[1]张成才,侯佳彤,王 蕊,等.基于无人机多光谱与热红外数据的冬小麦土壤水分反演[J].郑州大学学报(工学版),2024,45(05):111-118.[doi:10.13705/j.issn.1671-6833.2024.05.002]
　ZHANG Chengcai,HOU Jiatong,WANG Rui,et al.Soil Moisture Inversion for Winter Wheat Field Based on UAVMultispectral and Thermal Infrared Data[J].Journal of Zhengzhou University (Engineering Science),2024,45(05):111-118.[doi:10.13705/j.issn.1671-6833.2024.05.002]
点击复制

基于无人机多光谱与热红外数据的冬小麦土壤水分反演()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
45
期数:
2024年05期
页码:
111-118
栏目:
出版日期:
2024-08-08

文章信息/Info

Title:
Soil Moisture Inversion for Winter Wheat Field Based on UAVMultispectral and Thermal Infrared Data
文章编号:
1671-6833(2024)05-0111-08
作者:
张成才1;  侯佳彤1;  王 蕊1;  姜明梁1; 2;  祝星星1
1. 郑州大学 水利与交通学院,河南 郑州 450001;2. 中国农业科学院农田灌溉研究所,河南 新乡 453002
Author(s):
ZHANG Chengcai1;  HOU Jiatong1;  WANG Rui1;  JIANG Mingliang1; 2;  ZHU Xingxing1
1. School of Water Conservancy and Transportation, Zhengzhou University, Zhengzhou 450001, China; 2. Institute of Farmland Irrigation of CAAS, Xinxiang 453002, China
关键词:
土壤水分反演;  无人机遥感;  多光谱;  热红外;  随机森林;  干旱指数;  植被覆盖度
Keywords:
soil moisture retrieval;  UAV remote;  multispectral;  thermal infrared;  random forest;  drought indices; fractional vegetation cover
分类号:
TP79S152. 7
DOI:
10.13705/j.issn.1671-6833.2024.05.002
文献标志码:
A
摘要:
引入植被覆盖度会在一定程度上提高土壤水分反演模型的精度,但大多数研究均基于归一化植被指数NDVI 估算植被覆盖度,未深入研究基于其他植被指数估算植被覆盖度对模型的影响。 为此,以河南省驻马店市西平县人和乡冬小麦部分种植区域为实验区,基于分辨率高、机动性强的无人机平台搭载多光谱与热红外成像仪开展冬小麦覆盖地表的土壤水分反演研究,探究引入不同植被覆盖度参数后模型精度的变化,并弥补基于卫星遥感影像的土壤水分监测分辨率低、时效性差的不足。 基于随机森林算法,将温度植被干旱指数 TVDI、垂直干旱指数PDI 两种干旱指数分别与 7 种植被指数估算的植被覆盖度参数耦合搭建土壤水分反演模型,并根据最优模型的反演结果对实验区的土壤水分空间分布情况进行分析。 同时,建立耦合 TVDI 与 PDI 指数、不引入植被覆盖度的土壤水分反演模型 TP 模型为对照组。 结果表明:在 0 ~ 10 cm 和 > 10 ~ 20 cm 深度时,TP 模型的决定系数 R2 分别为0. 606、0. 670,均方根误差 RMSE 分别为 0. 045、0. 041。 7 种引入植被覆盖度的模型精度较 TP 模型精度均有一定程度的提升,其中最优模型 TP OSAVI 的 R2 较 TP 模型分别提高 0. 143、0. 158,RMSE 分别降低 0. 7 百分点、0. 8 百分点。基于干旱指数引入植被覆盖度能够提高模型精度,且不同植被覆盖度参数对模型精度的提升程度有差异。
Abstract:
The introduction of fractional vegetation cover could improve the accuracy of soil moisture inversion model to some extent, but most studies estimated fractional vegetation cover based on normalized difference vegetationindex NDVI, and without in-depth study on the impact of vegetation coverage based on other vegetation indices onthe model. Therefore, taking the winter wheat planting area in Xiping County, Zhumadian City, Henan Province asthe experimental area, based on the UAV platform with high resolution and strong mobility, the multi-spectral andTIR imaging apparatus were equipped to carry out the soil moisture inversion research of winter wheat covered surface, and to explore the changes of model accuracy after introducing different fractional vegetation cover parameters, so as to make up for the limitations of soil moisture monitoring caused by the low resolution and poor timelinessof satellite remote sensing images. The two drought indices of temperature vegetation dryness index TVDI and perpendicular drought index PDI were combined with the parameters of fractional vegetation cover estimated throughseven vegetation indices, respectively, and seven soil moisture inversion models were constructed based on the random forest algorithm, and the spatial distribution of soil moisture in the experimental area was analyzed according tothe inversion results of the optimal model. At the same time, the soil moisture inversion model TP model, whichcould integrateed TVDI and PDI indices and without introducing fractional vegetation cover, was built as the controlgroup. The results showed that the R2of the TP model was 0. 606,0. 670, the root mean square error RMSE was0. 045、0. 041 for the depths of 0 to10 cm and >10 to 20 cm respectively. The accuracy of the seven models introducing fractional vegetation cover was improved to some extent compared with that of the TP model. Among them, theR2of the optimal model TP OSAVI was improved by 0. 143,0. 158, the RMSE was reduced by 0. 7 percentage points,0. 8 percentage points respectively, compared with the TP model. It showed that the introduction of fractional vegetation cover based on the drought indices could improve the accuracy of model inversion and different fractional vegetation cover had different effects on the accuracy of the model.

参考文献/References:

[ 1 ] 冯珊珊, 梁雪映, 樊风雷, 等. 基于无人机多光谱数据的农田土壤水分遥感监测[ J] . 华南师范大学学报(自然科学版) , 2020, 52(6) : 74-81.

FENG S S, LIANG X Y, FAN F L, et al. Monitoring offarmland soil moisture based on unmanned aerial vehiclemultispectral data[J]. Journal of South China Normal University (Natural Science Edition), 2020, 52(6): 74-81.
[ 2 ] 李德仁, 李明. 无人机遥感系统的研究进展与应用前景[ J] . 武 汉 大 学 学 报 ( 信 息 科 学 版) , 2014, 39(5) : 505-513, 540.
LI D R, LI M. Research advance and application prospect of unmanned aerial vehicle remote sensing system[ J] . Geomatics and Information Science of Wuhan University, 2014, 39(5) : 505-513, 540.
[ 3 ] 金伟, 葛宏立, 杜华强, 等. 无人机遥感发展与应用概况[ J] . 遥感信息, 2009, 24(1) : 88-92.
JIN W, GE H L, DU H Q, et al. A review on unmannedaerial vehicle remote sensing and its application[J]. Remote Sensing Information, 2009, 24(1): 88-92.
[ 4 ] 马仪, 黄组桂, 贾江栋, 等. 基于无人机-卫星遥感升尺度的土壤水分监测模型研究[ J] . 农业机械学报, 2023, 54(6) : 307-318.
MA Y, HUANG Z G, JIA J D, et al. Soil moisture monitoring model based on UAV-satellite remote sensingscale-up[ J ] . Transactions of the Chinese Society forAgricultural Machinery, 2023, 54(6) : 307-318.
[ 5 ] 葛少青, 张剑, 孙文, 等. 三种干旱指数在干旱区沼泽湿地土壤水分遥感反演中的应用[ J] . 生态学报,2018, 38(7) : 2299-2307.
GE S Q, ZHANG J, SUN W, et al. Application of threedrought indexes in soil moisture inversion using remotesensing in marsh wetlands of arid area[ J] . Acta Ecologica Sinica, 2018, 38(7) : 2299-2307.
[ 6 ] 蔡亮红, 丁建丽, 魏阳. 基于多源数据的土壤水分反演及空 间 分 异 格 局 研 究 [ J] . 土 壤 学 报, 2017, 54(5) : 1057-1067.
CAI L H, DING J L, WEI Y. Spatial variation and inversion of soil moisture based on multi-source data[ J] .Acta Pedologica Sinica, 2017, 54(5) : 1057-1067.
[ 7 ] 张文江, 陆其峰, 高志强, 等. 基于水分距平指数的2006 年四川盆地东部特大干旱遥感响应分析[ J]. 中国科学:D 辑 地球科学, 2008, 38(2): 251-260.
ZHANG W J, LU Q F, GAO Z Q, et al. Remote sensing response analysis of the severe drought in eastern Sichuan Basin in 2006 based on moisture anomaly index[J]. Science in China: Series D Earth Sciences, 2008,38(2) : 251-260.
[ 8 ] 高磊, 覃志豪, 卢丽萍. 基于植被指数和地表温度特征空间的农业干旱监测模型研究综述[ J] . 国土资源遥感, 2007, 19(3) : 1-7.
GAO L, QIN Z H, LU L P. An overview on agriculturaldrought monitoring models based on vegetation index andsurface temperature feature space [ J] . Remote Sensingfor Land & Resources, 2007, 19(3) : 1-7.
[ 9 ] 杨志辉, 赵军, 朱国锋, 等. 顾及植被覆盖影响的石羊河流域土壤水分遥感估算及空间格局分析[ J] . 生态学报, 2020, 40(23) : 8826-8837.
YANG Z H, ZHAO J, ZHU G F, et al. Remote sensing estimation and spatial pattern analysis of soil moisture in Shiyang River Basin considering the influence of vegetation cover[J]. Acta Ecologica Sinica, 2020, 40(23): 8826-8837.
[10] GHULAM A,QIN Q,TEYIP T,et al. Modified perpendiculardrought index ( MPDI ): a real-time drought monitoringmethod[ J]. Photogrammetry and Remote Sensing,2007,62(2):150-164.
[11] 朱燕香, 潘剑君, 白浩然, 等. 基于 Sentinel-2A 影像的 OPTRAM 模型及其改进模型的土壤水分估算研究[ J] . 南京农业大学学报, 2020, 43(4) : 682-689.
ZHU Y X, PAN J J, BAI H R, et al. Soil moisture estimation with the OPTRAM model and its improved modelbased on Sentinel-2A data[ J] . Journal of Nanjing Agricultural University, 2020, 43(4) : 682-689.
[12] PRICE J C. Using spatial context in satellite data to inferregional scale evapotranspiration[J]. IEEE Transactions onGeoscience and Remote Sensing, 1990, 28(5): 940-948.
[13] 王正兴, 刘闯, ALFREDO H. 植被指数研究进展: 从AVHRR-NDVI 到 MODIS-EVI [ J ] . 生 态 学 报, 2003,23(5) : 979-987.
WANG Z X, LIU C, ALFREDO H. From AVHRR-NDVIto MODIS-EVI: advances in vegetation index research[ J] . Acta Ecologica Sinica, 2003, 23(5) : 979-987.
[14] 杨曦, 武建军, 闫峰, 等. 基于地表温度-植被指数特征空间的区域土壤干湿状况[ J] . 生态学报, 2009, 29(3) : 1205-1216.
YANG X, WU J J, YAN F, et al. Assessment of regionalsoil moisture status based on characteristics of surfacetemperature / vegetation index space [ J] . Acta EcologicaSinica, 2009, 29(3) : 1205-1216.
[15] 冯海霞, 秦其明, 李滨勇, 等. 基于 SWIR-Red 光谱特征空间的农田干旱监测新方法[ J] . 光谱学与光谱分析, 2011, 31(11) : 3069-3073.
FENG H X, QIN Q M, LI B Y, et al. The new methodmonitoring agricultural drought based on SWIR-red spectrum feature space[ J] . Spectroscopy and Spectral Analysis, 2011, 31(11) : 3069-3073.
[16] 丁艳玲. 植被覆盖度遥感估算及其真实性检验研j究[D] . 北京: 中国科学院大学, 2015.
DING Y L. Study on remote sensing estimation of vegetation coverage and its authenticity test[ D] . Beijing: University of Chinese Academy of Sciences, 2015.
[17] 敖登, 杨佳慧, 丁维婷, 等. 54 种植被指数研究进展综述[ J] . 安徽农业科学, 2023, 51(1) : 13-21, 28.
AO D, YANG J H, DING W T, et al. Review of 54 vegetation indices [ J ] . Journal of Anhui Agricultural Sciences, 2023, 51(1) : 13-21, 28.
[18] 方匡南, 吴见彬, 朱建平, 等. 随机森林方法研究综述[ J] . 统计与信息论坛, 2011, 26(3) : 32-38.F
ANG K N, WU J B, ZHU J P, et al. A review of technologies on random forests [ J] . Statistics & InformationForum, 2011, 26(3) : 32-38.
[19] 李爱民, 王海隆, 许有成. 优化随机森林算法的城市湖泊 DOC 质量浓度遥感反演[ J] . 郑州大学学报( 工学版) , 2022, 43(6) : 90-96.
LI A M, WANG H L, XU Y C. Remote sensing retrievalof urban lake DOC concentration based on optimized random forest algorithm[ J] . Journal of Zhengzhou University(Engineering Science) , 2022, 43(6) : 90-96.
[20] 张智韬, 许崇豪, 谭丞轩, 等. 覆盖度对无人机热红外遥感反演玉米土壤含水率的影响[ J] . 农业机械学报, 2019, 50(8) : 213-225.
ZHANG Z T, XU C H, TAN C X, et al. Influence ofcoverage on soil moisture content of field corn inversedfrom thermal infrared remote sensing of UAV[ J] . Transactions of the Chinese Society for Agricultural Machinery,2019, 50(8) : 213-225.

更新日期/Last Update: 2024-09-02