A SIFT-Based Fingerprint Verification System Using Cellular Neural Networks

Recently, with the increasing demand of high security, person identification has become more and more important in our everyday life. The purpose of establishing the identity is to ensure that only a legitimate user, and not anyone else, accesses the rendered services. The traditional identification methods are based on “something that you possess” and “something that you know” such as key, user-ID, password, PIN, etc. Examples of such applications include secure access to buildings, airports, computer systems, cellular phones and ATM machines. Another family of identification methods uses biometric characteristics. Biometric recognition, or simply biometrics, refers to the automatic recognition of individuals based on their physiological and/or behavioral characteristics. Biometrics allows us to confirm or establish an individual’s identity based on who she is, rather than by what she possesses (e.g., an ID card) or what she knows (e.g., a password). Current biometric systems make use of identifiers such as fingerprints, hand geometry, iris, face and voice to establish an identity. Biometric systems also introduce an aspect of user convenience. For example, they alleviate the need for a user to remember multiple passwords associated with different applications. Fingerprint characterization is the oldest and the prevalent member of the biometric family and has been extensively used for person identification in a number of commercial, civil and forensic applications. The question that is being asked about biometric technologies in general and about fingerprints in particular is that whether these technologies can work all the time, everywhere, and in all contexts for reliable person identification and authentication. One of the design criteria for building such completely automatic and reliable fingerprint identification (and verification) systems is that the underlying sensing, representation, and matching technologies must also be very robust. In practice, due to variations in impression conditions, ridge configuration, skin conditions (aberrant formations of epidermal ridges of fingerprints, postnatal marks, occupational marks), acquisition devices and non-cooperative attitude of subjects a significant percentage of acquired fingerprint images is of poor quality. In order to ensure that the performance of a feature extraction algorithm will be robust with respect to the quality of input fingerprint images, an enhancement algorithm which can improve the clarity of the ridge structures is useful. Most of the fingerprint image enhancement methods (Gabor, directional or anisotropic filter based) use convolution to obtain the results. Another way to address these O pe n A cc es s D at ab as e w w w .ite ch on lin e. co m