Skip to main content
ARS Home » Midwest Area » Columbia, Missouri » Cropping Systems and Water Quality Research » Research » Publications at this Location » Publication #381632

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

Location: Cropping Systems and Water Quality Research

Title: Early corn stand count of different cropping systems using UAV-imagery and deep learning

item VONG, CHIN NEE - University Of Missouri
item CONWAY, LANCE - University Of Missouri
item ZHOU, JIANFENG - University Of Missouri
item Kitchen, Newell
item Sudduth, Kenneth - Ken

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/13/2021
Publication Date: 5/23/2021
Citation: Vong, C., Conway, L.S., Zhou, J., Kitchen, N.R., Sudduth, K.A. 2021. Early corn stand count of different cropping systems using UAV-imagery and deep learning. Computers and Electronics in Agriculture. 186. Article 106214.

Interpretive Summary: Corn seed sometimes does not germinate well nor emerge uniformly from one plant to another, particularly when soils are cold and wet. When corn plants do not emergence at about the same time (e.g., within a day or two of each other) or when some seeds don’t emerge at all, grain yield can be reduced. Faced with these conditions farmers then have to decide whether they should replant portions or all of a field. To help in making replanting decisions, automated tools are needed to help farmers quickly evaluate early-season corn stands. This research evaluated unmanned aerial vehicle (UAV) images for estimating early-season corn stand in three different cropping systems (CS). These CS differed by tillage, rotation, and use of cover crops. One of the outcomes of this research was the development of a sophisticated modeling method that processed the UAV image details in a way that plant, soil, and dead plant residues could be identified. This step was needed before the small corn plants could be counted. Results were very good for all three CS, explaining more than 90% of the variation in corn stand. However, the CS that included a tillage operation before planting was best. Next in performance order was a no-till CS, and finally a no-till CS with cover crops. While the use of UAV imagery and modeling for estimating early corn stand count was successful, its accuracy was affected by soil and crop management practices. If this process could be more completely automated, farmers could utilize UAV scouting to identify fields or portions of fields that need replanting.

Technical Abstract: Optimum plant stand density and uniformity is vital in order to maximize corn (Zea mays L.) yield potential. Assessment of stand density can occur shortly after seedlings begin to emerge, allowing for timely replant decisions. The conventional methods for evaluating an early plant stand rely on manual measurement and visual observation, which are time consuming, subjective because of the small sampling areas used, and unable to capture field-scale spatial variability. This study aimed to evaluate the feasibility of an unmanned aerial vehicle (UAV)-based imaging system for estimating early corn stand count in three cropping systems (CS) with different tillage and crop rotation practices. A UAV equipped with an on-board RGB camera was used to collect imagery of corn seedlings (~14 days after planting) of CS, i.e. minimum-till corn-soybean rotation (MTCS), no-till corn-soybean rotation (NTCS), and no-till corn-corn rotation with cover crop implementation (NTCC). An image processing workflow based on a deep learning (DL) model, U-Net, was developed for plant segmentation and stand count estimation. Results showed that the DL model performed best in segmenting seedlings in MTCS, followed by NTCS and NTCC. Similarly, accuracy for stand count estimation was highest in MTCS (R^2 = 0.95), followed by NTCS (0.94) and NTCC (0.92). Differences by CS were related to amount and distribution of soil surface residue cover, with increasing residue generally reducing the performance of the proposed method in stand count estimation. Thus, the feasibility of using UAV imagery and DL modeling for estimating early corn stand count is qualified influenced by soil and crop management practices.