基于随机森林及遥感植被指数的无人农场水稻产量预测研究

doi:10.12398/j.issn.2096-7217.2025.02.009

摘要/Abstract

摘要：

为确保国家粮食安全、制定有效的粮食策略以及促进农业的可持续发展，对区域内水稻产量进行及时且准确的预测至关重要。尽管利用遥感数据结合机器学习技术进行产量预测已有一定进展，但现有研究在深入解析机器学习模型机理及高时空分辨率数据的应用方面仍显不足。本研究基于2023年的Sentinel和MODIS NDVI数据及县级产量信息，应用随机森林模型从县级扩展至像素级，以江苏射阳无人农场为例对水稻产量进行预测，并探讨不同特征对模型学习机理的影响。研究结果表明：（1）2023年8—10月的NDVI数据与产量之间存在显著相关性，此阶段为作物生长周期中的关键时期，植被覆盖和生长状态在此期间对产量的预测具有重要作用。随机森林模型能够有效预测县级水稻产量，其中RMSE值为339.5 kg/hm²。（2）此外，分析2023年射阳农场水稻产量空间分布图发现，水稻产量在该区域呈现显著的空间异质性，边缘地区产量较低，中心区域产量较高，产量主要集中在9 000~9 300 kg/hm²。其研究成果不仅加深了对机器学习及植被数据在产量预测中的应用理解，也为提高模型准确性及制定科学的粮食生产策略提供了理论支持。

关键词: 随机森林, 机器学习, 遥感, 植被指数, 产量预测, 生育期

Abstract:

In order to ensure national food security and promote sustainable agricultural development， it is crucial to accurately predict regional rice yield by formulating effective food strategies. While advancemnets have been achieved by integrating remote sensing data with machine learning technology for yield prediction， a more in-depth analysis of the machine learning model mechanism and the application of high spatio-temporal resolution data remains necessary. This study utilized Sentinel and MODIS Normalized Difference Vegetation Index （NDVI） data along with county-level yield statistics from 2023 to implement a Random Forest （RF） model for predicting yield from county level to pixel level with case study of Sheyang unmanned farm in Jiangsu Province， and investigated the impact of different features on the model's learning mechanism. The results demonstrated that there was a significant correlation between NDVI data and rice yield during the period of August to October 2023， which is a critical period in the crop phenological cycle. The RF model effectively predicted rice yield at the county level， with a Root Mean Square Error （RMSE） value of 339.5 kg /hm². Furthermore， spatial distribution mapping of the predicted yields within Sheyang County for 2023 revealed significant heterogeneity， indicating lower yields in marginal regions and higher yields in central areas ranging from 9 000 to 9 300 kg/hm². This study not only enhances understanding of machine learning application and vegetation data in yield prediction but also provides theoretical support for improving model accuracy and formulating scientific food production strategies.

Key words: Random Forest, machine learning, remote sensing： vegetation index, yield forecast, growth stage

中图分类号:

王俊, 吴振伟, 姜海, 柯娟, FOYEZ Ahmed Prodhan. 基于随机森林及遥感植被指数的无人农场水稻产量预测研究[J]. 智能化农业装备学报（中英文）, 2025, 6(2): 97-104.

WANG Jun, WU Zhenwei, JIANG Hai, KE Juan, FOYEZ Ahmed Prodhan. Rice yield prediction of unmanned farm based on random forest and remote sensing vegetation index[J]. Journal of Intelligent Agricultural Mechanization, 2025, 6(2): 97-104.

图/表 13

图1 随机森林模型框架

Figure 1 Random forest modelling framework

图2 随机森林模型预测水稻产量与统计产量比较

Figure 2 Comparison of Random Forest forecasting results of rice yield

表1 RF模型的参数值

Table 1 Parameter values of RF model

特征名称	特征值	最优值
N_estimators	[100, 200, 300, 400]	200
Max_depth	[10, 20, 30, 40]	20
Max_features	[sqrt, log2]	sqrt

图3 育秧移栽阶段大田作物状况

Figure 3 Field crop condition in seedling raising and transplanting stage

图4 分蘖生长期水稻长势状况

Figure 4 Rice growth status in tillering period

图5 拔节生长期水稻长势状况

Figure 5 Growth status of rice in jointing period

图6 孕穗生长期水稻长势状况

Figure 6 Growth status of rice during booting period

图7 抽穗扬花期水稻长势状况

Figure 7 Rice growth status at heading and flowering stage

图8 灌浆期至成熟期水稻长势状况

Figure 8 Rice growth from filling stage to mature stage

图9 收获期水稻大田状况

Figure 9 Rice planting conditions at harvest stage

图10 随机森林模型特征重要性分析

Figure 10 Analysis of the feature importance of Random Forest model

图11 2023年江苏射阳无人农场水稻产量分布图

Figure 11 The distribution of rice production from Sheyang unmanned farm of Jiangshu Province in 2023

图12 江苏射阳无人农场2023年水稻产量分布直方图

Figure 12 Histogram of rice yield distribution from Sheyang unmanned Farm of Jiangshu Province in 2023

参考文献 17

1	张佳华. 遥感技术在作物生长模式及农业气象预报中的应用研究［M］. 北京：中国农业科学技术出版社， 2021.
2	WANG J， CHEN J， ZHANG J， et al. Consistency and uncertainty of remote sensing-based approaches for regional yield gap estimation： A comprehensive assessment of process-based and data-driven models ［J］. Field Crops Research， 2023， 302： 109088.
3	张佳华，张健南，姚凤梅，等. 开放式增温对东北稻田生态系统作物生长与产量的影响［J］. 生态学杂志， 2013， 32（1）： 15-21.
	ZHANG Jiahua， ZHANG Jiannan， YAO Fengmei， et al. Effects of free air temperature increasing on the rice growth and grain yield in northeast China ［J］． Chinese Journal of Ecology， 2013， 32（1）： 15-21.
4	赵春江. 农业遥感研究与应用进展［J］. 农业机械学报， 2014， 45（12）： 277-293.
	ZHAO Chunjiang. Advances of research and application in remote sensing for agriculture ［J］. Transactions of the Chinese Society for Agricultural Machinery， 2014， 45（12）： 277-293.
5	聂鹏程，钱程，覃锐苗，等. 天空地一体化信息感知与融合技术发展现状与趋势［J］. 智能化农业装备学报（中英文）， 2023， 4（2）： 1-11.
	NIE Pengcheng， QIAN Cheng， QIN Ruimiao， et al. Development status and trends of space-air-ground integrated information sensing and fusion technology ［J］. Journal of Intelligent Agricultural Mechanization， 2023， 4（2）： 1-11.
6	BAI Y， ZHANG J， ZHANG S， et al. A remote sensing-based two-leaf canopy conductance model： Global optimization and applications in modeling gross primary productivity and evapotranspiration of crops ［J］. Remote Sensing of Environment， 2018， 215： 411-437.
7	ZHANG S， ZHANG J， BAI J， et al. Evaluation and improvement of the daily boreal ecosystem productivity simulator in simulating gross primary productivity at 41 flux sites across Europe ［J］. Ecological Modelling， 2018， 368： 205-232.
8	CAO D， FENG J， BAI L， et al. Delineating the rice crop activities in northeastern China through regional parametric synthesis using satellite remote sensing time-series data from 2000 to 2015 ［J］. Journal of Integrative Agriculture. 2021， 20（2）： 424-437.
9	WANG P J， SUN R， ZHANG J H， et al. Yield estimation of winter wheat in the north China Plain using the remote-sensing-photosynthesis-yield estimation for crops （RS-P-YEC） model ［J］. International Journal of Remote Sensing， 2011， 32： 6335-6348.
10	刘珊珊，牛超杰，边琳，等. 基于NDVI的水稻产量遥感估测［J］. 江苏农业科学， 2019， 47（3）： 193-198．
11	许童羽，洪雪，陈春玲，等. 基于冠层NDVI数据的北方粳稻产量模型研究［J］. 浙江农业学报， 2016， 28（10）： 1790-1795.
	XU Tongyu， HONG Xue， CHEN Chunling， et al. Study on northern japonica rice yield model based on canopy date of NDVI ［J］. Acta Agriculturae Zhejiangensis， 2016， 28（10）： 1790-1795.
12	YAO F， TANG Y， WANG P， et al. Estimation of maize yield by using a process-based model and remote sensing data in the northeast China Plain ［J］. Physics and Chemistry of the Earth， Parts A/B/C， 2015， 87： 142-152.
13	ZHANG S， BAI Y， ZHANG J H. Remote sensing-based quantification of the summer maize yield gap induced by suboptimum sowing dates over north China Plain ［J］. Remote Sensing， 2021， 13： 3582.
14	PRODHAN F A， ZHANG J， SHARMA T P P， et al. Projection of future drought and its impact on simulated crop yield over South Asia using ensemble machine learning approach ［J］. Science of the Total Environment， 2022， 807： 151029
15	李世隆，许辰一，王楠，等. 基于BWO-ELM的水稻氮素无人机高光谱反演研究［J］. 智能化农业装备学报（中英文）， 2024， 5（3）： 14-21.
	LI Shilong， XU Chenyi， WANG Nan， et al. Research on rice nitrogen unmanned aerial vehicle hyperspectral inversion based on BWO-ELM ［J］. Journal of Intelligent Agricultural Mechanization， 2024，5（3）： 14-21
16	AGHIGHI H， AZADBAKHT M， ASHOURLOO D， et al. Machine learning regression techniques for the silage maize yield prediction using time-series images of Landsat 8 OLI ［J］. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing， 2018， 11， 4563-4577.
17	YANG X， ZHANG J， YANG S， et al. Modelling the crop yield gap with a remote sensing-based process model： A case study of winter wheat in the north China Plain ［J］. Journal of Integrative Agriculture， 2023， 22（10）： 2993-3005.

[1]	高宁, 张安琪, 梅鹤波, 杨兴华, 甘蕾, 孟志军. 土壤墒情监测技术研究现状与发展趋势[J]. 智能化农业装备学报（中英文）, 2024, 5(3): 51-62.
[2]	徐畅, 张领先, 乔岩. 基于电子病历的作物病虫害关联挖掘及智能诊断[J]. 智能化农业装备学报（中英文）, 2023, 4(4): 1-10.
[3]	聂鹏程, 钱程, 覃锐苗, 邓水光, 孙崇德, 何勇. 天空地一体化信息感知与融合技术发展现状与趋势[J]. 智能化农业装备学报（中英文）, 2023, 4(2): 1-11.