An FPGA-based Embedded System For Fingerprint Matching Using Phase Only Correlation Algorithm

There is an increasing interest in inexpensive and reliable personal identification in many emerging civilian, commercial and financial applications. Traditional systems such as passwords, PINs, Badges, Smart Cards and Tokens may either be stolen or easy to guess but also to forget, in same cases they may be lost by the user who carries them; all this can lead to identified. Fingerprint-based identification is one of the most used biometric techniques in automated systems for personal identification and it is becoming socially acceptable and cost-effective, since a fingerprint is univocally related to a particular individual. Typical fingerprint identification methods employ feature-based image matching, where minutiae points in the ridge lines (i.e., ridge endings and bifurcations) are identified. Unfortunately this approach is highly influenced by fingertip surface condition. Fingerprint recognition is a complex pattern recognition problem. The efforts to make automatic the matching process based on digital representation of fingerprints, led to the development of Automatic Fingerprint Identification Systems (AFIS). Typically, there are millions of fingerprint records in a database which needs to be entirely searched for a match, to establish the identity of the individual. In order to provide a reasonable response time for each query, it will be better to develop special hardware solutions to implement matching and/or classification algorithms in a really efficient way. In this work we realised a system able to outperform modern PCs in recognising and classifying fingerprints and based on FPGA technology.