Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Crop Production Systems Research » Research » Publications at this Location » Publication #324137
Title: Cotton yield estimation using very high-resolution digital images acquired on a low-cost small unmanned aerial vehicle

Author
item Huang, Yanbo
item BRAND, HOWARD - Virginia Tech
item Sui, Ruixiu
item Thomson, Steven
item FURUKAWA, TOMONARI - Virginia Tech
item EBELHAR, M - Mississippi State University

Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/25/2016
Publication Date: 1/5/2017
Citation: Huang, Y., Brand, H., Sui, R., Thomson, S.J., Furukawa, T., Ebelhar, M.W. 2017. Cotton yield estimation using very high-resolution digital images acquired on a low-cost small unmanned aerial vehicle. Transactions of the ASABE. 59(6):1563-1574.

Interpretive Summary: Cotton yield estimation is important in cotton farming management. However, traditional methods for cotton yield estimation are time-consuming, laborious and difficult to cover large cotton fields. Remote sensing provides techniques to develop quick coverage over a field of any size for accurate cotton yield estimation. Scientists in USDA-ARS Crop Production Systems Research Unit, Stoneville, Mississippi, Virginia Tech and Mississippi State University collaboratively developed two new methods for estimating cotton yield using very high-resolution digital images acquired on a low-cost miniature unmanned aerial vehicle (UAV): 1) estimate cotton yield through estimating cotton plant height using 3D data generated from the UAV images over the cotton field and 2) estimate cotton yield through estimating cotton plot unit coverage using advanced Laplacian transform of the differentiated UAV images over the cotton field. The results indicated that the method 1 could well simulate the relationship between cotton plant height and the yield and the method 1 could directly estimate the yield well with little bias. This study provided proof and demonstrated the potential of UAV image data for accurate estimation of cotton yield at different growth stages of cotton.

Technical Abstract: Yield estimation is a critical task in crop management. A number of traditional methods are available for crop yield estimation but they are costly, time-consuming and difficult to expand to a relatively large field. Remote sensing provides techniques to develop quick coverage over a field at any scale for crop yield estimation. Satellite remote sensing is often used for large-scale earth observation. Remote sensing with manned airplanes flying at high altitudes cannot achieve the spatial resolution required for field-scale precision farming. Ground systems are typically used for point-measurements and their operation is often restricted to field conditions. UAVs (Unmanned Aerial Vehicles) provide a unique platform for high-resolution remote sensing and UAV-based remote sensing systems can be used to estimate crop yield in a cost-effective manner. The objective of this study was to develop and evaluate new methods for estimation of cotton yield for precision cotton farming. Experimental plots were laid out in a cotton field near Stoneville, Mississippi in 2014. Nitrogen fertilizer was applied to the plots at five different rates to generate cotton yield variation. Two methods were employed to estimate cotton yield using very high-resolution digital images (2.7 cm pixel-1) acquired from an inexpensive, small multirotor UAV: 1) Use 3 dimensional point cloud data derived from multiple digital images of the cotton field to estimate cotton plant height, and hence estimate yield; and 2) Segment cotton boll signatures from the background of the digital images of the defoliated cotton field just prior to harvest, and with the estimated cotton plot unit coverage, estimate the yield. The results indicated that inexpensive miniature UAV-based low-altitude remote sensing can be used to estimate cotton yield well through estimation of plant height (R2=0.43 compared with R2=0.42 for yield estimation through manually measured plant height). The results further indicated that the method can offer reliable cotton yield estimation through estimation of cotton boll coverage in each plot with Laplacian image processing while considering a few plots with poor light condition as outliers (R2=0.83). This study could benefit yield estimation of cotton, with similar methods used for other crops, in agricultural research and crop production.