A visual attention guided unsupervised feature learning for robust vessel delineation in retinal images.

•A novel visual attention guided unsupervised feature learning (VA-UFL) approach.•VA-UFL exploits both visual attention mechanism and multi-scale contextual information.•Results are superior to state-of-the-art methods on five publicly available datasets.•Method is robust against central vessel reflex, complex crossovers and pathological lesions.