Use of non-invasive RGB imaging to assess the canopy status in organic viticulture

Abstract

In organic viticulture, canopy features such as leaf area, canopy porosity and fruit exposure are critical due their impact on fungal disease incidence and grape composition. An adequate and accurate assessment of the canopy status is the first step towards appropriate and effective grapevine canopy management, therefore an easy, non-invasive, robust method to evaluate the main features of a grapevine canopy is needed. In this work a protocol for canopy status assessment based on non-invasive RGB imaging is presented and used to ascertain differences in the canopy status of grapevines grown under different degraded and non-degraded soil conditions. RGB images were processed using a classification algorithm based on the Mahalanobis distance, and then the pixels were classified in four classes: clusters, leaves, gaps and shoots. Overall, higher leaf exposure was observed in vines grown in non-degraded soil while the highest percent of gaps or canopy porosity corresponded to vines of the two cover crop treatments.