Biometrics Puts Technology and Identity Face to Face
Posted: November 27, 2006 at 1:00 am, Last Updated: November 30, -0001 at 12:00 am
A biometrics chart.
Photo courtesy of Harry Wechsler
One of the most well-known biometric identifiers is also the most basic — the human face. Every day we identify the people we know by their face. But what happens as people age? What if they grow their hair out or cut it short? Or what if they grow a full beard when once they were clean-shaven?
These changes occur all the time, creating challenges for biometric systems — especially when security is involved.
Computer science professor Harry Wechsler, an internationally known expert on the topic of facial recognition, is working to make the technology behind facial recognition more accurate.
The director of Mason’s Distributed and Intelligent Computation Center, Wechsler was involved in the development of the Facial Recognition Technology (FERET) database, which has become the standard facial database for benchmark studies, experimentation and performance evaluation. The FERET database made it possible for researchers to conduct analysis and develop algorithms using a common database.
One of the major challenges for facial recognition software is determining how to account for superficial changes, including changes that take place during the day.
“Two pictures of the same person can be more different from pictures of two different people,” says Wechsler. “I would look very different tomorrow morning after a transatlantic flight, unshaven and exhausted, than I would today when I board.”
Challenges to Recognition
Physical changes to someone’s face are not the only complicating factors. The distance at which an image is captured, along with lighting, pose and distortion, must also be considered when comparing facial images, says Wechsler.
Current security systems that use face biometrics are particularly ineffective when temporal changes, involuntary or not, occur. In addition, the working hypothesis for face recognition research so far has not been particularly concerned with the possibility that subjects would seek to foil their true biometric signatures.
Because the very purpose of biometrics is to provide security from impostors and those seeking to breach security, such subjects are obviously well motivated to interfere with the proper acquisition of their biometric signatures. Therefore, they will attempt to hide and/or alter the information needed for their identification.
Because occlusion and disguise usually affect only parts of the face, Wechsler has proposed a novel recognition-by-parts approach for reliable face recognition. This approach uses individual face components and their sequential recognition. With this approach, the face representation, or picture, spans a multi-resolution grid that captures partial information from the face at different scales to accommodate different surveillance scenarios, including human identification from a distance.
The face parts represent the eyes, nose, mouth, eye and nose, nose and mouth and the like. Using computer analysis, the parts that appear to be most relevant in facial appearance are captured and compared electronically, allowing the program to distinguish one face from another through the use of comparison and elimination.
Wechsler has found that the recognition-by-parts approach is reliable enough to cope with geometry, illumination and temporal changes, as well as missing or disguised parts of the face. He believes that reliable recognition of occluded and disguised faces will constitute a major step forward and make face recognition the medium of choice for biometric identification. In addition, surveillance in crowded environments using closed-circuit TV, when only parts of faces are visible from time to time, will become feasible and greatly benefit homeland security efforts.
Because face recognition technology requires little or no cooperation from the subject, it is becoming one of the top choices for biometrics and is starting to move into the commercial market. The government is already using the technology.
Yet experts say facial recognition will never be 100 percent accurate. False-positives (misidentifying someone) and false-negatives (not identifying someone) do still occur, but with Wechsler’s groundbreaking research, accurate results are becoming more prevalent.
In his recent book, “Reliable Face Recognition Methods: System Design, Implementation and Evaluation,” published by Springer, Wechsler examines the evolution of face recognition research and explores promising new avenues for research and development.
Over the years, Wechsler’s research on face recognition has been funded by the Navy Research Laboratory, Technical Support Working Group/Department of Defense and the Army Research Laboratory. He holds two patents and has two pending for his biometric-related work.