Skip to local navigation | Skip to main content

Facial Recognition Technology

Face recognition technologies can identify suspects and improve security and officer safety. These systems use multidimensional imaging techniques to create facial images. Recent breakthroughs in high-resolution imagery show the most potential for improving the accuracy of two-dimensional and three-dimensional facial recognition systems.

NIJ has funded the development and evaluation of still- and video-image approaches for facial recognition that will allow law enforcement and corrections officers and court officials to correctly identify cooperative or noncooperative individuals and to secure the identity of those individuals in the justice system.

Advancing the State of the Art in Facial Recognition Research

Facial recognition systems funded by NIJ continue to advance the technologies involved in suspect identification. In 2006, NIJ funded research that is:

  • Producing a prototype of facial recognition and audio analysis software. (Sonic Foundry, Advanced Media Analysis Project," grant number 2003–IJ–CX–K026.)
  • Developing a system that uses an advanced algorithm to solve complex facial pattern recognition problems. (New Jersey Institute of Technology, “Facial and Iris-Based Biometric System,” grant number 2006–IJ–CX–K033.)
  • Developing software that extracts facial images from surveillance video, rotates the images in three dimensions, and combines them to produce a recognizable image. (General Electric Company, “Active 3-D Face Capture,” grant number 2006–IJ–CX–K045.)

Evaluating the State of the Art

In partnership with the National Institute of Standards and Technology, NIJ sponsored two “technology potlucks” in 2006. Thirty domestic and international organizations came together to evaluate and test state-of-the-art face- and iris-recognition technologies at the “Face Recognition Vendor Test” and “Iris Challenge Evaluation.” The vendors brought their refined algorithms or software programs to help scientists determine which algorithms were most effective at matching faces or irises.

Facial Recognition in the Field

The Pinellas County Sheriff's Office in Florida has been using digital cameras to take mugshots in the field. Over the past two years, the images have helped to correctly identify 295 people who lied about their identity. Currently, 50 Pinellas County Sheriff's Office patrol cars are equipped with mobile facial recognition technology.