Digital Image Processing

P. Esther Rani ¹, B. Ratheesh Kannan ², S. Mohammed Ali ³, K. Chandirasekar ²

Affiliations
1 ECE Department, Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala Engineering College, Chennai, IN
2 ECE Stream, Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala Engineering College, Chennai, IN
3 Department of ECE, Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala Engineering College, Chennai, IN

Abstract

This paper proposes an efficient algorithm for fingerprint matching based human identification. Fingerprint biometrics finds its special position in biometric based identification due to the fact that they are easy to obtain and most importantly they are scientifically unique, reliable and universally accepted. A major approach in fingerprint recognition now-a-days is based on the distance metrics based minutiae matching. One important problem is false spurious minutiae in the finger features. One main reason for these false minutiae is presence of noise. In our proposed system we remove the noise by adaptive wiener filtering technique. Morphological thinning plays a main role in proposed system for thinning the ridges. We have also presented an accurate fingerprint identification technique for minutia extraction and a minutia matching. For minutia marking we considered two steps in post processing to remove false minutia and for minutiae matching we considered alignment based relax box matching method. MATLAB 7.9 is used for system verification. Experimental results show that this method achieves a much better and efficient matching performance.

Keywords

Adaptive Wiener Filter, Biometrics, Fingerprint Matching, MATLAB, Minutiae, PolyU.

Full Text

     

P. Esther Rani ¹, R. Shanmuga Lakshmi ²

Affiliations
1 Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala College of Engineering, Chennai, IN
2 Government College of Technology, Coimbatore, IN

Abstract

This paper presents a high performance palmprint personal authentication system implementation method with number of experimental result. A major approach for palmprint recognition today is to extract feature vectors corresponding to individual palmprint images and to perform palmprint matching based on some distance metrics. One of the difficult problems in feature-based recognition is that the matching performance is significantly influenced by many parameters in feature extraction process, which may vary depending on environmental factors of image acquisition. The Zernike moment feature extraction plays the main role in the proposed system for different orientation hand image ROI identification. Unsharp filtered palmprint images makes possible to achieve highly robust palmprint recognition. MATLAB 7.9 is used for the system verification. Experimental evaluation using a palmprint image database clearly demonstrates an efficient matching performance of the proposed system.

Keywords

PolyU, Pattern Matching, Unsharp Filter, MATLAB, Zernike Moments, Palmprint.

Full Text