Skip to main content
ARS Home » Midwest Area » Columbia, Missouri » Cropping Systems and Water Quality Research » Research » Publications at this Location » Publication #394095

Research Project: Sustainable Intensification of Cropping Systems on Spatially Variable Landscapes and Soils

Location: Cropping Systems and Water Quality Research

Title: Hay yield estimation using UAV-based imagery with machine learning technology

Author
item LEE, KYUHO - University Of Missouri
item Sudduth, Kenneth - Ken
item ZHOU, JIANFENG - University Of Missouri

Submitted to: International Conference on Precision Agriculture Abstracts & Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 5/1/2022
Publication Date: 6/26/2022
Citation: Lee, K., Sudduth, K.A., Zhou, J. 2022. Hay yield estimation using UAV-based imagery and a convolutional neural network. In: Proceedings of the 15th International Conference on Precision Agriculture, June 26-29, 2022, Minneapolis, Minnesota. Available: https://ispag.org/proceedings.

Interpretive Summary: Yield monitoring systems are widely used commercially in grain crops to map yields at a scale of a few meters. However, such high-resolution yield monitoring and mapping for hay and forage crops is not yet available to farmers, as commercial hay yield monitoring systems only obtain the weight of individual bales. This makes it difficult to map and understand the spatial variability in hay yield, as needed by farmers interested in applying precision agriculture principles to hay production. This study investigated the feasibility of an unmanned aerial vehicle (UAV)-based remote sensing system for estimation and mapping of hay yield by machine learning models. Data were obtained by UAV immediately before hay harvest and related to 110 manual measurements of hay yield from throughout the test field. Several datasets containing different combinations of image-derived variables provided good results, representing over 75% of the hay yield variability across the field. The results of this research provide information to aid in selection of an appropriate analysis method for hay estimation using UAV imagery. In future research, the models developed here will be tested for robustness on data from other fields and will also be applied to whole-field imagery for creating hay yield maps.

Technical Abstract: Yield monitoring systems are widely used commercially in grain crops to map yields at a scale of a few meters. However, such high-resolution yield monitoring and mapping for hay and forage crops has not been commercialized. Most commercial hay yield monitoring systems only obtain the weight of individual bales, making it difficult to map and understand the spatial variability in hay yield. This study investigated the feasibility of an unmanned aerial vehicle (UAV)-based remote sensing system for the estimation and mapping of hay yield by machine learning models. Data were obtained during harvest of a 35-ha hay field with mixture of red clover and timothy grass in June of 2021. A RGB camera consisting of three bands (red, blue, and green) attached to a UAV was used to acquire images at a flight height of 20 m For calibration, 110 ground truth hay yield measurements were collected from 1 m2 quadrats. Image features, such as color space components, vegetation indices and texture features, the proportion of grass in samples, and moisture content of samples were extracted from the images or ground truth samples, and were used to estimate the hay mass yield. For yield estimation, a simple random forest machine learning model was trained and tested with the stratified random sampling method using a split ratio of 70:30. Using the recursive feature elimination algorithm, we selected explanatory features for use in the random forest regression model. The most accurate model estimated hay wet mass with r2 = 0.79, RMSE = 251.05 g/m2, and MAE = 187.59 g/m2). The results of this research provide information to aid in selection of an appropriate analysis method for hay estimation using UAV imagery. In future research, the models developed here will be applied to whole-field imagery for creating hay yield maps.