The Benefit of Color Information in Digital Image Analysis for the Estimation of Legume Contribution in Legume–Grass Mixtures

Abstract

ABSTRACTEfficient detection of spatial legume dry matter (DM) contribution in legume–grass mixtures is of great importance for targeted management of legume‐based swards. Legume coverage may be an appropriate indicator for the contribution of legumes, because it can be assessed by on‐the‐go systems with digital image analysis (DIA). To examine the perspectives of DIA a pot experiment, under controlled conditions, was conducted across a wide range of legume species (white clover [Trifolium repens L.], red clover [Trifolium pratense L.], and alfalfa [Medicago sativa L.]), legume proportion (0–800 g kg−1), and growth stage (start of tillering to start of heading). In this study, an advanced procedure for the determination of legume DM contribution by DIA is suggested. The DIA procedure comprises the analysis of color images and applies an advanced function to predict legume DM contribution from legume coverage by considering total sward biomass. This resulted in an accurate prediction of legume contribution (grams per kilogram) with R2 of 0.90, 0.94, and 0.93 for red clover, white clover, and alfalfa, respectively. For validation, swards of a field experiment were used. It showed that legume detection is possible, but for practical field application some further adjustments are necessary.