Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Using RGB-based vegetation indices for monitoring guayule biomass, moisture content and rubber

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2016 ASABE Annual International Meeting  162380922.(doi:10.13031/aim.20162380922)
Authors:   Diaa Eldin M. Elshikha, Douglas Hunsaker, Kevin Bronson, Paul Sanchez
Keywords:   Biomass; camera based vegetation indices; guayule; rubber
Abstract. We investigated the use of a ground based digital camera to monitor plant moisture content, biomass and rubber content of guayule. These parameters are typically assessed using destructive plant sampling and lab analyses, which are labor intensive and time consuming. A method to predict these parameters by camera-based (red-green-blue [or RGB]) vegetation indices could offer an effective and affordable alternative to destructive sampling. RGB indices derived from camera images were related to plant moisture content, rubber content, and plant dry biomass weights using linear relationships. Certain vegetation indices derived from image color values were strongly correlated with percent plant moisture content, including the NGrRI index and the BGR index that had correlation coefficients (r) that ranged from 0.77 to 0.87. The RBNI index and the MRBI had the highest correlations (|r|>0.81) with plant biomass. The MSGrR index had the highest correlation with rubber content (r=0.82), however, the index also had essentially the same correlation as for plant biomass. Results suggest the possibility of utilizing RGB indices developed using data from a digital camera to monitor guayule plant moisture content and to estimate plant biomass. While rubber content did correlate to some indices, the strong multicollinearity of these indices to biomass indicates it may be difficult to separately predict rubber content.

(Download PDF)    (Export to EndNotes)