An Algorithm for Fingerprint Classification Using Template Matching Technique

Automatic fingerprint classification has received considerable attention over the past decade. Despite significant progress in this field, there are still rooms for improving the classification operation by continuing study and research in this field. This thesis describes a study of fingerprint classification using template matching technique. We have classified the fingerprints in four groups according to their pattern, which are Arch, Left loop, Right loop, and Whorl. We have discussed and explained the specification and the limitations of the fingerprint classification (the effect of corrupted and rotated input fingerprints on the accuracy of the classification operation). The thesis has analysed the mentioned technique and evaluated its strengths and limitation by comparing this technique with the singularities technique. This research has also included the pre-processing stage, which consist of enhancement, segmentation, and thinning of fingerprints.

Text FK_2002_26.pdf Download (1MB)

Actions (login required)

View Item