基于多源数据的小麦品种产量估测研究

doi:10.3964/j.issn.1000-0593(2023)07-2210-10

摘要
参考文献
相关文章 (15)

全文: PDF (8428 KB)
输出: BibTeX | EndNote (RIS)

摘要：小麦产量产前估测关乎农业生产计划制定、粮食安全保障、国家经济和宏观决策。应用无人机能够无损、快速准确、及时高效地估测小麦产量，通过多种机器学习方法充分挖掘无人机多源遥感数据对多个小麦品种进行籽粒产量估测的潜力，明确多源数据融合对模型估测精度的提升效果，对于作物田间管理保障小麦高产稳产具有重要意义。以黄淮麦区140个主栽小麦品种为材料开展冬小麦田间试验，采用搭载红绿蓝（RGB）和多光谱传感器的无人机平台对灌浆期的冠层信息进行采集，分别以岭回归、支持向量回归、随机森林回归、高斯过程、k-最邻近算法和Cubist等六种机器学习算法建立单传感器数据以及多源数据融合的产量估测模型，采用决定系数（R²）、均方根误差（RMSE）和相对均方根误差（RRMSE）对估算模型进行评价。结果表明，所选取的10个可见光植被指数及13个多光谱被指数特征值均与实测产量呈极显著相关（p<0.01），各特征值产量相关系数绝对值由高到低依次为多光谱植被指数（0.54～0.83）、可见光植被指数（0.45～0.61）、纹理特征（<0.45）。全部六种机器学习算法均在采用多源数据融合时产量估测模型精度最高，多源数据融合产量估测精度（平均决定系数R²=0.50～0.71）>多光谱传感器产量估测精度（R²=0.53～0.69）>RGB传感器产量估测精度（R²=0.35～0.51）。多源数据融合相对于RGB数据的R²提高0.17～0.23，平均均方根误差（RMSE）降低0.06～0.09 t·hm^-2；相对于多光谱数据的R²提高0.01～0.06，RMSE降低0.01～0.03 t·hm^-2。Cubist算法与其他5种算法相比，建立的多源数据融合模型产量估测精度最高，R²为0.71，RMSE为0.29 t·hm^-2。研究表明，相对于单一传感器数据产量估测模型，多源数据融合能够有效提升冬小麦品种产量的估测精度，并且Cubist算法能相对更好地处理多模态融合数据提高产量预测精度，为预测不同小麦品种的产量提供理论指导。

关键词：无人机；遥感；小麦估产；光谱指数；纹理特征

Abstract：Pre-production estimation of wheat production is related to the formulation of agricultural production plans, food security, national economy and macro-decision-making, and the application of drones can estimate wheat production in a non-destructive, fast, accurate, timely and efficient manner. The machine learning method is used to fully tap the potential of multi-source remote sensing data to estimate the grain yield of multiple wheat varieties and to clarify the effect of multi-source data fusion on improving the yield estimation accuracy of cultivars. It is significant for crop field management and ensuring a high and stable yield in wheat. In this study, field trials of winter wheat were carried out with 140 main wheat varieties in the Huanghuai wheat region as materials. The drone platform equipped with red green blue (RGB) and multispectral sensors were used to collect the canopy information of 140 winter wheat varieties during the grain filling period. Six machine learning algorithms were used, namely Ridge Regression (RR), support vector regression (SVR), Random Forest Regression (RFR), Gaussian Process (GP), k-Nearest Neighbor (k-NN) and Cubist, to build yield estimation models from single sensor data and multi-source data fusion. Coefficient of determination (R²), root mean square error (RMSE) and relative root mean square error (RRMSE) were used to evaluate the estimation model. The results showed that the selected 10 visible vegetation indices and 13 multispectral covered indices were significantly correlated with the measured yield (p<0.05), and the absolute value of the correlation coefficient from high to low was multispectral vegetation index (0.54～0.83), color index (0.45～0.61), texture feature (<0.45), all six machine learning algorithms have the highest yield estimation and prediction accuracy when using multi-source data fusion. Multi-source data fusion yield estimation accuracy (average coefficient of determination R²=0.50～0.71)>multi-spectral sensor yield estimation accuracy (R²=0.53～0.69)>RGB sensor yield estimation accuracy (R²=0.35～0.51). Compared with RGB data, the R² of multi-source data fusion increases by 0.17～0.23, and the mean root mean square error (RMSE) decreases by 0.06～0.09 t·hm^-2; compared with multi-spectral data, the R² increases by 0.01～0.06, and the RMSE decreases by 0.01～0.03 t·hm^-2. Compared with the other five algorithms, the multi-source data fusion model established by the Cubist algorithm has the highest yield estimation accuracy, with an R² of 0.71 and an RMSE of 0.29 t·hm^-2. It shows that compared with the yield estimation model of single sensor data, multi-source data fusion can effectively improve the yield estimation accuracy of winter wheat varieties, and the Cubist algorithm can better process multi-mode data to improve the yield prediction accuracy, providing theoretical guidance for predicting the yield of different wheat varieties.

Key words：Unmanned aerial vehicle;Remote sensing;Wheat yield estimation;Spectral index;Texture feature

收稿日期: 2022-02-22 修订日期: 2022-09-16

中图分类号:

S127

基金资助: 国家重点研发计划项目（2021YFD1200601-07），国家小麦产业技术体系项目（CARS-03-16）和新疆维吾尔自治区公益性科研院所基础科学活动专项资助项目（ky2019010）资助

通讯作者: 肖永贵，陶志强 E-mail: xiaoyonggui@caas.cn；taozhiqiang@caas.cn

作者简介: 宋成阳，1996年生，新疆农业大学农学院硕士研究生 e-mail: byyyscy@126.com

引用本文:

宋成阳，耿洪伟，费帅鹏，李雷，甘甜，曾潮武，肖永贵，陶志强. 基于多源数据的小麦品种产量估测研究[J]. 光谱学与光谱分析, 2023, 43(07): 2210-2219.
SONG Cheng-yang, GENG Hong-wei, FEI Shuai-peng, LI Lei, GAN Tian, ZENG Chao-wu, XIAO Yong-gui, TAO Zhi-qiang. Study on Yield Estimation of Wheat Varieties Based on Multi-Source Data. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(07): 2210-2219.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2023)07-2210-10 或 https://www.gpxygpfx.com/CN/Y2023/V43/I07/2210

[1] HE Zhong-hu, ZHUANG Qiao-sheng, CHENG Shun-he, et al（何中虎, 庄巧生, 程顺和, 等）. Journal of Agriculture（农学学报）, 2018, 8(1): 99.
[2] National Bureau of Statistics of China（中华人民共和国国家统计局）. CHINA STATISTICAL YEARBOOK （中国统计年鉴），2021-12-06：http://www.gov.cn/xinwen/2021-12/06/content_5656247.htm.
[3] Wu X, Feng H, Wu D, et al. Genome Biology, 2021, 22(1): 185.
[4] Schut A G T, Traore P C S, Blaes X, et al. Field Crops Research, 2018, 221: 98.
[5] Du M, Noguchi N. IFAC-PapersOnLine, 2016, 49(16): 5.
[6] Zhang C, Kovacs J M. Precision Agriculture, 2012, 13(6): 693.
[7] GUO Tao, YAN An, GENG Hong-wei（郭涛, 颜安, 耿洪伟）. Journal of Triticeae Crops（麦类作物学报）, 2020, 40(9): 1129.
[8] WU Bing-fang, ZHANG Miao, ZENG Hong-wei, et al（吴炳方, 张淼, 曾红伟, 等）. Journal of Remote Sensing（遥感学报）, 2016, 20(5): 1027.
[9] LIU Jian-gang, ZHAO Chun-jiang, YANG Gui-jun, et al（刘建刚, 赵春江, 杨贵军, 等）. Transactions of the Chinese Society of Agricultural Engineering （农业工程学报）, 2016, 32(24): 98.
[10] Hassan M A, Yang M, Rasheed A, et al. Plant Science, 2019, 282: 95.
[11] Hassan M, Yang M J, Rasheed A, et al. Remote Sensing, 2018, 10(6): 809.
[12] TAO Hui-lin, FENG Hai-kuan, YANG Gui-jun, et al (陶惠林, 冯海宽, 杨贵军, 等). Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2019, 35(23): 111.
[13] Zhu W X, Li S, Zhang X, et al. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34: 78.
[14] Hall D L，Llinas J. Proceedings of the IEEE, 1997, 85: 6.
[15] WANG Lai-gang, ZHENG Guo-qing, GUO Yan, et al（王来刚, 郑国清, 郭燕, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报）, 2022, 53(1): 198.
[16] Wei Q, Bioucas Dias J, Dobigeon N, et al. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(7): 3658.
[17] Bilgin G, Ustuner M. Journal of Applied Remote Sensing, 2015, 9(1): 096054.
[18] Maimaitijiang M, Sagan V, Sidike P, et al. Remote Sensing of Environment, 2020, 237: 111599.
[19] Jhan J P, Rau J Y, Haala N. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 137: 47.
[20] Bendig J, Yu K, Aasen H, et al. International Journal of Applied Earth Observation and Geoinformation, 2015, 39: 79.
[21] Metternicht G. International Journal of Remote Sensing, 2003, 24(14): 2855.
[22] Louhaichi M, Borman M M, Johnson D E. Geocarto International, 2001, 16(1): 65.
[23] Woebbecke D M, Meyer G E, Von Bargen K, et al. Transactions of the American Society of Agricultural Engineers, 1995, 38: 259.
[24] Kataoka T, Kaneko T, Okamoto H, et al. Crop Growth Estimation System Using Machine Vision，2003, 2: 1079.
[25] Gitelson A A, Kaufman Y J, Stark R, et al. Remote Sensing of Environment, 2002, 80(1): 76.
[26] Kawashima S, Nakatani M. Annals of Botany, 1998, 81(1): 49.
[27] Woebbecke D M, Meyer G E, Von Bargen K, et al. Transactions of the ASAE, 1995, 38(1): 259.
[28] ZHANG Pei-song, SUN Yi-ming, GUO Peng-tao, et al（张培松, 孙毅明, 郭澎涛, 等）. Journal of Tropical Crops（热带作物学报）, 2015, 36(12): 2120.
[29] Gitelson A A, Kaufman Y J, Merzlyak M N. Remote Sensing of Environment, 1996, 58(3): 289.
[30] Dash J, Curran P J. International Journal of Remote Sensing, 2004, 25(23): 5403.
[31] Liu X, Wei Y, Jiao Q, et al. Remote Sensing Technology and Application, 2019, 34(4): 756.
[32] CHEN La, HUANG Jing-feng, WANG Xiu-zhen（陈拉, 黄敬峰, 王秀珍）. Journal of Remote Sensing（遥感学报）, 2008, 1: 143.
[33] Xue L, Cao W, Luo W, et al. Agronomy Journal, 2004, 96(1): 135.
[34] Chen J M. Canadian Journal of Remote Sensing, 1996, 22(3): 229.
[35] Wang K, Shen Z Q, Wang R C. Journal of Zhejiang University (Agriculture and Life Sciences), 1998, 1: 95.
[36] NIU Qing-lin, FENG Hai-kuan, ZHOU Xin-guo, et al（牛庆林, 冯海宽, 周新国, 等）. Transactions of the Chinese Society for Agricultural Machinery（农业机械学报）, 2021, 52(8): 183.
[37] Gitelson A A, Viña A, Arkebauer T J, et al. Geophysical Research Letters, 2003, 30(5): 1248.
[38] LIU Chang, YANG Gui-jun, LI Zhen-hai, et al（刘畅, 杨贵军, 李振海, 等）. Scientia Agricultura Sinica（中国农业科学）, 2018, 51(16): 3060.
[39] Yue J B, YangG J, Li C, et al. Remote Sensing, 2017, 9: 708.
[40] Yang K, Gong Y, Fang S, et al. Remote Sensing, 2021, 13(15): 3001.
[41] LIU Xin-yi, ZHONG Xiao-chun, CHEN Chen, et al（刘欣谊, 仲晓春, 陈晨, 等）. Journal of Triticeae Crops（麦类作物学报）, 2020, 40(8): 1002.
[42] YAO Rui, LIU Jin-rong, LIU Pei-jiang, et al（姚睿, 刘金容, 刘培江, 等）. Mathematical Theory and Applications（数学理论与应用）, 2019, 39(1): 111.
[43] OU Qiang-xin, LI Hai-kui, LEI Xiang-dong, et al（欧强新, 李海奎, 雷相东, 等）. Chinese Journal of Applied Ecology（应用生态学报）, 2018, 29(6): 2007.