ENHANCING THE ACCURACY OF FINGERPRINT RECOGNITION

Abstract

In the era of the digital world everyone and everything is digitally connected with each other with the help of personnel digital assistants and internet of things (IOT).This will also give rise to the problem of identity theft whether making a banking transaction, sharing a confidential file, or making a fraud with company. So a highly secure system is thus required. Biometrics plays a dominant role for providing the security it includes face recognition, iris recognition and fingerprint recognition among all of this fingerprint provide high level of abstraction to the user security. We have discussed some approaches, comparison of the existing enhancement techniques and by using different machine learning algorithms how we can increase the accuracy of present recognition system. In this project I have used the standard dataset fvc2002 db1 and applying different machine learning algorithms on this database by first without applying the principal component analysis and then with principal component analysis (PCA) and realized that random forest classifier(without PCA) produce the best results among all of this algorithms.