Researchers from a multinational team claim to have created a biometric recognition system that uses three-dimensional images of faces and ears with an accuracy of 99.25% with an error rate threshold of 0.75%.
The two biometrics are merged to increase the accuracy of authentication, according to a paper by a team of scientists from India, South Korea, Saudi Arabia and Bangladesh. Their results indicate that the method is competitive with the state of the art.
Three-dimensional facial images have their challenges, not the least of which is file size, but they are better than two-dimensional images in that they are capable of strong facial affirmation and perform better in low light conditions. And there are more biometrics to authenticate with.
Three-dimensional images of the ear are stable over many years, but have problems with low light levels and pose variations, and large file sizes can incur high computational costs.
Principal component analysis was used for three-dimensional face recognition and independent component analysis was used for ear recognition.
For facial recognition, the team used the Face Recognition Grand Challenge database. Each of the 30 chosen subjects expressing anger, happiness, disgust, fear, surprise and sadness.
The F and G collections at the University of Notre Dame provided 30 pairs of ear images captured from different angles.
The researchers said they would tackle the biometric hardware next. They are considering ways to reduce the cost of a handheld 3D scanner into a portable device or laptop.
3D | precision | authentication | biometrics | biometric search | ear recognition | facial recognition