Another Pair of IT&E Researchers Earn Patent
Posted: December 20, 2004 at 1:00 am, Last Updated: November 30, -0001 at 12:00 am
By Rey Banks
Researchers in the School of Information Technology and Engineering (IT&E) continue to gain recognition for their inventions. Following on the heels of the 16th patent for the university is another one, this time awarded to Harry Wechsler, professor of computer science and director of the Center for Distributed and Intelligent Computation, and Chengjun Liu, PhD ’99, who is now an assistant professor of computer science at New Jersey Institute of Technology. The pair received the patent for work in feature-based classification, an application more commonly known by its use in the technology of facial recognition.
Facial recognition falls into a larger group of technologies known as biometrics, which use personal information to verify identity. The basic idea behind biometrics is that our bodies display unique properties that can be used to distinguish us from others. In addition to facial recognition, other authentication methods that use biometrics include fingerprint, retina scan, and voice identification.
“Facial recognition methods vary, but they generally involve a series of steps that capture and measure points of your face and compare those results to a database of stored images,” says Wechsler. “First, a picture is taken by a simple camera. That image is then digitized and measurements are taken.”
Face recognition is largely motivated by the need for surveillance, security, and telecommunication. But other uses can include human-computer interaction and smart environments. “Imagine a door that doesn’t require keys or numbers to open, but instead relies upon recognizing the face or faces of the occupants,” says Wechsler.
Feature-based classification isn’t yet perfect. Facial features change and accuracy can be affected by illumination, aging, and people’s attempts to fool the system through disguise. However, Wechsler and Liu’s invention is an important step in the process of perfecting the ability of computers to recognize facial images.