基于无人机数码图像的机采棉脱叶率监测模型构建

doi:10.11963/cs20210003

摘要/Abstract

摘要：

【目的】脱叶率是评价机采棉脱叶催熟效果的重要依据。以无人机RGB图像为基础数据源,通过从RGB图像中提取14种可见光植被指数,建立快速、准确监测棉花脱叶率的模型,为机采棉适时采收提供理论和技术支持。【方法】设置不同棉花品种,通过采集不同脱叶剂浓度及喷施时间处理下的棉花脱叶率数据,并利用无人机采集冠层RGB图像,提取可见光植被指数,分析其与棉花脱叶率的相关关系,进而采用一元线性回归（Simple linear regression,SLR）、多元线性回归（Multivariate linear regression,MLR）和偏最小二乘法回归（Partial least square regression,PLSR）构建棉花脱叶率监测模型,并进行模型评价。【结果】不同处理下的棉花脱叶率有明显差异,脱叶率与不同可见光植被指数存在较好相关性,其中三角形绿度值（Triangular greenness index,TGI）与棉花脱叶率的相关性最高（r=0.81）。建模结果表明,SLR模型中,以TGI指数建模效果最好（决定系数0.66,均方根误差10.44%,相对均方根误差12.87%）;MLR模型中,以过蓝指数（Excess blue index,ExB）、绿叶指数（Green leaf index,GLI）、TGI和过绿指数（Excess green index,ExG）4个植被指数组合建立的模型效果最好,其决定系数为0.70,均方根误差为10.26%,相对均方根误差为12.65%。PLSR模型中,以ExB、GLI、TGI、ExG、综合植被指数2和综合植被指数1建立的模型精度更高,其决定系数为0.70,均方根误差为10.02%,相对均方根误差为12.22%。外部验证表明,各模型实测值与预测值间有较好的拟合关系。【结论】以MLR和PLSR方法建立的模型精度较高,拟合程度较好。从计算量及模型复杂程度角度考虑,通过MLR方法以ExB、GLI、TGI、ExG建立的棉花脱叶率监测模型,能够更好地监测棉花脱叶率。

关键词: 棉花; 脱叶率; 可见光植被指数; RGB图像; 无人机

Abstract:

[Objective] Defoliation rate is an important basis for evaluating defoliation and ripening effect of machine-picked cotton. In this study, 14 kinds of color vegetation index were extracted from RGB images to establish a fast and accurate estimation model for cotton defoliation rate, which provides a theoretical and technical basis for timely harvesting of machine-picked cotton. [Methode] Different cotton varieties were set up. The data of cotton defoliation rates of different defoliant concentrations and spraying times were collected, and the canopy RGB images were collected by unmanned arerial vehicle(UAV). The correlation between color vegetation index and cotton defoliation rate was analyzed. Then, the estimation model of cotton defoliation rates was constructed by using the methods of simple linear regression (SLR), multivariate linear regression (MLR) and partial least square regression (PLSR). Meanwhile, the model was evaluated. [Result] The results showed different cotton defoliation rates after different treatments. There was a remarkable correlation between the defoliation rate and the visible light vegetation index; especially, the correlation coefficient between the triangular greenness index (TGI) and cotton defoliation rate is up to 0.81. The results showed that the model based on TGI index was the best in the linear regression model, coefficient of determination (R²) = 0.66, root mean squared error (RMSE) = 10.44%, relative RMSE (rRMSE) = 12.87%; the model based on excess blue index (ExB), green leaf index (GLI), TGI and excess green index (ExG) was outstanding in the multiple linear regression models (R² = 0.70, RMSE = 10.26%, rRMSE = 12.65%). In the PLSR models, the one with ExB, GLI, TGI, ExG, Comprehensive 2 and Comprehensive 1, had higher accuracy, R²= 0.70, RMSE = 10.02%, rRMSE = 12.22%. The external verification showed that there was a good fitting relationship between the measured values and the predicted values of each model. [Conclusion] The model established by MLR and PLSR has high accuracy and great fitting degree. Therefore, considering of the weight computation and complexity, the cotton defoliation rate estimation model established with ExB, GLI, TGI and ExG has excellent performance.

Key words: cotton; defoliation rate of cotton leaf; color vegetation index; RGB image; unmanned aerial vehicle

马怡茹,吕新,祁亚琴,张泽,易翔,陈翔宇,鄢天荥,侯彤瑜. 基于无人机数码图像的机采棉脱叶率监测模型构建[J]. 棉花学报, 2021, 33(4): 347-359.

Ma Yiru,Lü Xin,Qi Yaqin,Zhang Ze,Yi Xiang,Chen Xiangyu,Yan Tianying,Hou Tongyu. Estimation of the defoliation rate of cotton based on unmanned aerial vehicle digital images[J]. Cotton Science, 2021, 33(4): 347-359.

0
/ / 推荐

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

链接本文: http://journal.cricaas.com.cn/Jweb_mhxb/CN/10.11963/cs20210003

http://journal.cricaas.com.cn/Jweb_mhxb/CN/Y2021/V33/I4/347

图/表 10

表1

表2

图1

图2

表3

表4

表5

图3

图4

图5

参考文献 41

[1]	中国国家统计局. 国家统计局关于2020年棉花产量的公告[EB/OL]. (2020-12-18)[2021-01-10]. http://www.stats.gov.cn/tjsj/zxfb/202012/t20201218_1810113.html.
	National Bureau of Statistics of China. Announcement of the National Bureau of Statistics on cotton production in 2020[EB/OL]. (2020-12-18)[2021-01-10]. http://www.stats.gov.cn/tjsj/zxfb/202012/t20201218_1810113.html.
[2]	周婷婷, 肖庆刚, 杜睿, 等. 我国棉花脱叶催熟技术研究进展[J/OL]. 棉花学报, 2020, 32(2): 170-184[2021-01-10]. https://doi.org/10.11963/1002-7807.ztthxq.20200303.
	Zhou Tingting, Xiao Qinggang, Du Rui, et al. Research advances on cotton harvest aids in China[J/OL]. Cotton Science, 2020, 32(2): 170-184[2021-01-10]. https://doi.org/10.11963/1002-7807.ztthxq.20200303.
[3]	李琼, 刘强, 王海标, 等. 脱叶剂及用量对不同熟性棉花的脱叶效果和种子发芽率影响[J]. 棉花科学, 2020, 42(3):45-49.
	Li Qiong, Liu Qiang, Wang Haibiao, et al. Effects of different defoliants on the defoliation and germination rate of cotton[J]. Cotton Sciences, 2020, 42(3):45-49.
[4]	陈兵, 张国蕾, 王静, 等. 机采棉等行距密植模式下棉花脱叶催熟效果研究[J/OL]. 西北农业学报, 2019, 28(10): 1594-1601[2021-01-10]. https://doi.org/0.7606/j.issn.1004-1389.2019.10.005.
	Chen Bing, Zhang Guolei, Wang Jing, et al. Effect of different defoliants on machine-picked cotton under equal spacing and dense planting model[J/OL]. Acta Agriculturae Boreali-Occidentalis Sinica, 2019, 28(10): 1594-1601[2021-01-10]. https://doi.org/0.7606/j.issn.1004-1389.2019.10.005.
[5]	阿不都卡地尔·库尔班, 陈平, 程强, 等. 喷施脱叶剂对不同停水时间机采棉脱叶效果及纤维品质的影响[J/OL]. 新疆农业大学学报, 2018, 41(3): 157-163[2021-01-10]. https://doi.org/10.3969/j.issn.1007-8614.2018.03.001.
	Abudukadier Kuerban, Chen Ping, Cheng Qiang, et al. Influences of spraying defoliant on defoliation effect and fiber quality of machine-picked cotton under different water-shutdown time[J/OL]. Journal of Xinjiang Agricultural University, 2018, 41(3): 157-163[2021-01-10]. https://doi.org/10.3969/j.issn.1007-8614.2018.03.001.
[6]	Maes W H, Steppe K. Perspectives for remote sensing with unmanned aerial vehicles in precision agriculture[J/OL]. Trends in Plant Science, 2019, 24(2): 152-164[2021-01-10]. https://doi.rog/10.1016/j.tplants.2018.11.007.
[7]	易秋香. 基于Sentinel-2多光谱数据的棉花叶面积指数估算[J/OL]. 农业工程学报, 2019, 35(16): 189-197[2021-01-10]. https://doi.org/10.11975/j.issn.1002-6819.2019.16.021.
	Yi Qiuxiang. Remote estimation of cotton LAI using Sentinel-2 multispectral data[J/OL]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(16): 189-197[2021-01-10]. https://doi.org/10.11975/j.issn.1002-6819.2019.16.021.
[8]	Maimaitijiang M, Ghulam A, Sidike P, et al. Unmanned aerial system (UAS)-based phenotyping of soybean using multi-sensor data fusion and extreme learning machine[J/OL]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 134:43-58. https://doi.org/10.1016/j.isprsjprs.2017.10.011.
[9]	李亚兵. 棉花早衰程度诊断数码图像数字化指标的研究[D]. 棉花学报, 2006, 32(3): 160-163.
	Li Yabing. Study on the parameters of digital images of diagnosis on cotton premature senescence grade[J]. Cotton Science, 2006, 32(3): 160-163.
[10]	王康丽. 基于无人机可见光和热红外图像的棉花冠层信息识别[D]. 北京: 中国农业科学院, 2019.
	Wang Kangli. Cotton canopy information recognition based on visible light and thermal infrared image of UAV[D]. Beijing: Chinese Academy of Agricultural Sciences, 2019.
[11]	Liang G Z, Dong C W, Hu B, et al. Prediction of moisture content for congou black tea withering leaves using image features and nonlinear method[J/OL]. Scientific Reports, 2018, 8(1):7854[2021-01-10]. https://doi.org/10.1038/s41598-018-26165-2.
[12]	Niu Y X, Zhang L Y, Zhang H H, et al. Estimating above-ground biomass of maize using features derived from UAV-based RGB imagery[J/OL]. Remote Sensing, 2019, 11(11): 1261[2021-01-10]. https://doi.org/10.3390/rs11111261.
[13]	Hunt E R, Cavigelli M, Daughtry C S T, et al. Evaluation of digital photography from model aircraft for remote sensing of crop biomass and nitrogen status[J/OL]. Precision Agriculture, 2005, 6(4): 359-378[2021-01-10]. https://doi.org/10.1007/s11119-005-2324-5.
[14]	Louhaichi M, Borman M M Johnson D E. Spatially located platform and aerial photography for documentation of grazing impacts on wheat[J/OL]. Geocarto International, 2001, 16(1): 65-70[2021-01-10]. https://doi.org/10.1080/10106040108542184.
[15]	Gitelson A A, Kaufman Y J, Stark R, et al. Novel algorithms for remote estimation of vegetation fraction[J/OL]. Remote Sensing of Environment, 2002, 80(1): 76-87[2021-01-10]. https://doi.org/10.1016/S0034-4257(01)00289-9.
[16]	Meyer G E, Hindman T W, Laksmi K. Machine vision detection parameters for plant species identification[C/OL]// Proceedings volume 3543: Precision agriculture and biological quality. Boston: International Society for Optics and Photonics, 1999: 327-335[2021-01-10]. https://doi.org/10.1117/12.336896.
[17]	Woebbecke D M, Meyer G E, von Bargen K, et al. Color indices for weed identification under various soil, residue,lighting conditions[J/OL]. Transactions of the American Society of Agricultural Engineers, 1995, 38(1): 259-269[2021-01-10]. https://doi.org/10.13031/2013.27838.
[18]	Mao W, Wang Y, Wang Y. Real-time detection of between-row weeds using machine vision[C/OL]// 2003 American Society of Agricultural and Biological Engineers(ASABE) Annual Meeting. St, Joseph: ASABE, 2003: 031004[2021-01-10]. https://doi.org/10.13031/2013.15381.
[19]	Neto J C. A combined statistical-soft computing approach for classification and mapping weed species in minimum-tillage systems[D]. Lincoln: The University of Nebraska-Lincoln, 2004. 2003.1225492.
[20]	Kataoka T, Kaneko T, Okamoto H, et al. Crop growth estimation system using machine vision[C/OL]// Proceedings 2003 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2003). New Jersey: IEEE 2003, 2b1079-b1083[2021-01-10]. https://doi.org/10.1109/AIM.2003.1225492.
[21]	Hague T, Tillett N D, Wheeler H. Automated crop and weed monitoring in widely spaced cereals[J/OL]. Precision Agriculture, 2006, 7(1): 21-32[2021-01-10]. https://doi.org/10.1007/s11119-005-6787-1.
[22]	Burgos-Artizzu X P, Ribeiro A, Guijarro M, et al. Real-time image processing for crop/weed discrimination in maize fields[J/OL]. Computers and Electronics in Agriculture, 2011, 75(2): 337-346[2021-01-10]. https://doi.org/10.1016/j.compag.2010.12.011.
[23]	Kawashima S, Nakatani M. An algorithm for estimating chlorophyll content in leaves using a video camera[J/OL]. Annals of Botany, 1998, 81(1): 49-54[2021-01-10]. https://doi.org/10.1006/anbo.1997.0544.
[24]	Hunt E R, Jr, Daughtry C S T, Eitel J U H, et al. Remote sensing leaf chlorophyll content using a visible band index[J/OL]. Agronomy Journal, 2011, 103(4): 1090-1099[2021-01-10]. https://doi.org/10.2134/agronj2010.0395.
[25]	Guijarro M, Pajares G, Riomoros I, et al. Automatic segmentation of relevant textures in agricultural images[J/OL]. Computers and Electronics in Agriculture, 2011, 75(1): 75-83[2021-01-10]. https://doi.org/10.1016/j.compag.2010.09.013.
[26]	Guerrero J M, Pajares G, Montalvo M, et al. Support vector machines for crop/weeds identification in maize fields[J/OL]. Expert Systems with Applications, 2012, 39(12): 11149-11155[2021-01-10]. https://doi.org/10.1016/j.eswa.2012.03.040.
[27]	高丽丽, 李淦, 康正华, 等. 脱叶剂对棉花叶片叶绿素荧光动力学参数的影响[J/OL]. 棉花学报, 2016, 28(4): 345-352[2021-01-10]. https://doi.org/10.11963/issn.1002-7807.201604005.
	Gao Lili, Li Gan, Kang Zhenghua, et al. Effect of defoliants on chlorophyll fluorescence of cotton leaves[J/OL]. Cotton Science, 2016, 28(4): 345-352[2021-01-10]. https://doi.org/10.11963/issn.1002-7807.201604005.
[28]	宋兴虎, 徐东永, 孙璐, 等. 在不同棉区噻苯隆和乙烯利用量及配比对脱叶催熟效果影响[J/OL]. 棉花学报, 2020, 32(3): 247-257[2021-01-10]. https://doi.org/10.11963/1002-7807.sxhlzh.20200512.
	Song Xinghu, Xu Dongyong, Sun Lu, et al. Effect of thidiazuron and ethylene use and ratio on defoliation ripening in different cotton area[J/OL]. Cotton Science, 2020, 32(3): 247-257[2021-01-10]. https://doi.org/10.11963/1002-7807.sxhlzh.20200512.
[29]	Putra B T W, Soni P. Enhanced broadband greenness in assessing chlorophyll a and b,carotenoid, nitrogen in Robusta coffee plantations using a digital camera[J/OL]. Precision Agriculture, 2018, 19(2): 238-256[2021-01-10]. https://doi.org/10.1007/s11119-017-9513-x.
[30]	Buchaillot M, Gracia-Romero A, Vergara-Diaz O, et al. Evaluating maize genotype performance under low nitrogen conditions using RGB UAV phenotyping techniques[J/OL]. Sensors, 2019, 19(8): 1815[2021-01-10]. https://doi.org/10.3390/s19081815.
[31]	江杰, 张泽宇, 曹强, 等. 基于消费级无人机搭载数码相机监测小麦长势状况研究[J/OL]. 南京农业大学学报, 2019(4): 622-631[2021-01-10]. https://doi.org/10.7685/jnau.201904065.
	Jiang Jie, Zhang Zeyu, Cao Qiang, et al. Use of a digital camera mounted on a consumer-grade unmanned aerial vehicle to monitor the growth status of wheat[J/OL]. Journal of Nanjing Agricultural University, 2019, 42(4): 622-631[2021-01-10]. https://doi.org/10.7685/jnau.201904065.
[32]	Che Y P, Wang Q, Xie Z W, et al. Estimation of maize plant height and leaf area index dynamics using an unmanned aerial vehicle with oblique and nadir photography[J/OL]. Annals of Botany, 2020, 126(4): 765-773[2021-01-10]. https://doi.org/10.1093/aob/mcaa097.
[33]	Lu N, Zhou J, Han Z X, et al. Improved estimation of aboveground biomass in wheat from RGB imagery and point cloud data acquired with a low-cost unmanned aerial vehicle system[J/OL]. Plant Methods, 2019, 15: 17[2021-01-10]. https://doi.org/10.1186/s13007-019-0402-3.
[34]	杨俊, 丁峰, 陈晨, 等. 小麦生物量及产量与无人机图像特征参数的相关性[J/OL]. 农业工程学报, 2019, 35(23): 104-110[2021-01-10]. https://doi.org/10.11975/j.issn.1002-6819.2019.23.013.
	Yang Jun, Ding Feng, Chen Chen, et al. Correlation of wheat biomass and yield with UAV image characteristic parameters[J/OL]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(23): 104-110[2021-01-10]. https://doi.org/10.11975/j.issn.1002-6819.2019.23.013.
[35]	刘畅. 无人机影像‘图-谱’信息融合在作物氮素诊断中的应用研究[D]. 西安: 西安科技大学, 2018.
	Liu Chang. Unmanned aerial vehicle (UAV) based 'image and spectrum' information fusion for nitrogen diagnosis in winter wheat[D]. Xi’an: Xi’an University of Science and Technology, 2018.
[36]	Fathipoor H, Arefi H, Shah-Hosseini R, et al. Corn forage yield prediction using unmanned aerial vehicle images at mid-season growth stage[J/OL]. Journal of Applied Remote Sensing, 2019, 13(3): 034503[2021-01-10]. https://doi.org/10.1117/1.JRS.13.034503.
[37]	Tao H L, Feng H K, Xu L J, et al. Estimation of crop growth parameters using UAV-based hyperspectral remote sensing data[J/OL]. Sensors, 2020, 20(5): 1296[2021-01-10]. https://doi.org/10.3390/s20051296.
[38]	李天胜, 崔静, 王海江, 等. 基于高光谱特征波长的冬小麦水分含量估测模型[J/OL]. 新疆农业科学, 2019, 56(10): 1772-1782[2021-01-10]. https://doi.org/10.6048/j.issn.1001-4330.2019.10.002.
	Li Tiansheng, Cui jing, Wang Haijiang, et al. Study on water content estimation model of winter wheat based on hyperspectral characteristic wavelength[J/OL]. Xinjiang Agricultural Sciences, 2019, 56(10): 1772-1782[2021-01-10]. https://doi.org/10.6048/j.issn.1001-4330.2019.10.002.
[39]	Kim D W, Yun H S, Jeong S J, et al. Modeling and testing of growth status for Chinese cabbage and white radish with UAV-based RGB imagery[J/OL]. Remote Sensing, 2018, 10(4): 563[2021-01-10]. https://doi.org/10.3390/rs10040563.
[40]	田明璐, 班松涛, 常庆瑞, 等. 基于低空无人机成像光谱仪影像估算棉花叶面积指数[J/OL]. 农业工程学报, 2016, 32(21): 102-108[2021-01-10]. https://doi.org/10.11975/j.issn.1002-6819.2016.21.014.
	Tian Minglu, Ban Songtao, Chang Qingrui, et al. Use of hyperspectral images from UAV-based imaging spectroradiometer to estimate cotton leaf area index[J/OL]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(21): 102-108[2021-01-10]. https://doi.org/10.11975/j.issn.1002-6819.2016.21.014.
[41]	李梅, 徐明珠, 何勇, 等. 马铃薯晚疫病叶片SPAD值的高光谱研究[J/OL]. 农机化研究, 2016, 38(2): 136-142[2021-01-10]. https://doi.org/10.3969/j.issn.1003-188X.2016.12.027.
	Li Mei, Xu Mingzhu, He Yong, et al. SPAD measurement of potato leaves infected late blight based on hyper-spectral [J/OL]. Journal of Agricultural Mechanization Research, 2016, 38(2): 136-142[2021-01-10]. https://doi.org/10.3969/j.issn.1003-188X.2016.12.027.

植被指数 Vegetation index	公式 Formula	发明者 Inventor
归一化绿红差值植被指数Normalized green-red difference index (NGRDI)	NGRDI=	Hunt等^[13]
绿叶指数Green leaf index (GLI)	GLI=	Louhaichi等^[14]
可见光抗大气指数Visible atmospherically resistant index (VARI)	VARI=	Gltelson等^[15]
过红指数Excess red index (ExR)	ExR=1.4r-g	Meyer等^[16]
过绿指数Excess green index (ExG)	ExG=2g-r-b	Woebbecke等^[17]
过蓝指数Excess blue index (ExB)	ExB=	Mao等^[18]
过绿过红差分指数Excess green minus excess red index (ExGR)	ExGR=ExG-ExR	Neto等^[19]
植被颜色提取指数Color index of vegetation (CIVE)	CIVE=0.411R-0.811G+0.385B+18.787 45	Kataoka等^[20]
植被因子指数Vegetative index (VEG)	VEG=（α=0.667）	Hague等^[21]
改进过绿指数Modified excess green index (MExG)	MExG=1.262G-0.884R-0.311B	Burgos-Artizzu等^[22]
Kawashima指数Kawashima index (IKAW)	IKAW=	Kawashima等^[23]
三角绿色指数Triangular greenness index (TGI)	TGI=0.5[0.19（R-G）-0.12（R-B）]	Hunt等^[24]
综合植被指数1 Comprehensive 1 (COM1)	COM1=ExG+CIVE+ExGR+VEG	Guijarro等^[25]
综合植被指数2 Comprehensive 2 (COM2)	COM2=0.36ExG+0.47CIVE+0.17VEG	Guerrero等^[26]

样本 Sample	样本数量 Number	脱叶率统计指标Statistical indexes of defoliation rate
样本 Sample	样本数量 Number	最大值 Maximum/%	最小值 Minimum/%	平均值 Mean/%	标准偏差 Standard deviation/%	变异系数 Coefficient of variation/%
总样本 Total sample	48	100.00	29.16	81.97	17.94	21.89
训练集 Training set	36	99.78	29.16	83.75	9.38	11.20
验证集 Validation set	12	100.00	49.73	82.16	18.60	22.64

植被指数 Vegetation index	训练集Training set			验证集Validation set
植被指数 Vegetation index	R²	RMSE/%	rRMSE/%	R²	RMSE/%	rRMSE/%
GLI	0.62	10.99	13.55	0.93	5.02	6.11
VARI	0.21	15.86	19.55	0.53	13.37	16.27
ExR	0.32	14.74	18.17	0.72	10.36	12.61
ExG	0.59	11.37	14.01	0.90	6.01	7.32
ExB	0.62	10.99	13.55	0.89	6.57	8.00
ExGR	0.48	12.83	15.81	0.84	7.73	9.41
CIVE	0.36	14.50	17.87	0.74	9.95	12.11
VEG	0.45	13.24	16.32	0.84	7.86	9.57
MExG	0.33	14.61	16.32	0.73	10.08	12.27
IKAW	0.16	16.32	20.12	0.49	13.95	16.98
TGI	0.66	10.44	12.87	0.93	5.28	6.43
COM1	0.54	12.14	14.96	0.87	6.93	8.44
COM2	0.59	11.36	14.00	0.91	5.97	7.27

植被指数 Vegetation index	训练集Training set			验证集Validation set
植被指数 Vegetation index	R²	RMSE/%	rRMSE/%	R²	RMSE/%	rRMSE/%
GLI+TGI	0.66	10.59	13.05	0.92	5.58	6.79
ExB+GLI+TGI	0.69	10.28	12.67	0.94	4.88	5.94
ExB+GLI+TGI+ExG	0.70	10.26	12.65	0.93	5.19	6.32
ExB+GLI+TGI+ExG+COM2	0.70	10.43	12.86	0.94	4.75	5.78
ExB+GLI+TGI+ExG+COM2+COM1	0.70	10.60	13.07	0.93	5.31	6.46
ExB+GLI+TGI+ExG+COM2+COM1+ExGR	0.71	10.67	13.15	0.93	5.20	6.33

植被指数 Vegetation index	训练集Training set			验证集Validation set
植被指数 Vegetation index	R²	RMSE/%	rRMSE/%	R²	RMSE/%	rRMSE/%
GLI+TGI	0.66	10.59	12.92	0.92	5.58	6.79
ExB+GLI+TGI	0.68	9.84	12.34	0.94	4.97	6.05
ExB+GLI+TGI+ExG	0.68	10.11	12.34	0.93	5.00	6.10
ExB+GLI+TGI+ExG+COM2	0.69	10.11	12.23	0.93	5.01	6.10
ExB+GLI+TGI+ExG+COM2+COM1	0.70	10.02	12.22	0.93	5.01	6.11
ExB+GLI+TGI+ExG+COM2+COM1+ExGR	0.70	10.01	12.22	0.93	5.02	6.11