Abstract. Among the principal problems for realizing a robust Automated Fingerprint Identification System (AFIS) there are the images quality and matching algorithms. In this paper a fingerprint enhancement algorithm based on morphological filter and a triangular matching are introduced. The enhancement phase is based on tree steps: directional decomposition, morphological filter and composition. For the matching phase a global transformation to overcame the effects of rotation, displacement and deformation between acquired and stored fingerprint is performed using the number of similar triangular, having fingerprint minutiae as vertexes. The performance of the proposed approach has been evaluated on two set of images: the first one is DB3 database from Fingerprint Verification Competition (FVC) and the second one is self collected using an optical scanner. I. INTRODUCTIONAn Automated Fingerprint Identification System (AFIS) can work in two different ways [6] [7]:• Verification mode: the on-line biological signature is compared with the registered signature. With more details, a username is used to select the database item and perform a 1 -> 1 match.• Identification mode: the on-line biological signature is compared with the registered signatures. With more details 1 -> n matches must be performed in order to select the candidate with the highest matching score. Fingerprint are composed by a unique pattern of locally parallel ridge and valley with well defined orientation. To compare fingerprints in an Automatic Fingerprint Identification System (AFIS), local ridge structure characteristics, called minutiae, are usually used. The minutiae used in an AFIS are ridge ending and ridge bifurcation.Generally, the AFIS performance depends by image quality. The ridge-valley structures are rather well visible in a poor quality fingerprint image because a wrong fingerprint impression produces imperfections, genuine features alteration and many false features introduction.In addition, an incorrect image acquisition due to a wrong finger displacement produces partial fingerprint acquisition, roto-translation displacements and non linear deformation.In this paper, a directional morphological filter, in image enhancement process for improving the clarity of ridge/valley structures, is applied. In addiction, an enhanced matching algorithm with respect to [14], has been used to overcame the effects of rotation, displacement and deformation between acquired and stored fingerprint. The matching score is related to the number of similar triangular, having fingerprint minutia as vertexes, in the acquired and the stored images.In literature different approaches for fingerprint image enhancement were proposed [19][20]. Usually dedicated enhancement algorithms are time consuming applications. Most of fingerprint image enhancement approaches are based on Gabor filter [17] [18]. The Gabor approach is a frequency-selective and orientation-selective filter tuned by ridge directions and ridge frequencies in fingerprint. It works both on spa...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.