In 1996, Walt Disney World Resort in Florida installed fingerprint scanners to keep track of their season pass holders. This was put into place to combat fraud and to keep malevolent people from entering the theme park. After the terrorist attacks of September 11, 2001, many organizations followed suit and are now using various biometric devices to ensure the public’s safety and the nation’s security. Biometric devices use an individual’s physical and behavioral characteristics to prove their identity. To date, biometric technology can determine a person’s identity based on their iris, fingerprint, palm, hand geometry, signature, face, voice, and gait.
Biometric Data Can be Reconstructed from Biometric Templates
The companies that make biometric technologies have expressed that the devices will insure optimum protection because of its low cost, high accuracy, and its ability to keep people from masquerading as others. What they often fail to mention is how people’s biometric data will be stored within the database system. When people are enrolled into a biometric surveillance system, a biometric template is stored of them. This template is a digital representation of the person’s distinct characteristics, representing the information that was extracted from the biometric sample. Which means, for example, the scan of the individual's fingerprint is not stored, but instead specific data points called minutiae are captured from the picture and saved.
The way biometric templates are stored can vary between models and vendors. In general, the person’s data is stored as specific points and a mathematical algorithm is used to confirm or deny a person’s access. The primary concerns that arise from the storage of these templates are:
- can they be accessed by hackers?
- If so, will they be able to identify each individual associated with each template, and
- will they be able to reconstruct the original data?
It has been proven by many researchers that an individual’s raw biometric data (i.e. the original scan of their face, fingerprint, etc.) can be reconstructed from the template. Criminals can use the reconstructed templates to gain access to restricted areas and data. This makes biometric technology risky because a password can always be replaced but making a new biometric template is only limited to the number of fingers, eyes, hands, etc., a person has.
Distance, Lighting and Motion Blur are Problems Video-Based Face Recognition
Of all the various biometric devices available, the fingerprint scanner remains number one. This is largely because of its accuracy and ease of use. Recently, there has been a substantial increase in the use of video-based face recognition. To combat crime and terrorists attacks many law enforcement agencies are coupling their use of fingerprinting scanners with face recognition software. While face recognition based on still images has a high rate of accuracy, there are still tremendous problems with video-based face recognition. Rama Chellappa, Chairman of the Electrical and Computer Engineering Department at the University of Maryland-College Park, says that identifying a person based on video at distances greater than 50 feet can be difficult. He continues to explain that motion blur, lighting, and unconstrained subjects can cause major problems with identifying people through video. Below Dr. Chellappa discusses more reasons why this type of technology is not ready for extensive use:
Overall, the biggest issue that is associated with biometric technology is that of privacy. Although many people are open to the idea of having technology secure them, there are many who are wary of how the information collected can be used for other instances. Formally defined as the gradual widening of the use of technology or a system beyond the purpose for which it was originally intended, function creep is the broader underlying issue when it comes to biometric technology. Those opposed to the use of biometric devices will cite the Social Security Number (SSN) as the biggest example of function creep. At first the SSN was used for tax purposes, but it has transformed to a basis for identification for everyone within the United States. Religion, people's cultural and personal beliefs are other concerns that are prominent within this topic.
As mentioned before, biometrics can decrease fraud, lower costs, and provide better accuracy than humans, but there are still many dangers that are associated with it. Though the issues with video-based face recognition, biometric templates being duplicated, and privacy issues have yet to be resolved, there are major biometric programs that are currently in place and many others that will be implemented in the upcoming months. The FBI's billion dollar Next Generation Identification program will be the biggest. This program will depend heavily on video cameras to track and capture criminals, but will be able to also track non-wanted individuals.
Biometrics by John Woodward, Epic.org, Information Week and FBI