A Bayesian framework for joint structure and colour based pixel-wise classification of grapevine proximal images

•Joint structure-colour features based on structure tensors are efficient modelling tools.•Models enable a performant pixel-wise classification of. leaves, fruit, flowers and stems.•A unique framework can be applied to various phenological stages.•Precision and recall rates reach between 85% and 95%...

Ausführliche Beschreibung