基于固定翼无人机多光谱影像的水稻长势关键指标无损监测

doi:10.3864/j.issn.0578-1752.2023.21.004

摘要/Abstract

摘要：

【背景】近年来随着遥感技术的快速发展，实时无损监测作物生长状况已成为当前研究热点，遥感获取的农情信息将为实现大面积作物精确管理提供指导。在众多遥感监测平台里，无人机因其操作简单、使用成本低等特点而受到广泛关注，无人机搭载多光谱相机可以快速获取作物的长势信息。【目的】尝试将固定翼无人机多光谱影像纹理信息与光谱信息结合，探究“图谱”信息对水稻长势指标的监测效果。【方法】通过开展两年涉及不同播期、品种、播栽方式、施氮水平的水稻田间试验，在水稻关键生育期使用固定翼无人机搭载Sequoia多光谱相机获取水稻冠层遥感影像，同步进行地上部破坏性取样以获取水稻叶面积指数（LAI）、地上部生物量（AGB）和植株氮含量（PNC）等农学指标，采用简单线性回归、偏最小二乘回归和人工神经网络回归算法，构建基于固定翼无人机多光谱影像的水稻长势指标监测模型，比较分析光谱纹理信息在不同模型中的监测效果。【结果】利用简单线性回归方法探究了植被指数（VI）、单波段纹理特征与水稻LAI、AGB和PNC间的定量关系，结果表明植被指数与LAI和AGB之间有较强的相关性，表现最好的植被指数为CIRE和NDRE，R²分别为0.80和0.76，但对于PNC的监测，植被指数并未达到理想的效果，表现最好的RESAVI和NDRE与PNC的决定系数仅为0.13。通过简单线性回归进一步发现单波段的纹理特征在对水稻生长指标的监测中表现并不理想；为进一步分析影像纹理对上述3个指标的监测效果，参照植被指数的构建方法构建了归一化纹理指数（NDTI）、比值纹理指数（RTI）和差值纹理指数（DTI），通过相关性分析发现新构建的纹理指数（TI）相较于单波段纹理特征对水稻生长指标的监测精度有所提升，但效果并未好于植被指数。为实现光谱与纹理间的结合，采用偏最小二乘和人工神经网络的建模方法，以VI、VI+TI为不同的输入参数组合进行水稻LAI、AGB和PNC的监测模型构建，结果表明采用偏最小二乘和人工神经网络的建模方法与简单线性回归相比模型的监测精度均得到了大幅提升，以VI+TI为输入变量，采用人工神经网络构建的模型在模型验证中取得了最佳效果，LAI模型的验证R²由0.75提升至0.86，AGB和PNC的模型验证R²也分别由0.72和0.26提升至0.92和0.86，同时模型的RMSE均显著降低。【结论】利用固定翼无人机采集水稻冠层多光谱影像，通过人工神经网络算法融合光谱和纹理信息能够有效提升水稻LAI、AGB和PNC的监测精度，该研究结果将为快速大面积作物长势监测提供理论依据。

关键词: 无人机, 植被指数, 纹理特征, 长势, 水稻

Abstract:

【Background】In recent years, with the rapid development of remote sensing technology, real-time and non-destructive monitoring of crop growth status has become a research hotspot. Remote sensing-derived agricultural information will provide guidance for the precise management of large-scale crops. Among various remote sensing monitoring platforms, unmanned aerial vehicles (UAVs) have attracted wide attention due to their simple operation and low cost. UAVs equipped with multispectral cameras can quickly obtain crop growth conditions.【Objective】This study attempted to combine texture information and spectral information of multispectral images of fixed-wing UAVs to explore the monitoring effect of “atlas” information on rice growth indicators.【Method】A two-year rice field experiment involving different sowing dates, varieties, planting methods and nitrogen levels was conducted. During the key growth stages of rice, remote sensing images of the rice canopy were obtained using a Sequoia multispectral camera mounted on a fixed-wing UAV. Shoot destructive sampling was conducted simultaneously to obtain leaf area index (LAI), aboveground biomass (AGB), plant nitrogen content (PNC) and other agronomic indexes of rice. Simple regression, partial least squares regression and artificial neural network algorithms were used to construct rice growth index monitoring model based on multispectral images of fixed-wing UAV. The monitoring effects of spectral texture information in different models were compared and analyzed.【Result】The quantitative relationship between vegetation index (VI), single-band texture features and rice LAI, AGB, and PNC was explored using simple linear regression. The results showed that vegetation indexes had strong correlations with LAI and AGB, with the best-performing indexes being CIRE and NDRE, with R²values of 0.80 and 0.76, respectively. However, for PNC monitoring, vegetation indexes did not achieve ideal results, with the best-performing RESAVI and NDRE having R²values of only 0.13 with PNC. Further analysis using simple linear regression revealed that single-band texture features did not perform well in monitoring rice growth indicators. In order to further analyze the monitoring effect of image texture on the above three indexes, normalized texture indexes (NDTI), ratio texture indexes (RTI), and difference texture indexes (DTI) were constructed by referring to the construction method of VI. Correlation analysis showed that the newly constructed texture index (TI) improved the monitoring accuracy of rice growth indicators compared to single-band texture feature but did not perform better than vegetation indexes. To combine spectral and texture information, partial least squares and artificial neural network modeling methods were adopted in this paper. VI and VI+TI were used as different input parameter combinations to construct rice LAI, AGB and PNC monitoring models. The results showed that both partial least squares and artificial neural network modeling methods significantly improved the monitoring accuracy compared to simple linear regression. The best performance was achieved using VI+TI as input variables and an artificial neural network model for validation, with validation R²values for LAI, AGB, and PNC models increasing from 0.75, 0.72, and 0.26 to 0.86, 0.92, and 0.86, respectively, while RMSE values were significantly reduced.【Conclusion】The monitoring accuracy of rice LAI, AGB and PNC can be effectively improved by using the fixed-wing UAV to collect multispectral images of rice canopy and using the texture features and reflectance information as input parameters of the model through the model construction method of artificial neural network. The research results will provide a theoretical basis for rapid monitoring of large area crop growth.

Key words: unmanned aerial vehicle (UAV), vegetation index (VI), texture feature, growth, rice

王伟康, 张嘉懿, 汪慧, 曹强, 田永超, 朱艳, 曹卫星, 刘小军. 基于固定翼无人机多光谱影像的水稻长势关键指标无损监测[J]. 中国农业科学, 2023, 56(21): 4175-4191.

WANG WeiKang, ZHANG JiaYi, WANG Hui, CAO Qiang, TIAN YongChao, ZHU Yan, CAO WeiXing, LIU XiaoJun. Non-Destructive Monitoring of Rice Growth Key Indicators Based on Fixed-Wing UAV Multispectral Images[J]. Scientia Agricultura Sinica, 2023, 56(21): 4175-4191.

图/表 17

图1

图2

表1

表2

表3

图3

表4

图4

表5

图5

图6

图7

图8

图9

图10

图11

图12

参考文献 41

[1]	陈明江, 刘贵富, 余泓, 王冰, 李家洋. 水稻高产优质的分子基础与品种设计. 科学通报, 2018, 63(14): 1275-1289.
	CHEN M J, LIU G F, YU H, WANG B, LI J Y. Towards molecular design of rice plant architecture and grain quality. Chinese Science Bulletin, 2018, 63(14): 1275-1289. (in Chinese)
[2]	JIANG J, ATKINSON P, ZHANG J Y, LU R H, ZHOU Y Y, CAO Q, TIAN Y C, ZHU Y, CAO W X, LIU X J. Combining fixed-wing UAV multispectral imagery and machine learning to diagnose winter wheat nitrogen status at the farm scale. European Journal of Agronomy, 2022, 138: 126537. doi: 10.1016/j.eja.2022.126537
[3]	ZHAO B, ATA-UL-KARIM S T, DUAN A W, LIU Z D, WANG X L, XIAO J F, LIU Z G, QIN A Z, NING D F, ZHANG W Q, LIAN Y H. Determination of critical nitrogen concentration and dilution curve based on leaf area index for summer maize. Field Crops Research, 2018, 228: 195-203. doi: 10.1016/j.fcr.2018.09.005
[4]	ATA-UL-KARIM S T, ZHU Y, YAO X, CAO W X. Determination of critical nitrogen dilution curve based on leaf area index in rice. Field Crops Research, 2014, 167: 76-85. doi: 10.1016/j.fcr.2014.07.010
[5]	WANG X B, MIAO Y X, DONG R, ZHA H N, XIA T T, CHEN Z C, KUSNIEREK K, MI G H, SUN H, LI M Z. Machine learning-based in-season nitrogen status diagnosis and side-dress nitrogen recommendation for corn. European Journal of Agronomy, 2021, 123: 126193. doi: 10.1016/j.eja.2020.126193
[6]	BASCON M V, NAKATA T, SHIBATA S, TAKATA I, KOBAYASHI N, KATO Y, INOUE S, DOI K, MURASE J, NISHIUCHI S. Estimating yield-related traits using UAV-derived multispectral images to improve rice grain yield prediction. Agriculture, 2022, 12(8): 1141. doi: 10.3390/agriculture12081141
[7]	FU Z P, JIANG J, GAO Y, KRIENKE B, WANG M, ZHONG K, CAO Q, TIAN Y C, ZHU Y, CAO W X, LIU X J. Wheat growth monitoring and yield estimation based on multi-rotor unmanned aerial vehicle. Remote Sensing, 2020, 12(3): 508. doi: 10.3390/rs12030508
[8]	TAO H L, FENG H K, XU L J, MIAO M K, YANG G J, YANG X D, FAN L L. Estimation of the yield and plant height of winter wheat using UAV-based hyperspectral images. Sensors, 2020, 20(4): 1231. doi: 10.3390/s20041231
[9]	LU N, ZHOU J, HAN Z X, LI D, CAO Q, YAO X, TIAN Y C, ZHU Y, CAO W X, CHENG T. Improved estimation of aboveground biomass in wheat from RGB imagery and point cloud data acquired with a low-cost unmanned aerial vehicle system. Plant Methods, 2019, 15: 17. doi: 10.1186/s13007-019-0402-3 pmid: 30828356
[10]	陈鹏飞, TREMBLAY N, 王纪华, VIGNEAULT P, 黄文江, 李保国. 估测作物冠层生物量的新植被指数的研究. 光谱学与光谱分析, 2010, 30(2): 512-517.
	CHEN P F, TREMBLAY N, WANG J H, VIGNEAULT P, HUANG W J, LI B G. New index for crop canopy fresh biomass estimation. Spectroscopy and Spectral Analysis, 2010, 30(2): 512-517. (in Chinese)
[11]	YANG G J, LIU J G, ZHAO C J, LI Z H, HUANG Y B, YU H Y, XU B, YANG X D, ZHU D M, ZHANG X Y, ZHANG R Y, FENG H K, ZHAO X Q, LI Z H, LI H L, YANG H. Unmanned aerial vehicle remote sensing for field-based crop phenotyping: Current status and perspectives. Frontiers in Plant Science, 2017, 8: 1111. doi: 10.3389/fpls.2017.01111 pmid: 28713402
[12]	YANG K L, GONG Y, FANG S H, DUAN B, YUAN N G, PENG Y, WU X T, ZHU R S. Combining spectral and texture features of UAV images for the remote estimation of rice LAI throughout the entire growing season. Remote Sensing, 2021, 13(15): 3001. doi: 10.3390/rs13153001
[13]	MAIMAITIJIANG M, SAGAN V, SIDIKE P, HARTLING S, ESPOSITO F, FRITSCHI F B. Soybean yield prediction from UAV using multimodal data fusion and deep learning. Remote Sensing of Environment, 2020, 237: 111599. doi: 10.1016/j.rse.2019.111599
[14]	万亮, 杜晓月, 陈硕博, 于丰华, 朱姜蓬, 许童羽, 何勇, 岑海燕. 基于无人机多源图谱融合的水稻稻穗表型监测. 农业工程学报, 2022, 38(9): 162-170.
	WAN L, DU X Y, CHEN S B, YU F H, ZHU J P, XU T Y, HE Y, CEN H Y. Rice panicle phenotyping using UAV-based multi-source spectral image data fusion. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(9): 162-170. (in Chinese)
[15]	戴震. 基于无人机多源影像信息融合的玉米冠层氮含量遥感监测研究[D]. 合肥: 安徽大学, 2021.
	DAI Z. Nitrogen content monitoring of corn canopy based on UAV multi-source data fusion[D]. Hefei: Anhui University, 2021. (in Chinese)
[16]	ZHENG L, TAO J P, BAO Q, WENG S Z, ZHANG Y K, ZHAO J L, HUANG L S. Combining spectral and textures of digital imagery for wheat aboveground biomass estimation//International Conference on Electronic Information Technology (EIT 2022): SPIE, 2022: 304-307.
[17]	WULDER M A, LEDREW E F, FRANKLIN S E, LAVIGNE M B. Aerial image texture information in the estimation of northern deciduous and mixed wood forest leaf area index (LAI). Remote Sensing of Environment, 1998, 64(1): 64-76. doi: 10.1016/S0034-4257(97)00169-7
[18]	刘俊, 毕华兴, 朱沛林, 孙菁, 朱金兆, 陈涛. 基于ALOS遥感数据纹理及纹理指数的柞树蓄积量估测. 农业机械学报, 2014, 45(7): 245-254.
	LIU J, BI H X, ZHU P L, SUN J, ZHU J Z, CHEN T. Estimating stand volume of Xylosma racemosum forest based on texture parameters and derivative texture indices of ALOS imagery. Transactions of the Chinese Society for Agricultural Machinery, 2014, 45(7): 245-254. (in Chinese)
[19]	陈鹏飞, 梁飞. 基于低空无人机影像光谱和纹理特征的棉花氮素营养诊断研究. 中国农业科学, 2019, 52(13): 2220-2229. doi: 10.3864/j.issn.0571-1752.2019.13.003.
	CHEN P F, LIANG F. Cotton nitrogen nutrition diagnosis based on spectrum and texture feature of images from low altitude unmanned aerial vehicle. Scientia Agricultura Sinica, 2019, 52(13): 2220-2229. doi: 10.3864/j.issn.0571-1752.2019.13.003. (in Chinese)
[20]	陈鹏, 冯海宽, 李长春, 杨贵军, 杨钧森, 杨文攀, 刘帅兵. 无人机影像光谱和纹理融合信息估算马铃薯叶片叶绿素含量. 农业工程学报, 2019, 35(11): 63-74.
	CHEN P, FENG H K, LI C C, YANG G J, YANG J S, YANG W P, LIU S B. Estimation of chlorophyll content in potato using fusion of texture and spectral features derived from UAV multispectral image. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(11): 63-74. (in Chinese)
[21]	WANG Y Y, ZHANG K, TANG C L, CAO Q, TIAN Y C, ZHU Y, CAO W X, LIU X J. Estimation of rice growth parameters based on linear mixed-effect model using multispectral images from fixed-wing unmanned aerial vehicles. Remote Sensing, 2019, 11(11): 1371. doi: 10.3390/rs11111371
[22]	LIU X H, WANG L. Feasibility of using consumer-grade unmanned aerial vehicles to estimate leaf area index in mangrove forest. Remote Sensing Letters, 2018, 9(11): 1040-1049. doi: 10.1080/2150704X.2018.1504339
[23]	ZHANG J Y, QIU X L, WU Y T, ZHU Y, CAO Q, LIU X J, CAO W X. Combining texture, color, and vegetation indices from fixed-wing UAS imagery to estimate wheat growth parameters using multivariate regression methods. Computers and Electronics in Agriculture, 2021, 185: 106138. doi: 10.1016/j.compag.2021.106138
[24]	田庆久, 闵祥军. 植被指数研究进展. 地球科学进展, 1998, 13(4): 327-333.
	TIAN Q J, MIN X J. Advances in study on vegetation indices. Advance in Earth Sciences, 1998, 13(4): 327-333. (in Chinese)
[25]	杨福芹, 冯海宽, 肖天豪, 李天驰, 郭向前. 融合无人机影像光谱与纹理特征的冬小麦氮营养指数估算. 农业现代化研究, 2020, 41(4): 718-726.
	YANG F Q, FENG H K, XIAO T H, LI T C, GUO X Q. Nitrogen nutrition index estimation in winter wheat by UAV spectral information and texture feature fusion. Research of Agricultural Modernization, 2020, 41(4): 718-726. (in Chinese)
[26]	杭艳红, 苏欢, 于滋洋, 刘焕军, 官海翔, 孔繁昌. 结合无人机光谱与纹理特征和覆盖度的水稻叶面积指数估算. 农业工程学报, 2021, 37(9): 64-71.
	HANG Y H, SU H, YU Z Y, LIU H J, GUAN H X, KONG F C. Estimation of rice leaf area index combining UAV spectrum, texture features and vegetation coverage. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(9): 64-71. (in Chinese)
[27]	李岚涛, 郭宇龙, 韩鹏, 王丹丹, 王宜伦. 基于高光谱的冬小麦不同生育时期地上部生物量监测. 麦类作物学报, 2021, 41(7): 904-913.
	LI L T, GUO Y L, HAN P, WANG D D, WANG Y L. Estimation of shoot biomass at different growth stages of winter wheat based on hyperspectral reflectance. Journal of Triticeae Crops, 2021, 41(7): 904-913. (in Chinese)
[28]	YUAN H H, YANG G J, LI C C, WANG Y J, LIU J G, YU H Y, FENG H K, XU B, ZHAO X Q, YANG X D. Retrieving soybean leaf area index from unmanned aerial vehicle hyperspectral remote sensing: Analysis of RF, ANN, and SVM regression models. Remote Sensing, 2017, 9(4): 309. doi: 10.3390/rs9040309
[29]	赵晓阳, 张建, 张东彦, 周新根, 刘小辉, 谢静. 低空遥感平台下可见光与多光谱传感器在水稻纹枯病病害评估中的效果对比研究. 光谱学与光谱分析, 2019, 39(4): 1192-1198.
	ZHAO X Y, ZHANG J, ZHANG D Y, ZHOU X G, LIU X H, XIE J. Comparison between the effects of visible light and multispectral sensor based on low-altitude remote sensing platform in the evaluation of rice sheath blight. Spectroscopy and Spectral Analysis, 2019, 39(4): 1192-1198. (in Chinese)
[30]	LIU Y N, LIU S S, LI J, GUO X Y, WANG S Q, LU J W. Estimating biomass of winter oilseed rape using vegetation indices and texture metrics derived from UAV multispectral images. Computers and Electronics in Agriculture, 2019, 166: 105026. doi: 10.1016/j.compag.2019.105026
[31]	YUE J B, YANG G J, TIAN Q J, FENG H K, XU K J, ZHOU C Q. Estimate of winter-wheat above-ground biomass based on UAV ultrahigh-ground-resolution image textures and vegetation indices. ISPRS Journal of Photogrammetry and Remote Sensing, 2019, 150: 226-244. doi: 10.1016/j.isprsjprs.2019.02.022
[32]	ZHENG H B, CHENG T, ZHOU M, LI D, YAO X, TIAN Y C, CAO W X, ZHU Y. Improved estimation of rice aboveground biomass combining textural and spectral analysis of UAV imagery. Precision Agriculture, 2019, 20(3): 611-629. doi: 10.1007/s11119-018-9600-7
[33]	ZHENG H B, ZHOU M, ZHU Y, CHENG T. Exploiting the textural information of UAV multispectral imagery to monitor nitrogen status in rice//IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium, 2019: 7251-7253.
[34]	ALI A, IMRAN M. Evaluating the potential of red edge position (REP) of hyperspectral remote sensing data for real time estimation of LAI & chlorophyll content of kinnow mandarin (Citrus reticulata) fruit orchards. Scientia Horticulturae, 2020, 267: 109326. doi: 10.1016/j.scienta.2020.109326
[35]	XU T Y, WANG F M, XIE L L, YAO X P, ZHENG J Y, LI J L, CHEN S T. Integrating the textural and spectral information of UAV hyperspectral images for the improved estimation of rice aboveground biomass. Remote Sensing, 2022, 14(11): 2534. doi: 10.3390/rs14112534
[36]	CLARK M L, ROBERTS D A, CLARK D B. Hyperspectral discrimination of tropical rain forest tree species at leaf to crown scales. Remote Sensing of Environment, 2005, 96(3/4): 375-398. doi: 10.1016/j.rse.2005.03.009
[37]	刘秀英. 玉米生理参数及农田土壤信息高光谱监测模型研究[D]. 杨凌: 西北农林科技大学, 2016.
	LIU X Y. Monitoring models of physiological parameters of corn and farmland soil information based on hyper-spectral reflectance[D]. Yangling: Northwest A&F University, 2016. (in Chinese)
[38]	龚红菊, 姬长英. 基于图像处理技术的麦穗产量测量方法. 农业机械学报, 2007, 38(12): 116-119.
	GONG H J, JI C Y. Estimating yield of wheat spike with its texture features based on image processing technology. Transactions of the Chinese Society for Agricultural Machinery, 2007, 38(12): 116-119. (in Chinese)
[39]	郑恒彪. 水稻生育期及生长参数的近地面遥感监测研究[D]. 南京: 南京农业大学, 2018.
	ZHENG H B. Monitoring rice phenology and growth parameters using near-ground remote sensing platforms[D]. Nanjing: Nanjing Agricultural University, 2018. (in Chinese)
[40]	LU D. Aboveground biomass estimation using Landsat TM data in the Brazilian Amazon. International Journal of Remote Sensing, 2005, 26(12): 2509-2525. doi: 10.1080/01431160500142145
[41]	孙俊, 金夏明, 毛罕平, 武小红, 朱文静, 张晓东, 高洪燕. 基于高光谱图像光谱与纹理信息的生菜氮素含量检测. 农业工程学报, 2014, 30(10): 167-173.
	SUN J, JIN X M, MAO H P, WU X H, ZHU W J, ZHANG X D, GAO H Y. Detection of nitrogen content in lettuce leaves based on spectroscopy and texture using hyperspectral imaging technology. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(10): 167-173. (in Chinese)

植被指数Vegetation index	公式Formula
归一化差值植被指数Normalized difference vegetation index (NDVI)	(NIR-R)/(NIR+R)
归一化差异红边植被指数Normalized difference red-edge vegetation index (NDRE)	(NIR-RE)/(NIR+RE)
比值植被指数Ratio vegetation index (RVI)	NIR/R
差值植被指数Difference vegetation index (DVI)	NIR-R
红边土壤调节植被指数 Red-edge soil-adjusted vegetation index (RESAVI)	1.5*(NIR-RE)/(NIR+RE+0.5)
红边叶绿素指数Red-edge chlorophyll index (CIRE)	(NIR/RE)-1

纹理指数Texture index	公式Formula
NDTI (T1, T2)	(T1-T2)/(T1+T2)
RTI (T1, T2)	T1/T2
DTI (T1, T2)	T1-T2

指标 Indicator	取样数 Sample number	最小值 Minimum value	最大值 Maximum value	平均值 Mean value	标准差 SD	变异系数 C.V (%)
叶面积指数LAI	312	0.40	9.07	3.69	1.94	52.57
地上部生物量 AGB (t·hm^-2)	312	0.66	21.94	7.87	5.57	70.78
植株氮含量PNC (%)	312	0.56	3.88	1.93	0.66	34.20

植被指数 Vegetation index	叶面积指数 LAI	地上部生物量 AGB (t·hm^-2)	植株氮含量 PNC (%)
NDVI	0.49	0.37	0.08
NDRE	0.78	0.76	0.13
RVI	0.49	0.34	0.06
DVI	0.61	0.61	0.11
CIRE	0.80	0.75	0.12
RESAVI	0.77	0.74	0.13

纹理指数 Texture index	叶面积指数LAI			地上部生物量AGB (t·hm^-2)			植株氮含量PNC (%)
纹理指数 Texture index	λ1	λ2	R²	λ1	λ2	R²	λ1	λ2	R²
NDTI	MEA550	MEA790	0.59	MEA550	MEA790	0.48	VAR660	CON660	0.21
NDTI	ENT550	MEA790	0.55	ENT550	MEA790	0.45	ENT735	HOM660	0.17
RTI	MEA550	MEA790	0.56	MEA550	MEA790	0.40	COR550	SEC660	0.24
RTI	ENT550	MEA790	0.54	ENT550	MEA790	0.37	DIS660	ENT660	0.21
DTI	MEA550	MEA790	0.56	MEA550	MEA790	0.46	SEC735	ENT660	0.26
DTI	MEA790	CON735	0.51	MEA790	CON735	0.44	VAR660	CON660	0.22