Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning
The United States plays a pivotal role in global corn production, necessitating accurate corn yield estimation at the field scale for effective and timely field management. This study investigates a novel approach for estimating corn yield in well-watered and water-stressed fields using multisource...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-12-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772375525005751 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849225261838499840 |
|---|---|
| author | Huihui Zhang Yuting Zhou Shengfang Ma Kevin Yemoto |
| author_facet | Huihui Zhang Yuting Zhou Shengfang Ma Kevin Yemoto |
| author_sort | Huihui Zhang |
| collection | DOAJ |
| description | The United States plays a pivotal role in global corn production, necessitating accurate corn yield estimation at the field scale for effective and timely field management. This study investigates a novel approach for estimating corn yield in well-watered and water-stressed fields using multisource data acquired by Unmanned Aerial Systems (UAS). The approach integrates high-resolution Red-Green-Blue (RGB), multispectral reflectance (VNIR), and thermal (LWIR) imagery with machine learning techniques. Machine learning models, including LASSO regression, Random forest (RF), and Gradient Boosting (GB), were applied to predict yield based on data across multiple phenological stages. Results from one of the reproductive stages (R5 stage, at Day of Year (DOY) 259) showed that VNIR data can predict corn yield better than the high-resolution RGB, regardless of the model used. The more complex RF and GB models outperformed the simpler LASSO regression method. Adding LWIR can further improve the performance with RF and GB models, while LASSO did not benefit from this additional information. Moreover, LWIR data helped overcome the overestimation of corn yield in both fully- and deficit-irrigated fields, particularly when using GB. The time series of model performance indicated that the VNIR data could provide stable and accurate corn yield estimation as early as the V9 stage (DOY189) when using RF. In contrast, RGB data exhibited limited effectiveness in predicting corn yield during the reproductive growth stages, emphasizing the necessity of a multispectral sensor during these periods. Across all growth stages, combining LWIR data with VNIR data enhanced the accuracy of corn yield predictions in the deficit irrigated field with RF, especially during the reproductive stages, highlighting the importance of thermal sensors in water-stressed fields. This study presents a novel UAS-based machine learning approach to estimate corn yield in both well-watered and water-stressed fields using multisource and multi-temporal data. The integration of VNIR and LWIR imagery improves yield prediction accuracy, particularly in water-stressed conditions, highlighting the importance of advanced sensing technologies and complex models such as RF and GB. The potential of deep learning models needs to be explored to see if they could further increase the accuracy of corn yield prediction. It would be beneficial to investigate whether the integration of VNIR and LWIR across different sites, years, and cultivars enhances yield performance in the future. |
| format | Article |
| id | doaj-art-7c5b6830cef0425f8dd7f8ba23f6f509 |
| institution | Kabale University |
| issn | 2772-3755 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj-art-7c5b6830cef0425f8dd7f8ba23f6f5092025-08-25T04:14:55ZengElsevierSmart Agricultural Technology2772-37552025-12-011210134410.1016/j.atech.2025.101344Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learningHuihui Zhang0Yuting Zhou1Shengfang Ma2Kevin Yemoto3Water Management and Systems Research Unit, United States Department of Agriculture, Agricultural Research Service, Fort Collins, CO, USA; Corresponding authors.Department of Geography, Oklahoma State University, Stillwater, OK, USA; Corresponding authors.Oklahoma Water Resources Center, Oklahoma State University, Oklahoma City, Oklahoma, USAWater Management and Systems Research Unit, United States Department of Agriculture, Agricultural Research Service, Fort Collins, CO, USAThe United States plays a pivotal role in global corn production, necessitating accurate corn yield estimation at the field scale for effective and timely field management. This study investigates a novel approach for estimating corn yield in well-watered and water-stressed fields using multisource data acquired by Unmanned Aerial Systems (UAS). The approach integrates high-resolution Red-Green-Blue (RGB), multispectral reflectance (VNIR), and thermal (LWIR) imagery with machine learning techniques. Machine learning models, including LASSO regression, Random forest (RF), and Gradient Boosting (GB), were applied to predict yield based on data across multiple phenological stages. Results from one of the reproductive stages (R5 stage, at Day of Year (DOY) 259) showed that VNIR data can predict corn yield better than the high-resolution RGB, regardless of the model used. The more complex RF and GB models outperformed the simpler LASSO regression method. Adding LWIR can further improve the performance with RF and GB models, while LASSO did not benefit from this additional information. Moreover, LWIR data helped overcome the overestimation of corn yield in both fully- and deficit-irrigated fields, particularly when using GB. The time series of model performance indicated that the VNIR data could provide stable and accurate corn yield estimation as early as the V9 stage (DOY189) when using RF. In contrast, RGB data exhibited limited effectiveness in predicting corn yield during the reproductive growth stages, emphasizing the necessity of a multispectral sensor during these periods. Across all growth stages, combining LWIR data with VNIR data enhanced the accuracy of corn yield predictions in the deficit irrigated field with RF, especially during the reproductive stages, highlighting the importance of thermal sensors in water-stressed fields. This study presents a novel UAS-based machine learning approach to estimate corn yield in both well-watered and water-stressed fields using multisource and multi-temporal data. The integration of VNIR and LWIR imagery improves yield prediction accuracy, particularly in water-stressed conditions, highlighting the importance of advanced sensing technologies and complex models such as RF and GB. The potential of deep learning models needs to be explored to see if they could further increase the accuracy of corn yield prediction. It would be beneficial to investigate whether the integration of VNIR and LWIR across different sites, years, and cultivars enhances yield performance in the future.http://www.sciencedirect.com/science/article/pii/S2772375525005751Time-seriesRandom forestGradient boostingLWIRThermalWater deficit |
| spellingShingle | Huihui Zhang Yuting Zhou Shengfang Ma Kevin Yemoto Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning Smart Agricultural Technology Time-series Random forest Gradient boosting LWIR Thermal Water deficit |
| title | Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning |
| title_full | Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning |
| title_fullStr | Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning |
| title_full_unstemmed | Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning |
| title_short | Optimizing corn yield prediction: Integrating multi-temporal UAS data and machine learning |
| title_sort | optimizing corn yield prediction integrating multi temporal uas data and machine learning |
| topic | Time-series Random forest Gradient boosting LWIR Thermal Water deficit |
| url | http://www.sciencedirect.com/science/article/pii/S2772375525005751 |
| work_keys_str_mv | AT huihuizhang optimizingcornyieldpredictionintegratingmultitemporaluasdataandmachinelearning AT yutingzhou optimizingcornyieldpredictionintegratingmultitemporaluasdataandmachinelearning AT shengfangma optimizingcornyieldpredictionintegratingmultitemporaluasdataandmachinelearning AT kevinyemoto optimizingcornyieldpredictionintegratingmultitemporaluasdataandmachinelearning |