Skip to main content
ARS Home » Southeast Area » Houma, Louisiana » Sugarcane Research » Research » Research Project #418694

Research Project: Evaluation of Small Unmanned Aerial Systems (sUAS) for Sugarcane Monitoring

Location: Sugarcane Research

Project Number: 6052-21000-017-003-N
Project Type: Non-Funded Cooperative Agreement

Start Date: Aug 1, 2018
End Date: Jul 31, 2023

a) Identify the right sensor (RGB, multispectral, hyperspectral) and airframe (fixed wing v/s rotary) combination for sugarcane monitoring. b) Develop sUAS derived vegetation index metrics that best relates to sugarcane yield. c) Identify an optimal spatial and temporal resolution that is suitable for sugarcane growth and yield model development. d) Correlate sUAS derived crop heights and vegetation index to yield and biomass density. e) Compare conventional field sampling with sUAS based estimation techniques.

1) Field Study: A commercial sugarcane field in a near proximity to the USDA, ARS Sugarcane Research Unit in Houma, Louisiana, and Nicholls University will be selected for the study. Data will be collected with both fixed wing and rotary wing sUAS on a monthly basis beginning in June. The following sensors will be tested 1) SONY a7r 36.4 megapixel true color and infrared 2) MicaSense RedEdge multispectral and; 3) OceanOptics hyperspectral sensor. We will be using Trimble UX5HP and UX5MS fixed wing airframes; Trimble ZX5 and DJI Phantom rotary wing airframes in this project. 2) Image Preprocessing and Analysis: The use of sUAS captured digital image data requires several preprocessing procedures. A photogrammetric workflow is executed to generate a mosaic of the study area. Various vegetation indices will be derived from the orthomosaics and used as an additional image layer to analyze the plot yield by using image classification techniques. The image derived yield map will be correlated with field data. Similarly, the digital surface model will be used to derive crop heights which will be used to model the sugarcane biomass density. 3) Model development: The digitally derived data from sUAS images will be integrated with ground sampling plot data to model sugarcane yield and biomass maps in a Geographic Information System (GIS) environment. The conventional sampling data will serve two folds: namely to assess the accuracy of the sUAS digitally derived data and to refine the overall yield and biomass model. A second year follow up study will be required to test the replicability of the developed model.