The human brain has an amazing capacity for recognizing patterns, particularly faces. While we are able to differentiate different faces with apparent ease, computer facial recognition systems have a much harder time of it, relying on powerful computers and complex models to accurately identify the majority of differences between faces. This has held facial recognition systems back from being widely adopted, but now researchers at Florida Atlantic University (FAU) have developed a technique that significantly reduces the amount of computer power required without compromising accuracy.

Facial recognition systems have evolved from the systems that simply marked major facial features on a photograph and computed the distances from those features to a common reference point to approaches based on neural networks, dynamic link architectures (DLA), fisher linear discriminant model (FLD), hidden Markov models and Gabor wavelets. Next a way to create a ghost-like image that would succumb to an even more powerful analysis was developed that could accurately identify the majority of differences between faces.

If those techniques sound complicated it’s because they are, which means more powerful computers are required.

Now the team from FAU has cut the computer power needed by applying a one-dimensional filter to the two-dimensional data from conventional analyses, such as the Gabor method.

The team tested the performance of their new algorithm on a standard database of 400 images of 40 subjects. Images are grey scale and just 92 x 112 pixels in size. They found that their technique is not only faster and works with low resolution images, such as those produced by standard CCTV cameras, but also solves the variation problems caused by different light levels and shadows, viewing direction, pose, and facial expressions. It can even see through certain types of disguises such as facial hair and glasses.

Reducing the amount of computer power required could see facial recognition systems become a mainstream application, used for biometric authentication at border crossings, for access to buildings, for automated banking, crime investigation, and other applications.

The FAU team’s research, “A method towards face recognition”, was published in the International Journal of Intelligent Systems Technologies and Applications.