基于无人机多光谱和热红外影像信息融合的小麦白粉病监测

doi:10.3864/j.issn.0578-1752.2022.05.005

摘要/Abstract

摘要：

【目的】白粉病严重危害小麦生长及制约产量形成,确立实时监测小麦白粉病的多源数据融合方法,为精确防控及保证国家粮食安全提供技术支撑。【方法】在小麦开花和灌浆期,使用同时搭载多光谱仪和热成像仪的六旋翼无人机作为遥感数据获取平台,通过ENVI软件从小麦白粉病遥感影像中提取植被指数、纹理特征以及冠层温度信息,进而利用多元线性回归（MLR）、后向传播神经网络（BP）、随机森林（RF）、极限学习机（ELM）算法将植被指数（VIs）、纹理特征（TFs）和温度特征（T）进行结合,以构建小麦白粉病病情指数的监测模型。【结果】无论是单数据源建模,还是多数据源建模,随机森林（RF）的精度均高于其他模型;3种数据源中植被指数的RF模型（VIs-RF,R²=0.667,RMSE =5.712,RPD=1.572）更适宜白粉病监测,其次是温度特征（T-RF,R²=0.559,RMSE =6.563,RPD=1.430）,而纹理特征（TFs-RF,R²=0.495,RMSE =7.014,RPD=1.348）效果最差;多数据源协同建模间比较,RF协同植被指数和纹理特征的模型R²为0.701（VIs&TFs-RF,R²=0.701,RMSE =5.308,RPD=1.724）,仅比VIs-RF模型R²提升5.101%,RMSE降低7.073%,RPD提高9.672%,而RF协同植被指数和温度特征模型（VIs&T-RF）以及协同3种数据源模型（VIs&TFs&T-RF）的精度分别为R²=0.750,RMSE=4.704,RPD=1.912和R²=0.820,RMSE =4.677,RPD=1.996,较VIs-RF模型R²分别提升12.453%和23.181%,RMSE分别降低17.640%和18.113%,RPD分别提高21.667%和26.981%。同时对不同模型进行10折交叉验证,进一步证实了RF模型在多数据源融合建模中性能稳定,估算效果最好。【结论】采用多数据源协同建模能够提升小麦白粉病遥感监测精度,研究结果为实现大面积高精度遥感监测作物病害状况提供了思路与方法。

关键词: 小麦白粉病, 无人机, 机器学习, 信息融合, 遥感监测

Abstract:

【Objective】Wheat growth and yield can be seriously affected by powdery mildew. Establishing the multi-source data fusion method for real-time monitoring of powdery mildew of wheat could provide technical support for accurate prevention and control of diseases and guaranteeing national food security. 【Method】During the wheat flowering and filling period, a six-rotor UAV equipped with multi-spectral sensor and thermal imager was used as a remote sensing data acquisition platform to obtain remote sensing images of different degrees of wheat powdery mildew. Then, vegetation index (VIs), texture feature (TFs) and temperature feature (T) were extracted from multi-spectral and thermal infrared images of different disease degrees on a low-altitude drone platform by ENVI software. Finally, the wheat powdery mildew disease index model were built by multiple linear regression (MLR), back propagation neural network (BP), random forest (RF) and extreme learning machine (ELM). 【Result】The precision of the RF model based on both single and multiple data sources was higher than that of the other models. Among the three data sources of the RF model, the vegetation indices (VIs-RF, R ²= 0.667, RMSE=5.712, RPD=1.572) were the most suitable for powdery mildew monitoring, followed by the temperature feature (T-RF, R ²= 0.559, RMSE=6.563, RPD=1.430) and texture features (TFs-RF, R ² = 0.495, RMSE=7.014, RPD=1.348). When combining multiple data sources, a precision for the RF model combining vegetation indices and texture features (VIs & TFs-RF) of 0.701 could be obtained, which was 5.101% higher than that of the VIs-RF model, while RMSE was 7.073% lower and RPD was 9.672% higher, whereas the precision parameters of the RF model combining vegetation indices and the temperature feature (VIs & T-RF) were R ² = 0.750, RMSE = 4.704, RPD = 1.912. For all three remote sensing data sources (VIs & TFs & T-RF), the following accuracies resulted: R ² = 0.820, RMSE = 4.677, RPD=1.996. As compared to the VIs-RF model, R ² improved by 12.453%, RMSE by 17.640% and RPD by 21.667% for the (VIs & T-RF) model, whereas for the three remote sensing sources, R ²improved by 23.181%, RMSE by 18.113% and RPD by 26.981%. At the same time, 10 fold cross validation of different models was carried out, which further confirmed that RF model had stable performance and good estimation results in multi-data source fusion modeling. 【Conclusion】 The precision of wheat powdery mildew monitoring could be improved by using multi-data-sources collaborative ML modeling. This research provided technical support for large-area and high-precision remote sensing of crop diseases.

Key words: powdery mildew, UAV, machine learning, information fusion, remote sensing monitoring

冯子恒,宋莉,张少华,井宇航,段剑钊,贺利,尹飞,冯伟. 基于无人机多光谱和热红外影像信息融合的小麦白粉病监测[J]. 中国农业科学, 2022, 55(5): 890-906.

FENG ZiHeng,SONG Li,ZHANG ShaoHua,JING YuHang,DUAN JianZhao,HE Li,YIN Fei,FENG Wei. Wheat Powdery Mildew Monitoring Based on Information Fusion of Multi-Spectral and Thermal Infrared Images Acquired with an Unmanned Aerial Vehicle[J]. Scientia Agricultura Sinica, 2022, 55(5): 890-906.

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.0578-1752.2022.05.005

https://www.chinaagrisci.com/CN/Y2022/V55/I5/890

图/表 11

图1

表1

表2

图2

表3

图3

图4

表4

表5

图5

表6

参考文献 62

[1]	张凝, 杨贵军, 赵春江, 张竞成, 杨小冬, 潘瑜春, 黄文江, 徐波, 李明, 朱西存, 李振海. 作物病虫害高光谱遥感进展与展望. 遥感学报, 2021, 25(1):403-422.
	ZHANG N, YANG G J, ZHAO C J, ZHANG J C, YANG X D, PAN Y C, HUANG W J, XU B, LI M, ZHU X C, LI Z H. Progress and prospects of hyperspectral remote sensing technology for crop diseases and pests. National Remote Sensing Bulletin, 2021, 25(1):403-422. (in Chinese)
[2]	黄文江, 师越, 董莹莹, 叶回春, 邬明权, 崔贝, 刘林毅. 作物病虫害遥感监测研究进展与展望. 智慧农业, 2019, 1(4):1-11.
	HUANG W J, SHI Y, DONG Y Y, YE H C, WU M Q, CUI B, LIU L Y. Progress and prospects of crop diseases and pests monitoring by remote sensing. Smart Agriculture, 2019, 1(4):1-11. (in Chinese)
[3]	SHI Y, HUANG W J, GONZALEZ-MORENO P, LUKE B, DONG Y Y, ZHENG Q, MA H Q, LIU L Y. Wavelet-based rust spectral feature set (WRSFs): A novel spectral feature set based on continuous wavelet transformation for tracking progressive host- pathogen interaction of yellow rust on wheat. Remote Sensing, 2018, 10(4):525. doi: 10.3390/rs10040525
[4]	CHEN T T, ZHANG J L, CHEN Y, WAN S B, ZHANG L. Detection of peanut leaf spots disease using canopy hyperspectral reflectance. Computers and Electronics in Agriculture, 2019, 156:677-683. doi: 10.1016/j.compag.2018.12.036
[5]	FRANCESCHINI M H D, BARTHOLOMEUS H, APELDOORN D F, SUOMALAINEN J, KOOISTRA L. Feasibility of Unmanned Aerial Vehicle optical imagery for early detection and severity assessment of late blight in potato. Remote Sensing, 2019, 11(3):224. doi: 10.3390/rs11030224
[6]	MAHLEIN AK, RUMPF T, WELKE P, DEHNE HW, PLUMER L, STEINER U, OERKE E C. Development of spectral indices for detecting and identifying plant diseases. Remote Sensing of Environment, 2013, 128:21-30. doi: 10.1016/j.rse.2012.09.019
[7]	FENG W, SHEN W Y, HE L, DUAN J Z, GUO B B, LI Y X, WANG C Y, GUO T C. Improved remote sensing detection of wheat powdery mildew using dual-green vegetation indices. Precision Agriculture, 2016, 17(5):608-627. doi: 10.1007/s11119-016-9440-2
[8]	NAIDU R A, PERRY E M, PIERCE F J, MEKURIA T. The potential of spectral reflectance technique for the detection of Grapevine leafroll-associated virus-3 in two red-berried wine grape cultivars. Computers and Electronics in Agriculture, 2009, 66(1):38-45. doi: 10.1016/j.compag.2008.11.007
[9]	王利民, 刘佳, 杨福刚, 杨玲波, 姚保民, 高建孟. 基于GF-1/WFV数据的冬小麦条锈病遥感监测. 农业工程学报, 2017, 33(20):153-160.
	WANG L M, LIU J, YANG F G, YANG L B, YAO B M, GAO J M. Winter wheat stripe rust remote sensing monitoring based on GF-1/WFV data. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(20):153-160. (in Chinese)
[10]	白雪冰, 余建树, 傅泽田, 张领先, 李鑫星. 可见光谱图像联合区间的黄瓜白粉病分割与检测. 光谱学与光谱分析, 2019, 39(11):3592-3598.
	BAI X B, YU J S, FU Z T, ZHANG L X, LI X X. Segmentation and detection of cucumber powdery mildew based on visible spectrum and image processing. Spectroscopy and Spectral Analysis, 2019, 39(11):3592-3598. (in Chinese)
[11]	ASHOURLOO D, AGHIGHI H, MATKAN A A, MOBASHERI M R, RAD A M. An investigation into machine learning regression techniques for the leaf rust disease detection using hyperspectral measurement. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016, 9:1-8.
[12]	沈文颖, 李映雪, 冯伟, 张海艳, 张元帅, 谢迎新, 郭天财. 基于因子分析-BP神经网络的小麦叶片白粉病反演模型. 农业工程学报, 2015, 31(22):183-190.
	SHEN W Y, LI Y X, FENG W, ZHANG H Y, ZHANG Y S, XIE Y X, GUO T C. Inversion model for severity of powdery mildew in wheat leaves based on factor analysis-BP neural network. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(22):183-190. (in Chinese)
[13]	郭伟, 朱耀辉, 王慧芳, 张娟, 董萍, 乔红波. 基于无人机高光谱影像的冬小麦全蚀病监测模型研究. 农业机械学报, 2019, 50(9):162-169.
	GUO W, ZHU Y H, WANG H F, ZHANG J, DONG P, QIAO H B. Monitoring model of winter wheat take-all based on UAV hyperspectral imaging. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(9):162-169. (in Chinese)
[14]	简俊凡, 何宏昌, 王晓飞, 李月. 农作物病虫害遥感监测综述. 测绘通报, 2018(9):24-28.
	JIAN J F, HE H C, WANG X F, LI Y. Review of remote sensing monitoring of crop pests and diseases. Bulletin of Surveying and Mapping, 2018(9):24-28. (in Chinese)
[15]	刘畅, 杨贵军, 李振海, 汤伏全, 王建雯, 张春兰, 张丽妍. 融合无人机光谱信息与纹理信息的冬小麦生物量估测. 中国农业科学, 2018, 51(16):3060-3073.
	LIU C, YANG G J, LI Z H, TANG F Q, WANG J W, ZHANG C L, ZHANG L Y. Biomass estimation in winter wheat by UAV spectral information and texture information fusion. Scientia Agricultura Sinica, 2018, 51(16):3060-3073. (in Chinese)
[16]	梁栋, 刘娜, 张东彦, 赵晋陵, 林芬芳, 黄林生, 张庆, 丁玉婉. 利用成像高光谱区分冬小麦白粉病与条锈病(英文). 红外与激光工程, 2017, 46(1):50-58.
	LIANG D, LIU N, ZHANG D Y, ZHAO J L, LIN F F, HUANG L S, ZHANG Q, DING Y W. Discrimination of powdery mildew and yellow rust of winter wheat using high resolution hyper-spectra and imageries. Infrared and Laser Engineering, 2017, 46(1):50-58. (in Chinese)
[17]	张东彦, 张竞成, 朱大洲, 王纪华, 罗菊花, 赵晋陵, 黄文江. 小麦叶片胁迫状态下的高光谱图像特征分析研究. 光谱学与光谱分析, 2011, 31(4):1101-1105.
	ZHANG D Y, ZHANG J C, ZHU D Z, WANG J H, LUO J H, ZHAO J L, HUANG W J. Investigation of the hyperspectral image characteristics of wheat leaves under different stress. Spectroscopy and Spectral Analysis, 2011, 31(4):1101-1105. (in Chinese)
[18]	ZHENG H B, CHENG T, ZHOU M, LI D, YAO X, TIAN Y, CAO W, ZHU Y. Improved estimation of rice aboveground biomass combining textural and spectral analysis of UAV imagery. Precision Agriculture, 2018, 20:1-19. doi: 10.1007/s11119-018-09627-5
[19]	GUO A T, HUANG W J, DONG Y Y, YE H Q, MA H Q, LIU B, WU W B, REN Y, RUAN C, GENG Y. Wheat yellow rust detection using UAV-based hyperspectral technology. Remote Sensing, 2021, 13(1):123. (in Chinese) doi: 10.3390/rs13010123
[20]	杨成娅, 张艳, 赵明珠, 朱应燕. 基于红外热像的农作物早期病害检测识别技术的研究进展. 激光杂志, 2020, 41(6):1-4.
	YANG C Y, ZHANG Y, ZHAO M Z, ZHU Y Y. Research progress of infrared thermal imaging in early disease detection of crops. Laser Journal, 2020, 41(6):1-4. (in Chinese)
[21]	刘又夫, 肖德琴, 刘亚兰, 钟伯平, 周志艳. 褐飞虱诱导的水稻冠层热图像温度特征变异评估方法. 农业机械学报, 2020, 51(5):165-172.
	LIU Y F, XIAO D Q, LIU Y L, ZHONG B P, ZHOU Z Y. Temperature eigenvalues evaluation method of rice canopy thermal image induced by brown rice planthopper. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(5):165-172. (in Chinese)
[22]	朱文静, 陈华, 李林, 魏新华, 毛罕平, SPANER D. 基于红外热成像边缘检测算法的小麦叶锈病分级研究. 农业机械学报, 2019, 50(4):36-41, 48.
	ZHU W J, CHEN H, LI L, WEI X H, MAO H P, SPANER D. Grading of wheat leaf rust based on edge detection of infrared thermal imaging. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(4):36-41, 48. (in Chinese)
[23]	MAIMAITIJIANG M, SAGAN V, SIDIKE P, HARTING S, FRITSCHI F B. Soybean yield prediction from UAV using multimodal data fusion and deep learning. Remote Sensing of Environment, 2019, 237:111599. doi: 10.1016/j.rse.2019.111599
[24]	赵叶, 竞霞, 黄文江, 董莹莹, 李存军. 日光诱导叶绿素荧光与反射率光谱数据监测小麦条锈病严重度的对比分析. 光谱学与光谱分析, 2019, 39(9):2739-2745.
	ZHAO Y, JING X, HUANG W J, DONG Y Y, LI C J. Comparison of sun-induced chlorophyll fluorescence and reflectance data on estimating severity of wheat stripe rust. Spectroscopy and Spectral Analysis, 2019, 39(9):2739-2745. (in Chinese)
[25]	陈思媛, 竞霞, 董莹莹, 刘良云. 基于日光诱导叶绿素荧光与反射率光谱的小麦条锈病探测研究. 遥感技术与应用, 2019, 34(3):511-520.
	CHEN S Y, JING X, DONG Y Y, LIU L Y. Detection of wheat stripe rust using solar-induced chlorophyll fluorescence and reflectance spectral indices. Remote Sensing Technology and Application, 2019, 34(3):511-520. (in Chinese)
[26]	刘林毅, 黄文江, 董莹莹, 杜小平, 马慧琴. 基于概率模型的冬小麦白粉病监测研究. 中国农业信息, 2018, 30(1):79-92.
	LIU L Y, HUANG W J, DONG Y Y, DU X P, MA H Q. Wheat powdery mildew monitoring using probabilistic model. China Agricultural Informatics, 2018, 30(1):79-92. (in Chinese)
[27]	GITELSON A A, KAUFMAN Y J, MERZLYAK M N. Use of a green channel in remote sensing of global vegetation from EOS-MODIS. Remote Sensing of Environment, 1996, 58(3):289-298. doi: 10.1016/S0034-4257(96)00072-7
[28]	FILELLA I, SERRANO L, SERRA J, PENUELAS J. Evaluating wheat nitrogen status with canopy reflectance indices and discriminant analysis. Crop Science, 1995, 35(5):1400-1405. doi: 10.2135/cropsci1995.0011183X003500050023x
[29]	GITELSON A A, MERZLYAK M N, CHIVKUNOVA O B. Optical properties and nondestructive estimation of anthocyanin content in plant leaves. Photochemistry and Photobiology, 2001, 74(1):38-45. doi: 10.1562/0031-8655(2001)074<0038:OPANEO>2.0.CO;2
[30]	WANG Z J, WANG J H, LIU L Y, HUANG W J, ZHAO C J, WANG C Z. Prediction of grain protein content in winter wheat (Triticum aestivum L.) using plant pigment ratio (PPR). Field Crops Research, 2004, 90(2) : 311-321. doi: 10.1016/j.fcr.2004.04.004
[31]	ZARCO-TEJADA P J, BERJÓN A, LÓPEZ-LOZANO R, MILLER J R, MARTÍN P, CACHORRO V, GONZÁLEZ M R, FRUTOS A D. Assessing vineyard condition with hyperspectral indices: Leaf and canopy reflectance simulation in a row-structured discontinuous canopy. Remote Sensing of Environment, 2005, 99(3):271-287. doi: 10.1016/j.rse.2005.09.002
[32]	HABOUDANE D, MILLER J R, TREMBLAY N, ZARCO-TEJADA P J, DEXTRAZE L. Integrated narrow-band vegetation indices for prediction of crop chlorophyll content for application to precision agriculture. Remote Sensing of Environment, 2002, 81(2):416-426. doi: 10.1016/S0034-4257(02)00018-4
[33]	NAIDU R A, PERRY E M, PIERCE F J, MEKURIA T. The potential of spectral reflectance technique for the detection of Grapevine leafroll-associated virus-3 in two red-berried wine grape cultivars. Computers and Electronics in Agriculture, 2009, 66(1):38-45. doi: 10.1016/j.compag.2008.11.007
[34]	MERZLYAK M N, GITELSON A A, CHIVKUNOVA O B, RAKITIN V Y. Non-destructive optical detection of pigment changes during leaf senescence and fruit ripening. Physiologia Plantarum, 1999, 106(1):135-141. doi: 10.1034/j.1399-3054.1999.106119.x
[35]	HARALICK R M, SHANMUGAM K, DINSTEIN I. Textural features for image classification. Studies in Media and Communication, 1973, 3(6):610-621.
[36]	VANI K, POONGODI S, HARIKRISHNA B. K-Means cluster based leaf disease identification in cotton plants. Indian Journal of Public Health Research and Development, 2018, 9(10):1117.
[37]	LI S Y, YUAN F, ATA-UI-KARIM S T, ZHENG H B, CAO Q. Combining color indices and textures of UAV-based digital imagery for rice LAI estimation. Remote Sensing, 2019, 11(15):1763. doi: 10.3390/rs11151763
[38]	XU J X, MA J, TANG Y N, WU W X. Estimation of sugarcane yield using a machine learning approach based on UAV-LiDAR data. Remote Sensing, 2020, 12(17):2823. doi: 10.3390/rs12172823
[39]	FU Z P, JIANG J, GAO Y, KRIENKE B. Wheat growth monitoring and yield estimation based on multi-rotor unmanned aerial vehicle. Remote Sensing, 2020, 12(3):508. doi: 10.3390/rs12030508
[40]	HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: Theory and applications. Neurocomputing, 2006, 70(1/3):489-501. doi: 10.1016/j.neucom.2005.12.126
[41]	孙立影, 杨晨, 赵海士, 常志勇. 基于极限学习机的遥感地球化学反演模型. 吉林大学学报(地球科学版), 2020, 50(6):1929-1938.
	SUN L Y, YANG C, ZHAO H S, CHANG Z Y. Remote sensing geochemical inversion model by using extreme learning machine. Journal of Jilin University (Earth Science Edition), 2020, 50(6):1929-1938. (in Chinese)
[42]	王鑫梅, 张劲松, 孟平, 杨洪国, 孙圣. 基于无人机遥感影像的核桃冠层氮素含量估算. 农业机械学报, 2021, 52(2):178-187.
	WANG X M, ZHANG J S, MENG P, YANG H G, SUN S. Estimation of nitrogen content in walnut canopy based on UAV remote sensing image. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(2):178-187. (in Chinese)
[43]	BREIMAN L. Random forest. Machine Learning, 2001, 45:5-32. doi: 10.1023/A:1010933404324
[44]	刘冀, 张特, 魏榕, 张茜, 刘艳丽, 董晓华. 基于随机森林偏差校正的农业干旱遥感监测模型研究. 农业机械学报, 2020, 51(7):170-177.
	LIU J, ZHANG T, WEI R, ZHANG Q, LIU Y L, DONG X H. Development of agricultural drought monitoring model using remote sensing based on bias-correcting random forest. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(7):170-177. (in Chinese)
[45]	冯伟, 王晓宇, 宋晓, 贺利, 王永华, 郭天财. 基于冠层反射光谱的小麦白粉病严重度估测. 作物学报, 2013, 39(8):1469-1477.
	FENG W, WANG X Y, SONG X, HE L, WANG Y H, GUO T C. Estimation of severity level of wheat powdery mildew based on canopy spectral reflectance. Acta Agronomica Sinica, 2013, 39(8):1469-1477. (in Chinese)
[46]	沈文颖, 冯伟, 李晓, 王晓宇, 贺利, 郭天财, 李映雪. 基于叶片高光谱特征的小麦白粉病严重度估算模式. 麦类作物学报, 2015, 35(1):129-137.
	SHEN W Y, FENG W, LI X, WANG X Y, HE L, GUO T C, LI Y X. Estimation model of wheat powdery mildew severity based on leaves hyperspectral characteristics. Journal of Triticeae Crops, 2015, 35(1):129-137. (in Chinese)
[47]	GRAEFF S, LINK J, CLAUPEIN W. Identification of powdery mildew (Erysiphe graminis sp. tritici) and take-all disease (Gaeumannomyces graminis sp. tritici) in wheat (Triticum aestivum L.) by means of leaf reflectance measurements. Central European Journal of Biology, 2006, 1(2):275-288.
[48]	PEÑUELAS J, FILELLA I. Visible and near-infrared reflectance techniques for diagnosing plant physiological status. Trends in Plant Science, 1998, 3(4):151-156. doi: 10.1016/S1360-1385(98)01213-8
[49]	MAIRE G L, FRANÇOIS C, DUFRÊNE E. Towards universal broad leaf chlorophyll indices using PROSPECT simulated database and hyperspectral reflectance measurements. Remote Sensing of Environment, 2004, 89(1):1-28. doi: 10.1016/j.rse.2003.09.004
[50]	白宗璠, 竞霞, 张腾, 董莹莹. MDBPSO算法优化的全波段光谱数据协同冠层SIF监测小麦条锈病. 作物学报, 2020, 46(8):1248-1257. doi: 10.3724/SP.J.1006.2020.01004
	BAI Z F, JING X, ZHANG T, DONG Y Y. Canopy SIF synergize with total spectral reflectance optimized by the MDBPSO algorithm to monitor wheat stripe rust. Acta Agronomica Sinica, 2020, 46(8):1248-1257. (in Chinese) doi: 10.3724/SP.J.1006.2020.01004
[51]	李鑫星, 朱晨光, 白雪冰, 毛富焕, 傅泽田, 张领先. 基于可见光谱和支持向量机的黄瓜叶部病害识别方法研究. 光谱学与光谱分析, 2019, 39(7):2250-2256.
	LI X X, ZHU C G, BAI X B, MAO F H, BO Z T, ZHANG L X. Recognition method of leaves based on visual spectrum and support vector machine. Spectroscopy and Spectral Analysis, 2019, 39(7):2250-2256. (in Chinese)
[52]	GUO A T, HUANG W J, YE H C, DONG Y Y, MA H Q, REN Y, RUAN C. Identification of wheat yellow rust using spectral and texture features of hyperspectral images. Remote Sensing, 2020, 12(9):1419. doi: 10.3390/rs12091419
[53]	CAO W X, QIAO Z Y, GAO Z Y, LU S H, TIAN F. Use of unmanned aerial vehicle imagery and a hybrid algorithm combining a watershed algorithm and adaptive threshold segmentation to extract wheat lodging. Physics and Chemistry of the Earth, 2021(6):103016.
[54]	CHANDEL A K, KHOT L R, YU L X. Alfalfa (Medicago sativa L.) crop vigor and yield characterization using high-resolution aerial multispectral and thermal infrared imaging technique. Computers and Electronics in Agriculture, 2021, 182(2):105999. doi: 10.1016/j.compag.2021.105999
[55]	FRANCESCONI S, HARFOUCHE A, MAESANO M, BALESTRA G M. UAV-based thermal, RGB imaging and gene expression analysis allowed detection of fusarium head blight and gave new insights into the physiological responses to the disease in durum wheat. Frontiers in Plant Science, 2021, 12:628575. doi: 10.3389/fpls.2021.628575
[56]	马慧琴, 黄文江, 景元书. 遥感与气象数据结合预测小麦灌浆期白粉病. 农业工程学报, 2016, 32(9):165-172.
	MA H Q, HUANG W J, JING Y S. Wheat powdery mildew forecasting in filling stage based on remote sensing and meteorological data. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(9):165-172. (in Chinese)
[57]	张帅堂, 王紫烟, 邹修国, 钱燕, 余磊. 基于高光谱图像和遗传优化神经网络的茶叶病斑识别. 农业工程学报, 2017, 33(22):200-207.
	ZHANG S T, WANG Z Y, ZOU X G, QIAN Y, YU L. Recognition of tea disease spot based on hyperspectral image and genetic optimization neural network. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(22):200-207. (in Chinese)
[58]	ZHENG Q, HUANG W J, CUI X M, SHI Y, LIU L Y. New spectral index for detecting wheat yellow rust using sentinel-2 multispectral imagery. Sensors, 2018, 18(3):868. doi: 10.3390/s18030868
[59]	张春兰, 杨贵军, 李贺丽, 汤伏全, 刘畅, 张丽妍. 基于随机森林算法的冬小麦叶面积指数遥感反演研究. 中国农业科学, 2018, 51(5):855-867.
	ZHANG C L, YANG G J, LI H L, TANG F Q, LIU C, ZHANG L Y. Remote sensing inversion of leaf area index of winter wheat based on random forest algorithm. Scientia Agricultura Sinica, 2018, 51(5):855-867. (in Chinese)
[60]	岳继博, 杨贵军, 冯海宽. 基于随机森林算法的冬小麦生物量遥感估算模型对比. 农业工程学报, 2016, 32(18):175-182.
	YUE J B, YANG G J, FENG H K. Comparative of remote sensing estimation models of winter wheat biomass based on random forest algorithm. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(18):175-182. (in Chinese)
[61]	竞霞, 白宗璠, 高媛, 刘良云. 利用随机森林法协同SIF和反射率光谱监测小麦条锈病. 农业工程学报, 2019, 35(13):154-161.
	JING X, BAI Z F, GAO Y, LIU L Y. Wheat stripe rust monitoring by random forest algorithm combined with SIF and reflectance spectrum. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(13):154-161. (in Chinese)
[62]	赵晋陵, 金玉, 叶回春, 黄文江, 董莹莹, 范玲玲, 马慧琴, 江静. 基于无人机多光谱影像的槟榔黄化病遥感监测. 农业工程学报, 2020, 36(8):54-61.
	ZHAO J L, JIN Y, YE H C, HUANG W J, DONG Y Y, FAN L L, MA H Q, JIANG J. Remote sensing monitoring of areca yellow leaf disease based on UAV multi-spectral images. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(8):54-61. (in Chinese)

植被指数 Vegetation index	计算公式 Calculation formula	参考文献References
绿度归一化植被指数GNDVI	GNDVI=(R_NIR-R_Green)/(R_NIR+R_Green)	[27]
氮反应指数NRI	NRI=(R_Green-R_Red)/(R_Green+R_Red)	[28]
花青素反射指数ARI	ARI=(1/R_Green)-(1/R_Red)	[29]
叶斑病指数 CLSI	CLSI=(R_Re-R_Green)/(R_Re+R_Green)-R_Re	[6]
植物色素比例PPR	PPR=(R_Green-R_Blue)/(R_Green+R_Blue)	[30]
绿度指数GI	GI=R_Green/R_Red	[31]
优化土壤调节植被指数TCARI	TCARI=3×[(R_NIR-R_Red)-0.2×(R_NIR-R_Green)×R_NIR/R_Red]	[32]
可见光大气阻抗指数VARI	VARI=(R_Green-R_Red)/(R_Green+R_Red-R_Blue)	[33]
植被衰减指数PSRI	PSRI=(R_Red-R_Green)/R_NIR	[34]

光谱参数 Spectral parameter	相关系数 Correlation coefficient (r)	光谱参数 Spectral parameter	相关系数 Correlation coefficient (r)
R₄₅₀	0.352**	ARI	0.580**
R₅₅₀	0.489**	CLSI	0.586**
R₆₈₅	0.413**	PPR	-0.578**
R₇₂₅	0.286*	GI	-0.614**
R₇₈₀	-0.321**	TCARI	-0.568**
GNDVI	0.504**	VARI	-0.552**
NRI	-0.602**	PSRI	0.640**

纹理特征Texture feature	Blue	Green	Red	REDedge	NIR
方差 Variance (Var)	0.381**	0.277*	0.186	0.294**	0.267*
对比度Contrast ratio (Con)	0.289*	0.281*	0.227*	0.370**	0.273*
差异性Difference (Dis)	0.328**	0.274*	0.242*	0.353**	0.260*
熵 Entropy (Ent)	0.387**	0.240*	0.208	0.277*	0.219

自变量类型 Independent variable type	变量个数 Number of variables	模型算法 Model algorithm	训练集 Training set			验证集Validation set
自变量类型 Independent variable type	变量个数 Number of variables	模型算法 Model algorithm	R²	RMSE	RPD	R²	RMSE	RPD
植被指数VIs	9	MLR	0.509	5.796	1.437	0.528	8.594	0.957
		BP	0.613	5.780	1.450	0.596	6.917	1.353
		ELM	0.637	5.581	1.517	0.624	6.663	1.413
		RF	0.672	5.398	1.580	0.661	6.025	1.563
纹理特征TFs	4	MLR	0.247	7.651	1.210	0.216	10.515	0.703
		BP	0.279	8.004	1.064	0.289	9.241	0.864
		ELM	0.466	6.461	1.354	0.420	9.698	0.796
		RF	0.499	7.534	1.242	0.491	6.493	1.453
温度特征T	2	MLR	0.351	7.550	1.249	0.416	9.114	0.834
		BP	0.431	6.944	1.322	0.404	8.154	1.061
		ELM	0.540	6.065	1.368	0.495	8.495	0.980
		RF	0.556	6.988	1.324	0.561	6.138	1.536

自变量 Independent variable type	变量个数 Number of variables	模型算法 Model algorithm	训练集Training set			验证集Validation set
自变量 Independent variable type	变量个数 Number of variables	模型算法 Model algorithm	R²	RMSE	RPD	R²	RMSE	RPD
温度特征结合纹理特征T&TFs	6	MLR	0.504	7.527	1.251	0.501	7.492	1.217
		BP	0.563	6.205	1.407	0.601	6.384	1.478
		ELM	0.620	5.821	1.489	0.613	7.124	1.304
		RF	0.667	6.180	1.474	0.647	6.565	1.436
植被指数结合纹理特征VIs&TFs	13	MLR	0.555	6.600	1.516	0.541	6.861	1.366
		BP	0.587	5.274	1.725	0.607	6.187	1.525
		ELM	0.649	5.381	1.717	0.652	5.830	1.609
		RF	0.725	5.242	1.730	0.676	5.373	1.717
植被指数结合温度特征VIs&T	11	MLR	0.618	6.120	1.645	0.613	5.367	1.718
		BP	0.627	6.110	1.646	0.705	5.907	1.591
		ELM	0.695	5.924	1.736	0.744	6.190	1.574
		RF	0.741	4.686	1.953	0.758	4.722	1.871
植被指数结合温度和纹理特征VIs&T&TFs	15	MLR	0.603	6.499	1.694	0.638	4.932	1.821
		BP	0.667	5.404	1.896	0.718	5.820	1.796
		ELM	0.753	4.754	1.972	0.786	5.445	1.812
		RF	0.806	4.853	1.968	0.836	4.501	2.023