基于高光谱和激光雷达遥感的水稻产量监测研究

doi:10.3864/j.issn.0578-1752.2021.14.004

摘要/Abstract

摘要：

【背景】快速、准确地估算水稻产量对于肥水精确管理及国家粮食政策的制定至关重要。高光谱与激光雷达遥感作为2种不同的主被动监测技术,为水稻长势信息获取提供了多样化手段。【目的】对比2种遥感监测手段在不同生态点的独立数据集中的验证精度,寻求可移植性强的产量估算模型,对水稻长势监测提供理论与技术支撑,及为精确农业提供科学指导具有重要意义。【方法】本研究通过实施3年（2016—2018年）包含不同地点、不同品种与不同氮素水平的水稻田间试验,在抽穗后各时期同步获取点云数据和光谱数据,结合线性回归与随机森林回归来估算产量,探究抽穗后点云数据与光谱数据估算水稻产量的差异;同时评估产量模型在不同数据集的时空可移植性,寻求可移植性强的产量估算模型。【结果】利用点云数据估算产量的精度（R² = 0.64—0.69）优于光谱数据的估算精度（R² = 0.20—0.58）;基于线性回归的产量估算模型,其验证精度明显优于基于随机森林回归的产量模型;产量模型在同一生态点的可移植性更强（不同生态点：RRMSE 16.69%—17.85%;同一生态点：RRMSE 11.37%—12.41%）。【结论】本研究为抽穗后水稻产量监测提供了新的方法和不同遥感手段的性能比较,为收获前作物产量的实时估算提供重要支撑。激光雷达技术凭借其全天候工作的特点,在长江中下游水稻产量实时监测中有着较好的应用前景。

关键词: 水稻, 产量, 激光雷达, 高光谱, 回归模型

Abstract:

【Background】Non-destructive and accurate estimation of crop biomass and yield is crucial for the quantitative diagnosis of growth status and planning of national food policies. Hyperspectral and Terrestrial Laser Scanning (TLS) remote sensing provide convenient and effective ways to monitor crop condition.【Objective】The aim of this study was to examine the feasibility of developing models with various independent datasets to build a universal yield monitoring model. The expected results can provide theoretical and technical support for rice growth monitoring and scientific guidance of precision agriculture.【Method】Field plot experiments were conducted in 2016, 2017 and 2018 and involved different study sites, nitrogen (N) rates, planting techniques and rice varieties. Linear regression (LR) and random forest (RF) were evaluated in estimating yield with TLS and spectral data collected since the heading stage, and the feasibility of developing models with various independent datasets was examined to build a universal yield monitoring model.【Result】 The results showed that TLS models exhibited higher estimation accuracies for yield in the heading stage, early-filling stage and late-filling stage (R²: 0.64-0.69) than hyperspectral models (R²: 0.20-0.58). Compared to RF, LR modeling yielded significantly higher validation accuracies for growth stages after heading. While the predictive model was transferred to other datasets, the validation accuracies from the same site (RRMSE: 11.37%-12.41%) were higher than those from a different site (RRMSE: 16.69%-17.85%).【Conclusion】The results suggested that TLS was a promising technique to monitor yield at post-heading stages with high accuracy and to overcome the saturation of canopy reflectance signals encountered in optical remote sensing.

Key words: rice, yield, LiDAR, hyperspectral, regression model

李朋磊, 张骁, 王文辉, 郑恒彪, 姚霞, 朱艳, 曹卫星, 程涛. 基于高光谱和激光雷达遥感的水稻产量监测研究[J]. 中国农业科学, 2021, 54(14): 2965-2976.

LI PengLei, ZHANG Xiao, WANG WenHui, ZHENG HengBiao, YAO Xia, ZHU Yan, CAO WeiXing, CHENG Tao. Assessment of Terrestrial Laser Scanning and Hyperspectral Remote Sensing for the Estimation of Rice Grain Yield[J]. Scientia Agricultura Sinica, 2021, 54(14): 2965-2976.

0
/ / 推荐

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: https://www.chinaagrisci.com/CN/10.3864/j.issn.0578-1752.2021.14.004

https://www.chinaagrisci.com/CN/Y2021/V54/I14/2965

图/表 18

图1

图2

图3

表1

表2

表3

表4

表5

图4

图5

图6

图7

图8

表6

图9

图10

图11

图12

参考文献 37

[1]	ALOLA A, ALOLA U V. The dynamic nexus of crop production and population growth: housing market sustainability pathway. Environmental Science and Pollution Research, 2018, 26:6472-6480. doi: 10.1007/s11356-018-04074-1
[2]	UN Food Agriculture Organization. How to feed the word in 2050. Discussion paper prepared for expert forum: 12-13. October 2009, released 23.
[3]	INTERPRETERS S. FAO-food and agriculture organization of the united nations. Science, 2013, 118(3077):13-23.
[4]	JULIAN P, MARK B, SARAH M. Food waste within food supply chains: quantification and potential for change to 2050. Philosophical Transactions of the Royal Society of London, 2010, 365(1554):3065-3081.
[5]	LIU J G, PATTEY E, MILLER J R, MCNAIRN H, SMITH A, HU B X. Estimating crop stresses, aboveground dry biomass and yield of corn using multi-temporal optical data combined with a radiation use efficiency model. Remote Sensing of Environment, 2010, 114(6):1167-1177. doi: 10.1016/j.rse.2010.01.004
[6]	MUTANGA O, ADAM E, CHO M A. High density biomass estimation for wetland vegetation using WorldView-2 imagery and random forest regression algorithm. International Journal of Applied Earth Observations and Geoinformation, 2012, 18(1):399-406. doi: 10.1016/j.jag.2012.03.012
[7]	亓雪勇, 田庆久. 光学遥感大气校正研究进展. 国土资源遥感, 2005, 45(4):4-9.
	QI X Y, TIAN Q J. The advance in the study of atmospheric correction for optical remote sensing. Remote Sensing for Land and Resources, 2005, 45(4):4-9. (in Chinese)
[8]	唐延林, 黄敬峰, 王人潮, 王福民. 水稻遥感估产模拟模式比较. 农业工程学报, 2004, 3(1):167-172.
	TANG Y L, HUANG J F, WANG R C, WANG F M. Comparsion of yield estimation simulated models of rice by remote sensing. Transactions of The Chinese Society of Agricultural Engineering, 2004, 3(1):167-172. (in Chinese)
[9]	OWERS, C J, ROGERS K, WOODROFFE C D. Terrestrial laser scanning to quantify above-ground biomass of structurally complex coastal wetland vegetation. Estuarine Coastal and Shelf Science, 2018, 204:164-176. doi: 10.1016/j.ecss.2018.02.027
[10]	EITEL J U H, MAGNEY T S, VIERLING L A, BROWN T T, HUGGINS D R. LiDAR based biomass and crop nitrogen estimates for rapid, non-destructive assessment of wheat nitrogen status. Field Crops Research, 2014, 159:21-32. doi: 10.1016/j.fcr.2014.01.008
[11]	TILLY N, HOFFMEISTER D, CAO Q, LENZ-WIEDEMANN V, BARETH G. Precise plant height monitoring and biomass estimation with Terrestrial Laser Scanning in paddy rice. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2013, II-5/W2:295-300. doi: 10.5194/isprsannals-II-5-W2-295-2013
[12]	TILLY N, AASEN H, BARETH G. Fusion of plant height and vegetation indices for the estimation of barley biomass. Remote Sensing, 2015, 7:11449-11480. doi: 10.3390/rs70911449
[13]	GUO Q H, WU F F, PANG S X, ZHAO X Q, CHEN L H, LIU J, XUE B L, XU G C, LI L, JING H C. Crop 3D: a platform based on LiDAR for 3D high-throughput crop phenotyping. Scientia Sinica, 2016, 12:121-141.
[14]	LUMME J, KARJALAINEN M, KAARTINEN H, KUKKO A, HYYPPÄ J, HYYPPÄ H, JAAKKOLA A, KLEEMOLA J. Terrestrial laser scanning of agricultural crops. International Journal of Remote Sensing, 2008, 26(7):563-566. doi: 10.1080/01431160512331299270
[15]	MCKINION J M, WILLERS J L, JENKINS J N. Comparing high density LIDAR and medium resolution GPS generated elevation data for predicting yield stability. Computers and Electronics in Agriculture, 2010, 74(2):244-249. doi: 10.1016/j.compag.2010.08.011
[16]	LI Z H, WANG J H, XU X G, ZHAO C J, JIN X L, YANG G J, FENG H K. Assimilation of two variables derived from hyperspectral data into the DSSAT-CERES model for grain yield and quality estimation. Remote Sensing, 2015, 35(9):12400-12418.
[17]	王延颐. 植被指数与水稻长势及产量结构要素关系的研究. 国土资源遥感, 1996, 2(1):56-59.
	WANG Y Y. The relationship between vegetation index and rice growth and rice yield components. Remote Sensing for Land and Resources, 1996, 2(1):56-59. (in Chinese)
[18]	李卫国, 王纪华, 赵春江, 李存军, 王永华. 基于生态因子的冬小麦产量遥感估测研究. 麦类作物学报, 2009, 29(5):213-220.
	LI W G, WANG J H, ZHAO C J, LI C J, WANG Y H. Study on remote sensing estimation of winter wheat yield based on eco-environmental factors. Journal of Triticeae Crop, 2009, 29(5):213-220. (in Chinese)
[19]	薛利红, 曹卫星, 罗卫红. 基于冠层反射光谱的水稻产量预测模型. 遥感学报, 2005, 7(1):102-107.
	XUE L H, CAO W X, LUO W H. Rice yield forecasting model with canopy reflectance spectra. Journal of Remote Sensing, 2005, 7(1):102-107. (in Chinese)
[20]	CHENG T, RIVARD B, SÁNCHEZ-AZOFEIFA A G, FÉRET J B, JACQUEMOUD S, USTIN S L. Deriving leaf mass per area (LMA) from foliar reflectance across a variety of plant species using continuous wavelet analysis. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 87:28-38. doi: 10.1016/j.isprsjprs.2013.10.009
[21]	宋开山, 刘殿伟, 王宗明, 吕冬梅, 张柏, 任春颖, 杜嘉. 基于小波分析的玉米叶绿素a与LAI高光谱反演模型研究. 农业系统科学与综合研究, 2011, 3(2):28-34.
	SONG K S, LIU D W, WANG Z M, LÜ D M, ZHANG B, REN C Y, DU J. Corn chlorophyll-a concentration and LAI estimation models based on wavelet transformed canopy hyperspectral reflectance. System Sciences and Comprehensive Studies in Agriculture, 2011, 3(2):28-34. (in Chinese)
[22]	宋开山, 张柏, 王宗明, 刘殿伟, 刘焕军. 基于小波分析的大豆叶绿素a含量高光谱反演模型. 植物生态学报, 2008, 2(1):157-165.
	SONG K S, ZHANG B, WANG Z M, LIU D W, LIU H J. Soybean chlorophyll a concentration estimation models based on wavelet- transformed, in situ collected, canopy hyperspectral data. Journal of Plant Ecology, 2008, 2(1):157-165. (in Chinese)
[23]	洪雪. 基于水稻高光谱遥感数据的植被指数产量模型研究[D]. 沈阳: 沈阳农业大学, 2017.
	HONG X. Rice yield model research based on vegetation index of hyperspectral remote sensing data[D]. Shenyang: Shenyang Agricultural University, 2017. (in Chinese)
[24]	王娣. 高光谱与多光谱遥感水稻估产研究[D]. 武汉: 武汉大学, 2017.
	WANG D. Hyperspectral and multispectral remote sensing study on yield estimation of rice[D]. Wuhan: Wuhan University, 2017. (in Chinese)
[25]	黄春燕, 王登伟, 陈冠文, 袁杰, 祁亚琴, 陈燕, 程诚. 基于高光谱植被指数的棉花干物质积累估算模型研究. 棉花学报, 2006, 18(2):115-119.
	HUANG C Y, WANG D W, CHEN G W, YUAN J, QI Y Q, CHEN Y, CHENG C. Estimation modeling of cotton dry matter accumulation based on hyperspectral vegetation index. Cotton Science, 2006, 18(2):115-119. (in Chinese)
[26]	许童羽, 洪雪, 陈春玲, 周云成, 曹英丽, 于丰华, 李娜. 基于冠层NDVI数据的北方粳稻产量模型研究. 浙江农业学报, 2016, 28(10):1790-1795.
	XU T Y, HONG X, CHEN C L, ZHOU Y C, CAO Y L, YU F H, LI N. Study on northern japonica rice yield model based on canopy date of NDVI. Acta Agriculturae Zhejiangensis, 2016, 28(10):1790-1795. (in Chinese)
[27]	宋红燕, 胡克林, 彭希. 基于高光谱技术的覆膜旱作水稻植株氮含量及籽粒产量估算. 中国农业大学学报, 2016, 4(8):27-34.
	SONG H Y, HU K L, PENG X. Crop nitrogen content diagnosis and yield estimation in ground cover rice production system based on hyperspectral data. Journal of China Agricultural University, 2016, 4(8):27-34. (in Chinese)
[28]	SHIBAYAMA M. AKIYAMA T. Estimating grain yield of maturing rice canopies using high spectral resolution reflectance measurements. Remote Sensing of Environment, 1991, 36(1):45-53. doi: 10.1016/0034-4257(91)90029-6
[29]	BAI J H, LI S K, WANG K R, SUI X Y, CHEN B, WANG F Y. Estimating aboveground fresh biomass of different cotton canopy types with homogeneity models based on hyper spectrum parameters. Agricultural Sciences in China, 2007, 6(4):437-445. doi: 10.1016/S1671-2927(07)60067-4
[30]	LI P, ZHANG X, WANG W H, ZHENG H B, YAO X, TIAN Y C, ZHU Y, CAO W X, CHEN Q, CHENG T. Estimating aboveground and organ biomass of plant canopies across the entire season of rice growth with terrestrial laser scanning. International Journal of Applied Earth Observation and Geoinformation, 2020, 91:102-132.
[31]	GITELSON A A, VIÑA A, ARKEBAUER T J, RUNDQUIST D C, KEYDAN G P, LEAVITT B, KEYDAN G. Remote estimation of leaf area index and green leaf biomass in maize canopies. Geophysical Research Letters, 2003, 30:335-343.
[32]	HUDAK A T, LEFSKY M A, COHEN W B, BERTERRETCHE M. Integration of LiDAR and Landsat ETM+ data for estimating and mapping forest canopy height. Remote Sensing of Environment, 2015, 82(2/3):397-416. doi: 10.1016/S0034-4257(02)00056-1
[33]	WEI F, HUI F, YANG W N, DUAN L F, CHEN G X, XIONG L Z, LIU Q. High-throughput volumetric reconstruction for 3D wheat plant architecture studies. Journal of Innovative Optical Health Sciences, 2016, 9(5):1650037. doi: 10.1142/S1793545816500371
[34]	HUANG J F, BLACKBURN G A. Optimizing predictive models for leaf chlorophyll concentration based on continuous wavelet analysis of hyperspectral data. International Journal of Remote Sensing, 2011, 32(24):9375-9396. doi: 10.1080/01431161.2011.558130
[35]	CHEN Q. Modeling aboveground tree woody biomass using national- scale allometric methods and airborne LiDAR. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 106:95-106. doi: 10.1016/j.isprsjprs.2015.05.007
[36]	DAI B, GU C S, ZHAO E, QIN X N. Statistical model optimized random forest regression model for concrete dam deformation monitoring. Structural Control and Health Monitoring, 2018, 25(6):1-15.
[37]	TILLY N, HOFFMEISTER D, CAO Q, LENZ-WIEDEMANN V, MIAO Y X, BARETH G. Transferability of models for estimating paddy rice biomass from spatial plant height data. Agriculture, 2015, 5:538-552. doi: 10.3390/agriculture5030538

指数 Vegetation index	计算公式 Equation	文献 Reference
差值Deviation
DVI [1200,680]	R₁₂₀₀-R₆₈₀	[9]
DVI [1200,440]	R₁₂₀₀-R₄₄₀	[9]
DVI [800,550]	R₈₀₀-R₅₅₀	[9]
比值Ratio
SR [609,518]	(R₆₀₉/R₅₁₈)-1	[24]
SR [1971,2018]	(R₁₉₇₁/R₂₀₁₈)-1	[25]
SR [750,673]	R₇₅₀/R₆₇₃	[26]
归一化Normalization
NDVI [1200,550]	(R₁₂₀₀-R₅₅₀)/(R₁₂₀₀+R₅₅₀)	[27]
NDVI[800,680]	(R₈₀₀-R₆₈₀)/(R₈₀₀+R₆₈₀)	[28]
NDVI [608,518]	(R₆₀₉-R₅₁₈)/(R₆₀₉+R₅₁₈)	[29]

结构参数 Structural parameter	描述 Description
Height mean ( H_mean)	高度平均值 Mean of height
Height min (H_min)	高度最小值 Minimum of height
Height max (H_max)	高度最大值 Maximum of height
Height standard deviation (H_std)	高度标准偏差 Standard deviation of height
Height coefficient of variation (H_cov)	高度变异系数 Variable coefficient of height
Height kurtosis (H_k)	高度峰度 Kurtosis of height
Height skewness (H_s)	高度偏度 Skewness of height
Height percentile (H_1st, H_5th, H_10th, H_25th, H_50th, H_75th, H_95th, H_99th)	高度1st, 5th, 10th, 25th, 50th, 75th, 95th, 和99th 百分位 Percent of 1st, 5th, 10th, 25th, 50th, 75th, 95th, and 99th height

数据集 Dataset	年份 Year	试验地点 Site	样本数 Number of samples	品种 Variety	播栽方式 Planting technique
训练数据集 Training dataset	2017	兴化Xinghua	72	南粳9108 & 甬优2640 Nanjing 9108 & Yongyou 2640	钵苗移栽Tray seeding transplanting、毯苗移栽Blanket seeding transplanting、直播Direct seeding
验证数据集1 Validation dataset 1	2016	如皋Rugao	36	武运粳24 & Y两优1号 Wuyunjing 24 & Y Liangyou 1	旱育秧人工移栽 Dried-seedling manual transplanting
验证数据集2 Validation dataset 2	2018	兴化Xinghua	48	南粳9108 & 甬优2640 Nanjing 9108 & Yongyou 2640	钵苗移栽Tray seedling transplanting、毯苗移栽Blanket seeding transplanting

植被指数 Vegetation index	抽穗期 Heading	灌浆前期 Early filling	灌浆后期 Late filling
R₁₂₀₀-R₆₈₀	0.12*	0.41**	0.23**
R₁₂₀₀-R₄₄₀	0.10*	0.46**	0.24**
R₈₀₀-R₅₅₀	0.17**	0.41**	0.52**
(R₆₀₉/R₅₁₈）-1	0.20**	ns	0.01
R₇₅₀/R₆₇₃	0.16**	0.29**	0.27**
(R₁₉₇₁/R₂₀₁₈）-1	0.03	0.04	0.02
(R₁₂₀₀-R₅₅₀)/(R₁₂₀₀+R₅₅₀)	0.17**	0.31**	0.28**
(R₈₀₀-R₆₈₀)/(R₈₀₀+R₆₈₀)	0.12**	0.32**	0.23**
(R₆₀₉-R₅₁₈)/(R₆₀₉+R₅₁₈)	0.16**	ns	0.01

	抽穗期 Heading	灌浆前期 Early filling	灌浆后期 Late filing
最优小波特征 Optimal wavelet feature	WF_730,3	WF_1200,3	WF_1185,3
R²	0.34**	0.58**	0.58**