Résumé
Maize is a pillar of the French agricultural system, providing biomass for animal feed which is crucial for milk production. The maize stem comprises distinct tissues with unique cell-wall matrices that affect dry matter digestibility, which impacts milk production, and respond differently to drought1. During harvest, the stem is ground into particles. These particles are either pure (composed of a single tissue) or mixed (composed of multiple tissues, whose proportion could explain a part of their digestibility). In direct relation to their specific biochemical composition, the unique autofluorescence properties of these tissues have been emphasised2. Multispectral imaging was successfully used to identify the given tissue on maize microscopic cross-sections by exploiting these autofluorescence properties, but its potential must be evaluated on dried particles found in the agricultural sector. To study this, particles were produced from isolated pure tissues extracted from stems grown from a panel of forage maize inbred lines and hybrid varieties cultivated in Southern France. In this poster, we present early results regarding differences between the tissue types observed through multispectral images. We then explore how these spectral differences could feed predictive models of particle tissular origin, and discuss how it could complement tissue-specific biochemistry and near-infrared spectroscopy methods routinely used in forage quality evaluation. We conclude that multispectral imaging could provide complementary information to breeding programs, offering a new dimension for improving forage maize digestibility.