Your browser version is outdated. We recommend that you update your browser to the latest version.

Manuscript Title: Implementation of Easy Fingerprint Image Authentication with Traditional Euclidean and Singular Value Decomposition Algorithms

Author : M. James Stephen, P.V.G.D Prasad Reddy


Abstract: Fingerprints are impression of the friction ridges of the finger. They are used as biometric feature for person identification and verification in the field of biometric identification. In spite of decades of research in fingerprints, reliable finger print recognition is still an open problem. Minutia based fingerprint recognition algorithms have been widely accepted as a standard for single finger recognition applications.  This technology has proved to be a reliable form of enrollment and matching in a corporate environment under ideal circumstances. In this paper, an algorithm based on SVD (singular value decomposition) is proposed for feature extraction of a fingerprint and Identification of finger prints is done by using Traditional Euclidean Distance. This paper consists of three main stages    (i) Data Acquisition (ii) Feature Extraction and (iii) Authentication / Identification. The image acquisition is done by means of Futronic FS88 fingerprint Scanner. The obtained image is of high quality, 256 gray scale values in every single pixel. The algorithm developed for feature extraction of a fingerprint is based on singular values which are obtained by performing SVD. Identification of finger prints is done by using Euclidean Distance Algorithm. Our experimental results are encouraging. The developed system is light-weighted and simple. The experimental results show that the developed system is efficient and reliable.

Keywords: Image Processing, Finger Print Identification, Single Value Decomposition algorithm, Euclidean distance.

Vol 3 (2)