Authors : B. Jayaraman , C. Puttamadappa , E. Anbalagan , E. Mohan and Srinivasarao Madane
Abstract: The objective of this research is to incorporate finger print authentication with the widely used PIN authentication scheme currently used for online transactions, to achieve a foolproof system. A complete minutiae extraction scheme for automatic fingerprint recognition systems with intelligent method is presented. A critical step is to reliably extract minutiae from the fingerprint images and process with artificial neural network method. Fingerprint images are rarely of perfect quality. They may be degraded and corrupted due to variations in skin and impression conditions. So, fingerprint minutiae enhancement techniques are employed prior to minutiae extraction to obtain a more reliable estimation of minutiae locations. In the first stage, image normalization and orientation field of the fingerprint are calculated. The local orientation of the ridges serves as the parameter for the next processing stages. Details of the adaptive morphological filtering used for ridge extraction and background noise elimination are described. Minutiae patterns of the scanned and stored fingerprints are then compared and evaluation results are obtained by processing through supervised artificial neural network. Greater security is achieved with PIN authentication.
B. Jayaraman , C. Puttamadappa , E. Anbalagan , E. Mohan and Srinivasarao Madane , 2008. Fingerprint Authentication Using Back-Propagation Algorithm. International Journal of Soft Computing, 3: 282-287.