Adaptive and Faster Approach to Fingerprint Minutiae Extraction and Validation

Fingerprint is a very vital index in the field of security. Series of Automatic FingerprintIdentification Systems (AFIS) have been developed for human identification. These systemscompare each of the features of a template fingerprint image with each of the features in thefeature sets in the reference database to determine whether the template and each of thereference images are from the same source. Comparison is done on the basis of presetparameters such as feature type, location, orientation and so on. Getting the features used for theconstruction of a reference database from the images involve the implementation of a soundfingerprint feature detection, validation and extraction algorithm. In this paper, the process ofdetecting and extracting true and false feature points in a fingerprint image is discussed. Attentionis also given to the elimination of the false feature points through the process of validation. Someof the existing fingerprint feature extraction and validation algorithms were firstly modified and theresulting algorithms were implemented. The implementation was carried out in an environmentcharacterized by Window Vista Home Basic as platform and Matrix Laboratory (MatLab) asfrontend engine. Fingerprints images of different qualities obtained from the manual (ink andpaper) and electronic (fingerprint scanner) methods were used to test the adequacy of theresulting algorithms. The results obtained show that with the modified algorithm, valid and trueminutiae points were extracted from the images with greater speed and accuracy.