Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data

This study investigates a multimodal deep learning framework that integrates unmanned aerial vehicle (UAV) multispectral imagery with meteorological data to predict cotton yield. The study analyzes the impact of different neural network architectures, including the CNN feature extraction layer, the...

Full description

Saved in:
Bibliographic Details
Main Authors: Chunbo Jiang, Xiaoshuai Guo, Yongfu Li, Ning Lai, Lei Peng, Qinglong Geng
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Agronomy
Subjects:
Online Access:https://www.mdpi.com/2073-4395/15/5/1217
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849711324372664320
author Chunbo Jiang
Xiaoshuai Guo
Yongfu Li
Ning Lai
Lei Peng
Qinglong Geng
author_facet Chunbo Jiang
Xiaoshuai Guo
Yongfu Li
Ning Lai
Lei Peng
Qinglong Geng
author_sort Chunbo Jiang
collection DOAJ
description This study investigates a multimodal deep learning framework that integrates unmanned aerial vehicle (UAV) multispectral imagery with meteorological data to predict cotton yield. The study analyzes the impact of different neural network architectures, including the CNN feature extraction layer, the depth of the fully connected layer, and the method of integrating meteorological data, on model performance. Experimental results show that the model combining UAV multispectral imagery with weekly meteorological data achieved optimal yield prediction accuracy (RMSE = 0.27 t/ha; R<sup>2</sup> = 0.61). Specifically, models based on AlexNet (Model 9) and CNN2conv (Model 18) exhibited superior accuracy. ANOVA results revealed that deeper fully connected layers significantly reduced RMSE, while variations in CNN architectural complexity had no statistically significant effect. Furthermore, although the models exhibited comparable prediction accuracy (RMSE: 0.27–0.33 t/ha; R<sup>2</sup>: 0.61–0.69 across test datasets), their yield prediction spatial distributions varied significantly (e.g., Model 9 predicted a mean yield of 3.88 t/ha with a range of 2.51–4.89 t/ha, versus Model 18 at 3.74 t/ha and 2.33–4.76 t/ha), suggesting the need for further evaluation of spatial stability. This study underscores the potential of deep learning models integrating UAV and meteorological data for precision agriculture, offering valuable insights for optimizing spatiotemporal data integration strategies in future research.
format Article
id doaj-art-1acc1d49679e45e0bbf26d78a3be3bdc
institution DOAJ
issn 2073-4395
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Agronomy
spelling doaj-art-1acc1d49679e45e0bbf26d78a3be3bdc2025-08-20T03:14:39ZengMDPI AGAgronomy2073-43952025-05-01155121710.3390/agronomy15051217Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological DataChunbo Jiang0Xiaoshuai Guo1Yongfu Li2Ning Lai3Lei Peng4Qinglong Geng5Agricultural Engineering and Information Technology, College of Resources and Environment, Xinjiang Agricultural University, Urumqi 830052, ChinaAgricultural Engineering and Information Technology, College of Resources and Environment, Xinjiang Agricultural University, Urumqi 830052, ChinaXinjiang Academy of Agricultural Sciences, Resource and Environmental Information Technology Innovation Team, Urumqi 830091, ChinaXinjiang Academy of Agricultural Sciences, Resource and Environmental Information Technology Innovation Team, Urumqi 830091, ChinaXinjiang Academy of Agricultural Sciences, Resource and Environmental Information Technology Innovation Team, Urumqi 830091, ChinaXinjiang Academy of Agricultural Sciences, Resource and Environmental Information Technology Innovation Team, Urumqi 830091, ChinaThis study investigates a multimodal deep learning framework that integrates unmanned aerial vehicle (UAV) multispectral imagery with meteorological data to predict cotton yield. The study analyzes the impact of different neural network architectures, including the CNN feature extraction layer, the depth of the fully connected layer, and the method of integrating meteorological data, on model performance. Experimental results show that the model combining UAV multispectral imagery with weekly meteorological data achieved optimal yield prediction accuracy (RMSE = 0.27 t/ha; R<sup>2</sup> = 0.61). Specifically, models based on AlexNet (Model 9) and CNN2conv (Model 18) exhibited superior accuracy. ANOVA results revealed that deeper fully connected layers significantly reduced RMSE, while variations in CNN architectural complexity had no statistically significant effect. Furthermore, although the models exhibited comparable prediction accuracy (RMSE: 0.27–0.33 t/ha; R<sup>2</sup>: 0.61–0.69 across test datasets), their yield prediction spatial distributions varied significantly (e.g., Model 9 predicted a mean yield of 3.88 t/ha with a range of 2.51–4.89 t/ha, versus Model 18 at 3.74 t/ha and 2.33–4.76 t/ha), suggesting the need for further evaluation of spatial stability. This study underscores the potential of deep learning models integrating UAV and meteorological data for precision agriculture, offering valuable insights for optimizing spatiotemporal data integration strategies in future research.https://www.mdpi.com/2073-4395/15/5/1217UAV multispectral imagerymeteorological data integrationcottonmultimodal deep learningprecision agriculture
spellingShingle Chunbo Jiang
Xiaoshuai Guo
Yongfu Li
Ning Lai
Lei Peng
Qinglong Geng
Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data
Agronomy
UAV multispectral imagery
meteorological data integration
cotton
multimodal deep learning
precision agriculture
title Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data
title_full Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data
title_fullStr Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data
title_full_unstemmed Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data
title_short Multimodal Deep Learning Models in Precision Agriculture: Cotton Yield Prediction Based on Unmanned Aerial Vehicle Imagery and Meteorological Data
title_sort multimodal deep learning models in precision agriculture cotton yield prediction based on unmanned aerial vehicle imagery and meteorological data
topic UAV multispectral imagery
meteorological data integration
cotton
multimodal deep learning
precision agriculture
url https://www.mdpi.com/2073-4395/15/5/1217
work_keys_str_mv AT chunbojiang multimodaldeeplearningmodelsinprecisionagriculturecottonyieldpredictionbasedonunmannedaerialvehicleimageryandmeteorologicaldata
AT xiaoshuaiguo multimodaldeeplearningmodelsinprecisionagriculturecottonyieldpredictionbasedonunmannedaerialvehicleimageryandmeteorologicaldata
AT yongfuli multimodaldeeplearningmodelsinprecisionagriculturecottonyieldpredictionbasedonunmannedaerialvehicleimageryandmeteorologicaldata
AT ninglai multimodaldeeplearningmodelsinprecisionagriculturecottonyieldpredictionbasedonunmannedaerialvehicleimageryandmeteorologicaldata
AT leipeng multimodaldeeplearningmodelsinprecisionagriculturecottonyieldpredictionbasedonunmannedaerialvehicleimageryandmeteorologicaldata
AT qinglonggeng multimodaldeeplearningmodelsinprecisionagriculturecottonyieldpredictionbasedonunmannedaerialvehicleimageryandmeteorologicaldata