Sign In

Communications of the ACM

ACM TechNews

Recognition at Last

View as: Print Mobile App Share:

Florida Atlantic University researchers have significantly reduced the amount of computer power that is needed to operate face recognition systems without compromising their accuracy. Lin Huang and colleagues Hanqi Zhuang and Salvatore Morgera in the Department of Electrical Engineering have used a one-dimensional filter with the two-dimensional data from conventional analyses. Testing the algorithm on a standard database of 400 images of 40 subjects, gray scale, and 92 x 112 pixels in size, the team found the approach was faster and works with low resolution images, such as those produced by standard CCTV cameras.

The technique can also handle variation issues caused by different lighting levels and shadows, viewing direction, pose, and facial expressions. As a result, the algorithm would enable face recognition systems to see through disguises such as facial hair and glasses.

From EurekAlert
View Full Article


Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA


No entries found