智能化农业装备学报（中英文）

基于无人机多维特征的水稻估产模型对比与归因解析

预览PDF | 下载PDF

兰玉彬 ¹^,³^,⁴ , 吴奇梁 ¹^,⁴ , 张彬 ² , 徐伟诚 ² , 张雷 ³^,⁴^,⁵ , 杨炜光 ¹^,³^,⁴^,^*^,^*

智能化农业装备学报（中英文） | 2026,7(2): 11-20

收起

智能化农业装备学报（中英文） | 2026 , 7 (2) : 11 -20

基于无人机多维特征的水稻估产模型对比与归因解析

全屏

兰玉彬¹^,³^,⁴, 吴奇梁¹^,⁴, 张彬², 徐伟诚², 张雷³^,⁴^,⁵, 杨炜光¹^,³^,⁴^,^*^,^*

作者信息

^1.华南农业大学人工智能与低空技术学院，广东广州，510642

^2.广东省农业科学院水稻育种新技术重点实验室，广东广州，510640

^3.绿色农药全国重点实验室，广东广州，510642

^4.国家精准农业航空施药技术国际联合研究中心，广东广州，510642

^5.华南农业大学农学院，广东广州，510642

通讯作者:

^* 杨炜光，男，1996年生，甘肃张掖人，博士，助理研究员；研究方向为农业多源遥感与作物长势监测。E-mail：wgyang@scau.edu.cn

作者简介:

兰玉彬，男，1961年生，吉林农安人，博士，教授；研究方向为精准农业航空技术。E-mail：ylan@scau.edu.cn

Comparison and attribution analysis of rice yield estimation models based on multi-dimensional UAV features

Yubin Lan¹^,³^,⁴, Qiliang Wu¹^,⁴, Bin Zhang², Weicheng Xu², Lei Zhang³^,⁴^,⁵, Weiguang Yang¹^,³^,⁴

Affiliations

^1.College of Artificial Intelligence and Low-Altitude Technology, South China Agricultural University, Guangzhou 510642, China

^2.Guangdong Key Laboratory of New Technology for Rice Breeding, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, China

^3.State Key Laboratory of Green Pesticide, Guangzhou 510642, China

^4.National Center for International Collaboration Research on Precision Agricultural Aviation Pesticide Spraying Technology, Guangzhou 510642, China

^5.College of Agriculture, South China Agricultural University, Guangzhou 510642, China

Corresponding Author:

^* Yang Weiguang, E-mail: wgyang@scau.edu.cn

Lan Yubin, E-mail: ylan@scau.edu.cn

出版时间: 2026-05-15 doi: 10.12398/j.issn.2096-7217.2026.02.002

文章导航

摘要

收起

作物产量的精准定量预测是保障国家粮食安全与制定智慧农业决策的重要基础。在水稻生长周期中，齐穗期是生物量积累最关键的阶段，其冠层光谱特征蕴含着丰富的生化参数与产量潜力信息。本研究探讨了齐穗期多维无人机遥感特征与先进机器学习算法组合在水稻产量预测中的应用优势。研究利用多光谱无人机获取影像，系统构建了可见光（red-green-blue，RGB）、多光谱（multispectral，MS）、重建高光谱（multispectral to hyperspectral，MSTPP）、空间纹理（spatial texture，Texture）、植被指数（vegetation index，VI）及多维融合特征（multi-dimensional fusion，Fusion）在内的6类特征集。算法层面全面评估了8种机器学习算法，重点对比了支持向量回归（support vector regression，SVR）等核方法及多类非线性集成模型。试验结果表明，MS数据在单维特征中展现出突出的预测潜力。融合光谱与纹理信息的策略取得了优异的预测稳定性，其中SVR模型结合多维融合特征实现了理想的拟合效果（决定系数R²为0.828），校准了因齐穗期冠层郁闭而受损的光谱信号，有效修正了单维模型在极端产量区间的系统性估算偏差。此外，本实验通过引入沙普利加和解释（Shapley additive explanations，SHAP）框架验证了绿光归一化植被指数（green normalized difference vegetation index，GNDVI）在产量预测中起到了核心驱动作用，客观揭示了群体长势均衡度对最终产量的保底机理。本研究探索的“多维融合结合非线性算法与归因解析”路径，为构建兼具高鲁棒性与农学解释力的定量估产模型提供了一种可行的思路。

关键词

产量预测 / 无人机遥感 / 特征集 / 特征融合 / 机器学习 / SHAP框架

Abstract

收起

Accurate quantitative prediction of crop yield is a crucial foundation for ensuring national food security and formulating smart agricultural decisions. During the rice growth cycle, the full heading stage is the most critical period for biomass accumulation, and its canopy spectral characteristics contain rich information on biochemical parameters and yield potential. This study explores the application advantages of combining multi-dimensional UAV remote sensing features with advanced machine learning algorithms for rice yield prediction at the full heading stage. Using a multispectral UAV to acquire images, this research systematically constructed six categories of feature sets, including red-green-blue (RGB), multispectral (MS), multispectral to hyperspectral (MSTPP), spatial texture (Texture), vegetation index (VI), and multi-dimensional fusion (Fusion). At the algorithmic level, eight machine learning algorithms were comprehensively evaluated, with a focus on comparing kernel methods represented by support vector regression (SVR) and various nonlinear ensemble models. The experimental results indicate that MS data exhibit outstanding prediction potential among single-dimensional features. The strategy of fusing spectral and texture information achieved excellent prediction stability, wherein the SVR model combined with multi-dimensional fusion features achieved an ideal fitting effect (coefficient of determination R²=0.828), calibrating the spectral signals impaired by canopy closure during the full heading stage, and effectively correcting the systematic estimation bias of single-dimensional models in extreme yield intervals. Furthermore, this experiment verified through the introduction of the Shapley additive explanations (SHAP) framework that the green normalized difference vegetation index (GNDVI) played a core driving role in yield prediction, objectively revealing the safeguard mechanism of population growth balance on the final yield. The approach of "multi-dimensional fusion combined with nonlinear algorithms and attribution analysis" explored in this study provides a feasible approach for constructing quantitative yield estimation models with both high robustness and agronomic interpretability.

Key words

yield prediction / UAV remote sensing / feature sets / feature fusion / machine learning / SHAP framework

引用本文

兰玉彬, 吴奇梁, 张彬, 徐伟诚, 张雷, 杨炜光. 基于无人机多维特征的水稻估产模型对比与归因解析. 智能化农业装备学报（中英文）, 2026 , 7 (2) : 11 -20 . DOI: 10.12398/j.issn.2096-7217.2026.02.002

Yubin Lan, Qiliang Wu, Bin Zhang, Weicheng Xu, Lei Zhang, Weiguang Yang. Comparison and attribution analysis of rice yield estimation models based on multi-dimensional UAV features[J]. Journal of Intelligent Agricultural Mechanization, 2026 , 7 (2) : 11 -20 . DOI: 10.12398/j.issn.2096-7217.2026.02.002

基金

收起

国家自然科学基金(32501765)
中国博士后科学基金(2025M782467)
岭南现代农业实验室科研项目(NT2021009)
高等学校学科创新引智基地资助(D18019)
国家棉花产业技术体系项目(CARS-15-23)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

[1]

张伟，王松寒，和玉璞，等．遥感技术驱动的作物产量估算方法研究进展[J]．南京农业大学学报，2026（1）：1-17.

Zhang Wei, Wang Songhan, He Yupu, et al. Advances in remote sensing-based methods for crop yield estimation [J]. Journal of NanjingAgricultural University, 2026(1): 1-17.

[2]

吉文翰，郑恒彪，王迪，等．基于无人机影像和卷积神经网络的水稻育种材料产量预测研究[J]．南京农业大学学报，2025，48（5）：1001-1012.

Ji Wenhan, Zheng Hengbiao, Wang Di, et al. Yield prediction in rice breeding materials using UAV-based images and convolutional neural network [J]. Journal of Nanjing Agricultural University, 2025, 48(5): 1001-1012.

[3]

薛利红，曹卫星，罗卫红．基于冠层反射光谱的水稻产量预测模型[J]．遥感学报，2005（1）：100-105.

Xue Lihong, Cao Weixing, Luo Weihong. Rice yield forecasting model with canopy reflectance spectra [J]. Journal of Remote Sensing, 2005(1): 100-105.

[4]

Kganyago M, Adjorlolo C, Mhangara P, et al. Optical remote sensing of crop biophysical and biochemical parameters: An overview of advances in sensor technologies and machine learning algorithms for precision agriculture[J]. Computers and Electronics in Agriculture, 2024, 218: 108730.

[5]

兰玉彬，王天伟，陈盛德，等．农业人工智能技术：现代农业科技的翅膀[J]．华南农业大学学报，2020，41（6）：1-13.

Lan Yubin, Wang Tianwei, Chen Shengde, et al. Agricultural artificial intelligence technology: Wings of modern agricultural science and technology [J]. Journal of South China Agricultural University, 2020, 41(6): 1-13.

[6]

Quille-Mamani J, Ramos-Fernández L, Huanuqueño-Murillo J, et al. Rice yield prediction using spectral and textural indices derived from UAV imagery and machine learning models in Lambayeque, Peru [J]. Remote Sensing, 2025, 17(4): 632.

[7]

孙竹，顾伟，崔龙飞，等．智能植保装备关键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（4）：1-23.

Sun Zhu, Gu Wei, Cui Longfei, et al. Research status and development trends of key technologies for intelligent plant protection equipment [J]. Journal of Intelligent Agricultural Mechanization, 2024, 5(4): 1-23.

[8]

Yang W G, Zhang B, Xu W C, et al. Impact of hyperspectral reconstruction techniques on the quantitative inversion of rice physiological parameters: A case study using the MST++ model [J]. Journal of Integrative Agriculture, 2025, 24(7): 2540-2557.

[9]

Yang W G, Fu H Y, Xu W C, et al. Optimized modular transfer learning framework integrating PROSAIL and UAV-based hyperspectral reconstruction for cotton canopy water and nitrogen content retrieval [J]. Artificial Intelligence in Agriculture, 2026, 16(2): 764-787.

[10]

朱俊科，赵红建，吕垚，等．基于无人机植被指数与纹理-颜色特征融合的冬小麦冻害评价[J]．农业工程学报，2025，41（14）：165-173.

Zhu Junke, Zhao Hongjian, Lü Yao, et al. Evaluating freeze damage in winter wheat using vegetation index and texture-color features of unmanned aerial vehicle [J]. Transactions of the Chinese Society of Agricultural Engineering, 2025, 41(14): 165-173.

[11]

赵静，闫春雨，杨东建，等．基于无人机多光谱遥感的台风灾后玉米倒伏信息提取[J]．农业工程学报，2021，37（24）：56-64.

Zhao Jing, Yan Chunyu, Yang Dongjian, et al. Extraction of maize lodging information after typhoon based on UAV multispectral remote sensing [J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(24): 56-64.

[12]

王俊，吴振伟，姜海，等．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, et al. 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.

[13]

Brandt P, Beyer F, Borrmann P, et al. Ensemble learning-based crop yield estimation: A scalable approach for supporting agricultural statistics [J]. GIScience & Remote Sensing, 2024, 61(1): 2367808.

[14]

Wang B J, Zhu J K, Sun S, et al. Multimodal remote sensing combination for maize LAI estimation: Stacking model development and phenology-specific feature sensitivity analysis [J]. Artificial Intelligence in Agriculture, 2026, 16(2): 873-888.

[15]

赵瑞雪，杨潇，张丹丹，等．农业智能科研新范式：关键要素、面临挑战与路径建议[J]．智慧农业（中英文），2025，7（3）：35-47.

Zhao Ruixue, Yang Xiao, Zhang Dandan, et al. Artificial intelligence for agricultural science (AI4AS): Key elements, challenges and pathways [J]. Smart Agriculture, 2025, 7(3): 35-47.

[16]

Hohl A, Obadic I, Fernandez-Torres M A, et al. Opening the black box: A systematic review on explainable AI in remote sensing [J]. IEEE Geoscience and Remote Sensing Magazine, 2024, 12(4): 261-304.

[17]

于丰华，许童羽，郭忠辉，等．水稻智慧无人农场关键技术研究现状与展望[J]．智慧农业（中英文），2024，6（6）：1-22.

Yu Fenghua, Xu Tongyu, Guo Zhonghui, et al. Research status and prospects of key technologies for rice smart unmanned farms [J]. Smart Agriculture, 2024, 6(6): 1-22.

[18]

黄光群，韩鲁佳．基于非线性径向基核函数支持向量机的堆肥产品近红外光谱分析研究[J]．光学学报，2009，29（12）：3556-3560.

Huang Guangqun, Han Lujia. Near infrared reflectance spectroscopy analysis of compost products using nonlinear support vector machine with RBF nucleus [J]. Acta Optica Sinica, 2009, 29(12): 3556-3560.

[19]

尹儒，门昌骞，王文剑．一种模型决策森林算法[J]．计算机科学与探索，2020，14（1）：108-116.

Yin Ru, Men Changqian, Wang Wenjian. Model decision forest algorithm [J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(1): 108-116.

[20]

赵胜利，王国宾，胡连槟，等．基于无人机多光谱遥感的棉花生长参数和产量估算[J]．中国农机化学报，2024，45（2）：227-234.

Zhao Shengli, Wang Guobin, Hu Lianbin, et al. Estimation of cotton growth parameters and yield based on UAV multispectral remote sensing [J]. Journal of Chinese Agricultural Mechanization, 2024, 45(2): 227-234.

[21]

El-Shahat D, Tolba A, Abouhawwash M, et al. Machine learning and deep learning models based grid search cross validation for short-term solar irradiance forecasting [J]. Journal of Big Data, 2024, 11(1): 134.

[22]

Lundberg S M, Lee S I. A unified approach to interpreting model predictions [J]. Advances in Neural Information Processing Systems, 2017, 30: 4765-4774.

[23]

常硕其，粟琳，欧阳翔．水稻产量提高与光合作用之间的关系[J]．生命科学，2024，36（10）：1305-1310.

Chang Shuoqi, Su Lin, Ouyang Xiang. The relationship between the enhancement of rice grain yield and photosynthesis [J]. Chinese Bulletin of Life Sciences, 2024, 36(10): 1305-1310.

[24]

徐孟亮，丁绿萍，罗佳，等．超级稻超高产形成的光合机制研究进展[J]．生命科学研究，2021，25（5）：386-392.

Xu Mengliang, Ding Lüping, Luo Jia, et al. Research progress of photosynthetic mechanisms of super high-grain-yield rice [J]. Frontiers in Plant Science, 2021, 25(5): 386-392.

[25]

徐恩恩，郭颖，陈尔学，等．基于无人机LiDAR和高空间分辨率卫星遥感数据的区域森林郁闭度估测模型[J]．武汉大学学报（信息科学版），2022，47（8）：1298-1308.

Xu Enen, Guo Ying, Chen Erxue, et al. An estimation model for regional forest canopy closure combined with UAV LiDAR and high spatial resolution satellite remote sensing data [J]. Geomatics and Information Science of Wuhan University, 2022, 47(8): 1298-1308.

2026年第7卷第2期

预览PDF | 下载PDF

引用本文

BibTeX

文章信息

doi: 10.12398/j.issn.2096-7217.2026.02.002

接收时间：2026-04-10
首发时间：2026-11-05
出版时间：2026-05-15

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2026-04-10
修回日期：2026-04-30

基金

National Natural Science Foundation of China(32501765)

国家自然科学基金(32501765)

China Postdoctoral Science Foundation(2025M782467)

中国博士后科学基金(2025M782467)

Laboratory of Lingnan Modern Agriculture Project(NT2021009)

岭南现代农业实验室科研项目(NT2021009)

the Programme of Introducing Talents of Disciplines to Universities of China(D18019)

高等学校学科创新引智基地资助(D18019)

National Cotton Industry Technology System Projects(CARS-15-23)

国家棉花产业技术体系项目(CARS-15-23)

作者信息

^1.华南农业大学人工智能与低空技术学院，广东广州，510642

^2.广东省农业科学院水稻育种新技术重点实验室，广东广州，510640

^3.绿色农药全国重点实验室，广东广州，510642

^4.国家精准农业航空施药技术国际联合研究中心，广东广州，510642

^5.华南农业大学农学院，广东广州，510642

通讯作者:

^* 杨炜光，男，1996年生，甘肃张掖人，博士，助理研究员；研究方向为农业多源遥感与作物长势监测。E-mail：wgyang@scau.edu.cn

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/znhnyzbxb/CN/10.12398/j.issn.2096-7217.2026.02.002

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT