Conventional vessel enhancement approaches used in literature are Hessian-based filters, which are sensitive to noise and sometimes give discontinued vessels due to junction suppression. In this paper, we propose a new approach incorporating the use of linear directional features of vessels to get more precise estimates of the Hessian eigenvalues in noisy environment. The directional features are extracted from a set of directional images which are obtained by decomposing the input image using a Directional Filter Bank. In addition, the directional image decomposition helps to avoid junction suppression, which in turn, yields continuous vessel tree. Experimental results show that the proposed filter generates better performance in comparison against conventional Hessian-based approaches.