Background
Fingerprinting is one of the oldest and most consistently used forms of physiological biometrics. [1] Every single person has their own set of unique fingerprints, making fingerprint biometrics an excellent and proven way of identifying individuals.
The uniqueness of fingerprints is due to the series of ridges and furrows on the fingers. The pattern of the ridges and furrows is identified using the Henry Classification in most English-speaking countries. The three basic classifications for fingerprints are arch, loop, and whorl. These can be broken down into further sub-categories. Other unique points on the finger called minutiae points occur at ridge bifurcation or end points.
 |
 |
 |
| Arch | Loop | Whorl |
Matching Techniques
Fingerprint matching techniques fall into one of two categories: minutae-based and correlation-based. Minutae-based matching maps the location of minutiae points on the finger. However, low quality prints may make this unreliable. Minutae-based matching also does not take into account the ridges and furrows of the finger. The correlation-based method measures the unique points of the fingerprint relative to a registration point. However, the registration point is affected by image translation and rotation, making it difficult to assure proper alignment.
All fingerprint matching based on minutiae is made difficult by different sized minutiae patterns. Also, the ridge structure cannot be totally defined by the minutiae. [2]
 |
Advantages
There are many major advantages in the use of fingerprints for biometric security, such as:
- Inexpensive
- Fast results
- Reliable
- Established systems
- Effective
All of these reasons make fingerprinting one of the best methods for quickly identifying individuals. [3]
Problems
Some criticisms of fingerprinting as a “bulletproof” identification method have emerged. Many people feel that the subjectivity of collecting and identifying the unique characteristics of fingerprints makes them unreliable as the sole method of identification. This is an issue mostly associated with criminal identification, although false positives and negatives could be a major problem for business security. [4]
Also, as portrayed in many movies, it is possible to fake fingerprint identification by making imprints on wax-like substances. This is not widespread but it is a possible threat.
Implementation Enhancement
Fingerprinting can be done on a single or multiple fingers, making the ability to scale the verification excellent. Fingerprint identification could also easily be combined with hand geometrics to produce a more effective method of scanning. This method could potentially eliminate the risk of fingerprint falsification.
Business Application
Fingerprinting is already a very popular method of identification in business due to its inexpensive and effective implementation. One of the best examples of a current widespread use is on IBM/Lenovo Thinkpads. The fingerprint implementation on Thinkpads allows users to replace all their passwords with a simple fingerprint scan, including BIOS, Windows, wireless and web passwords. This simple, yet effective system ensures that confidential business information is kept secure. [5]
Fingerprint identification systems are also in place for high-security installations and even in some public schools. [6]
