An Improved Binarization Based Algorithm using Minutiae Approach for Fingerprint Identification
S. M Rajbhoj1, P. B. Mane2
1S. M. Rajbhoj, Electronics Dept, Bharati Vidyapeeth Deemed University, Pune, M.S, India.
2Dr. P. B. Mane, Principal, AISSMS IOIT, Pune, M.S. India.
Manuscript received on July 17, 2012. | Revised Manuscript received on August 25, 2012. | Manuscript published on August 30, 2012. | PP: 219-222 | Volume-1 Issue-6, August 2012. | Retrieval Number: F0671081612/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The long history of fingerprint, their extensive use in forensics and with need of automatic personal identification in recent years, fingerprints is receiving a lot of attention. There is misconception that fingerprint identification is a fully solved problem. However numerous fingerprint systems currently available which use minutiae based approach still do not meet performance requirement of several civilian applications. Performance of these systems degrades with deterioration in the quality of fingerprint image. In absence of an a priori enhancement step most of the binarization based techniques do not provide satisfactory results when applied to low quality images. Thus trying to eliminate these shortcomings we present an improved approach for fingerprint recognition providing accurate automatic personal identification. In this approach we use optical sensor which captures image of excellent quality with large capture area and superior reliability. The recognition algorithm first use histogram equalization technique to improve the global contrast of an image, then enhancement of the image is done by an efficient enhancement technique. We then use binarization based method to extract minutiae. False minutiae are removed using thresholding technique. The matching is based on determining the total number of matched minutiae based on Euclidian distances. This system is tested on two different databases. The experimental result shows that incorporating a fast enhancement technique and using an optical scanner increase the accuracy of the system for lower values of False accept rate.
Keywords: Fingerprint identification, Minutiae, Enhancement, Binarzation, Extraction, Thresholding, Euclidian distances.