Skip to main content
ARS Home » Plains Area » Akron, Colorado » Central Great Plains Resources Management Research » Research » Publications at this Location » Publication #317718

Research Project: Sustainable Dryland Cropping System for the Central Great Plains

Location: Central Great Plains Resources Management Research

Title: Increasing the accuracy and automation of fractional vegetation cover estimation from digital photographs

Author
item COY, ANDRE - University Of The West Indies
item RANKINE, DALE - University Of The West Indies
item TAYLOR, MICHAEL - University Of The West Indies
item Nielsen, David
item COHEN, JAN - University Of The West Indies

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/22/2016
Publication Date: 6/23/2016
Publication URL: http://handle.nal.usda.gov/10113/62749
Citation: Coy, A.D., Rankine, D.R., Taylor, M.A., Nielsen, D.C., Cohen, J. 2016. Increasing the accuracy and automation of fractional vegetation cover estimation from digital photographs. Remote Sensing. 8/474-487.

Interpretive Summary: Estimates of canopy cover are essential for quantifying crop growth and development and are used to calibrate and validate crop models. Canopy cover estimates can be made with inexpensive digital cameras, but require analysis to determine the fraction of the photograph that is covered by leaf area. This paper describes the development, validation and use of the Automated Canopy Estimator to determine canopy cover estimates for oat, corn, rapeseed, and flax. The ACE was found to outperform eight other methods for accurately and automatically analyzing digital photographs for canopy cover.

Technical Abstract: The use of automated methods to estimate canopy cover (CC) from digital photographs has increased in recent years given its potential to produce accurate, fast and inexpensive CC measurements. Wide acceptance has been delayed because of the limitations of these methods. This work introduces a novel technique, the Automated Canopy Estimator (ACE) that overcomes many of these challenges to produce accurate estimates of canopy cover using an unsupervised segmentation process. ACE is shown to outperform eight other segmentation algorithms in the segmentation of photographs of four different crops (oat, corn, rapeseed and flax) with an overall accuracy of 89.2 % averaged across all four crops. ACE is also shown to produce CC estimates that are strongly correlated with ground truth CC values.