School of Advanced Sciences, VIT University, Vellore 632014, India Finger vein recognition is a cost efficient and convenient biometrics having commercial applications. Expert enhancement techniques and reliable matching algorithms are essential stages in the process of acquiring reliable finger vein patterns for identification. This work proposes a finger vein matching system using scale invariant feature descriptors for matching after enhancing the finger vein images using a fuzzy theory based algorithm. A fuzzy aggregate operator is used for combining the membership function for enhancement. Features extracted for matching are translational and scale invariant and a matching score based feature matching is proposed. Error rates of the proposed matching is observed to be much lower than that of the existing SURF matching and minutiae matching algorithms. Performance of the three algorithms are analysed using the ROC curves for changes in translation and the results emphasize the efficiency of the proposed algorithm in translation invariance. This study emphasise that designed finger vein matching system is an effective solution to the poor quality as well as the translation issues that are usually present in the finger vein images. Copyright © 2018 American Scientific Publishers. All rights reserved.