一特征参数数据质量参差不齐、可靠性差,模型反演结果不稳定的问题,本文提出了一种融合孤立森林和深度学习的GNSS-IR土壤湿度反演方法。试验结果表明,GNSS SNR的频率特征参数不适合土壤湿度的反演,而其振幅、相位特征参数与土壤湿度的相关性较高,可用于土壤湿度的反演;CNN、DBN和GRU 3种深度学习模型融合振幅和相位特征参数的反演结果与实测土壤湿度吻合度都较高;相比于仅利用振幅或相位的单一特征参数反演方法,本文方法反演精度提高了21.4%～55.8%,相关系数提高了4%～9.1%。

关键词: 土壤湿度, GNSS-IR, 深度学习, 孤立森林

Abstract: Aiming at the problems of uneven quality,poor reliability and unstable model inversion results of single characteristic parameter data in GNSS reflected signal remote sensing,this paper proposes a GNSS-IR soil moisture inversion method that combines isolated forest and deep learning. The experimental results show that the frequency characteristic parameters of GNSS SNR are not suitable for the inversion of soil moisture,while the amplitude and phase characteristic parameters are highly correlated with soil moisture,which can be used for the inversion of soil moisture. The inversion results of the fusion amplitude and phase characteristic parameters of the three deep learning models of CNN,DBN and GRU are in good agreement with the measured soil moisture. Compared with the single feature parameter inversion method using only amplitude or phase,the inversion accuracy of the proposed method is improved by 21.4%～55.8%,and the correlation coefficient is improved by 4%～9.1%.

Key words: soil moisture, GNSS-IR, deep learning, isolated forest

中图分类号:

P228

杨晓峰, 魏浩翰, 张强, 向云飞. 融合孤立森林和深度学习的GNSS-IR土壤湿度反演[J]. 测绘通报, 2024, 0(7): 88-94.

YANG Xiaofeng, WEI Haohan, ZHANG Qiang, XIANG Yunfei. GNSS-IR soil moisture inversion integrating isolated forest and deep learning[J]. Bulletin of Surveying and Mapping, 2024, 0(7): 88-94.

参考文献

[1] 林洁,陈效民,张勇.气候变化与土壤湿度关系的研究进展[J].土壤通报,2012,43(5):1271-1276.
[2] 严颂华,龚健雅,张训械,等.GNSS-R测量地表土壤湿度的地基试验[J].地球物理学报,2011,54(11):2735-2744.
[3] LARSON K M,GUTMANN E D,ZAVOROTNY V U,et al.Can we measure snow depth with GPS receivers？[J].Geophysical Research Letters,2009,36(17):L17502.
[4] LARSON K M,LÖFGREN J S,HAAS R.Coastal sea level measurements using a single geodetic GPS receiver[J].Advances in Space Research,2013,51(8):1301-1310.
[5] DUAN J J,SCHMUDE J M,LARSON K M.Effects of low temperature exposure on diapause,development,and reproductive fitness of the emerald ash borer (Coleoptera:Buprestidae):implications for voltinism and laboratory rearing[J].Journal of Economic Entomology,2021,114(1):201-208.
[6] LARSON K M,RAY R D,NIEVINSKI F G,et al.The accidental tide gauge:a GPS reflection case study from kachemak bay,Alaska[J].IEEE Geoscience and Remote Sensing Letters,2013,10(5):1200-1204.
[7] LARSON K M,SMALL E E,GUTMANN E D,et al.Use of GPS receivers as a soil moisture network for water cycle studies[J].Geophysical Research Letters,2008,35(24):L24405.
[8] BILICH A,LARSON K M.Correction published 29 March 2008:mapping the GPS multipath environment using the signal-to-noise ratio (SNR)[J].Radio Science,2007,42(6):1-16.
[9] 任超,张志刚,梁月吉,等.基于小波分析辅助GPS-IR反演土壤湿度研究[J].大地测量与地球动力学,2020,40(1):77-81.
[10] CHEW C C,SMALL E E,LARSON K M,et al.Effects of near-surface soil moisture on GPS SNR data:development of a retrieval algorithm for soil moisture[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(1):537-543.
[11] 梁月吉,任超,黄仪邦,等.利用GPS-IR监测土壤湿度的多星线性回归反演模型[J].测绘学报,2020,49(7):833-842.
[12] 丰秋林,郑南山.机器学习算法辅助的GPS信噪比观测值土壤湿度反演[J].测绘通报,2018(7):106-111.
[13] ZHANG Sibo,ROUSSEL N,BONIFACE K,et al.Use of reflected GNSS SNR data to retrieve either soil moisture or vegetation height from a wheat crop[J].Hydrology and Earth System Sciences,2017,21(9):4767-4784.
[14] REN Chao,LIANG Yueji,LU Xianjian,et al.Research on the soil moisture sliding estimation method using the LS-SVM based on multi-satellite fusion[J].International Journal of Remote Sensing,2019,40(5/6):2104-2119.
[15] 郭斐,陈惟杰,朱逸凡,等.一种融合相位、振幅与频率的GNSS-IR土壤湿度反演方法[J].武汉大学学报(信息科学版),2024,49(5):715-721.
[16] 陈堃,沈飞,曹新运,等.基于深度置信网络的GNSS-IR土壤湿度反演[J].测绘通报,2020(9):100-105.
[17] ROBERTS T M,COLWELL I,SHAH R,et al.GNSS-R soil moisture retrieval with a deep learning approach[C]//Proceedings of 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS.Brussels:IEEE,2021:147-150.
[18] 梁月吉,赖建民,任超,等.基于GNSS-IR双系统组合的土壤湿度多星线性回归反演模型[J].大地测量与地球动力学,2022,42(5):477-482.
[19] LIM B,ZOHREN S.Time-series forecasting with deep learning:a survey[J].Philosophical Transactions Series A,Mathematical,Physical,and Engineering Sciences,2021,379(2194):20200209.
[20] LÜ Jichao,ZHANG Rui,TU Jinsheng,et al.A GNSS-IR method for retrieving soil moisture content from integrated multi-satellite data that accounts for the impact of vegetation moisture content[J].Remote Sensing,2021,13(13):2442.
[21] 杨晓峰,魏浩翰,张强,等.小波与BP神经网络联合反演GNSS-IR高精度水库水位变化[J].导航定位与授时,2023,10(1):54-64.
[22] CHEW C,SMALL E E,LARSON K M.An algorithm for soil moisture estimation using GPS-interferometric reflectometry for bare and vegetated soil[J].GPS Solutions,2016,20(3):525-537.
[23] 王笑蕾,张勤,张双成.基于EMD和WD联合算法的GPS水汽时间序列的周期性振荡分析[J].武汉大学学报(信息科学版),2018,43(4):620-628.
[24] 黄伟建,李永涛,黄远.基于混合神经网络和注意力机制的混沌时间序列预测[J].物理学报,2021,70(1):235-243.
[25] 牛哲文,余泽远,李波,等.基于深度门控循环单元神经网络的短期风功率预测模型[J].电力自动化设备,2018,38(5):36-42.
[26] 贺小伟,徐靖杰,王宾,等.基于GRU-LSTM组合模型的云计算资源负载预测研究[J].计算机工程,2022,48(5):11-17.
[27] WANG Xiaolei,ZHANG Shuangcheng,WANG Lifu,et al.Analysis and combination of multi-GNSS snow depth retrievals in multipath reflectometry[J].GPS Solutions,2020,24(3):77.
[28] WANG Xiaolei,HE Xiufeng,XIAO Ruya,et al.Millimeter to centimeter scale precision water-level monitoring using GNSS reflectometry:application to the South-to-North Water Diversion Project,China[J].Remote Sensing of Environment,2021,265:112645.