Fingerprint Advances Will Fight Cyber Crime
A team of University-at-Buffalo biometrics researchers has determined the degree of security afforded by fingerprint scans that typically only capture a partial print, which had been a major stumbling block for the practical implementation of biometric identification in lieu of passwords.
“Thus research paves the way toward efficient methods of preventing unauthorized access to handheld devices, such as cell phones, wireless handheld devices, and electronic audio players, as well as to secure Web sites,” said Buffalo computer science and engineering professor Venu Govindaraju, adding that the technology could also have applications in forensics.
The researchers’ technique, the Automated Partial Fingerprint Identification System, defines the keypad sensor dimensions to specify the level of required security, pinning down for the first time how much of a fingerprint that is required to provide a degree of security comparable to a six-letter password.
The system relies on an algorithm that determines if two images are a close enough match to verify the identity, recognizing that fingerprints and most other biometric information are incomplete, and that, unlike passwords, biometric data can be slightly different with each use.
Securely matching biometric scans requires an algorithm that can adjust for factors that could cause variations in a person’s fingerprints, such as how firmly the person pressed and the level of moisture in the finger.
Govindaraju notes that the system relies on a transformation of the image, rather than the image of the fingerprint itself, making reverse engineering all but a mathematical impossibility.
Abstracted by the National Law Enforcement and Corrections Technology Center(NLECTC) from the University at Buffalo News (02/22/06).