Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #416937
Research Project: Improving Crop Performance and Precision Irrigation Management in Semi-Arid Regions through Data-Driven Research, AI, and Integrated Models

Location: Water Management and Systems Research

Title: Enhancing corn yield prediction with UAV multisource data and machine learning

Author
item Zhang, Huihui
item ZHOU, YUTING - Oklahoma State University
item MA, SHENGFANG - Oklahoma State University

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 8/29/2024
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Accurate corn yield estimation is crucial for US agriculture and global food security. This study explores a novel framework utilizing high-resolution Unmanned Aerial Systems (UAS) data and machine learning to predict corn yield under well-watered and water-stressed field conditions (2022 growing season). Red-Green-Blue (RGB), multispectral reflectance, and thermal (LWIR) imagery were integrated with machine learning models (LASSO regression, Random Forest, Gradient Boosting). All models achieved high accuracy (>0.83 R-squared) for both irrigation scenarios. Notably, combining thermal data with RGB or multispectral data significantly improved yield prediction by capturing water stress responses. Further analysis compared traditional ensemble learning (Random Forest, Gradient Boosting) with deep learning models (ResNet 18, ResNet34, ViT) for yield prediction using RGB and multispectral data across growth stages. Deep learning methods outperformed ensemble methods in early and late stages with RGB data, but performance became comparable in the middle stage. This research provides valuable insights for optimizing corn yield prediction across growth stages, informing agricultural decisions and harvest planning.