Facial Recognition
Home Up

Facial Recognition Biometric Identification Technology and Techniques

Facial Recognition 

By defining a dozen or more points on a person's face, a video machine can measure the distance between the points and use them as a means of recognition. It may be easy to collect this information, but a person can change appearance to circumvent the system. There is a lower accuracy when images are captured outdoors.

This technology requires a person to sit in front of a digital camera while it tracks about 80 facial characteristics. The lighting must be perfect and the camera must line up the image perfectly.

Essentially, the technology measures the peaks and valleys of the face, such as the tip of the nose and the depth of the eye sockets, which are known as nodal points - the human face has 80 nodal points, only 14 to 22 are needed for recognition - concentrating on the inner region, which runs from temple to temple and just over the lip. It then comes up with a face print.

Face prints can also be stored on a smart card that users swipe through a door without looking into a camera.

The technology has been around since the early 1990s and is used in more than 100 casinos in the United States. It got a lot of attention last February when authorities used it at Super Bowl XXXV in Tampa to search for felons among the crowd of 100,000 spectators.

Improvements in 3-D facial recognition make it more accurate than 2-D and very promising for surveillance where iris or finger scanning isn't possible. Casinos, airports, and high-security facilities are likely candidates for facial-recognition technology.

The most significant attributes of 3-D facial-recognition technology are speed and accuracy. 3-D facial images are captured by a number of digital cameras positioned around the subject's face or by using a structured light grid that captures facial-structure data. This data is then stored in a back-end database, where it can be retrieved and compared with other facial images.

More work remains before facial recognition, much less 3-D facial recognition, becomes a mainstream technology for securing access to facilities or identifying people captured by video surveillance. "In video-surveillance environments, facial recognition requires very specific lighting and very specific facial poses," says David Fisch, an International Biometric Group consultant. "I'm not sure how ready 3-D is to replace 2-D facial recognition."

One of the keys to further developing 3-D facial-recognition biometrics is to create standards that govern how the technology is developed and deployed. A4Vision said Monday that the American National Standards Institute has adopted the company's proposal, co-sponsored by Motorola, Oracle, and Unisys, to amend the U.S. standard for facial-recognition format and data interchange to include 3-D facial-image data. A4Vision and its partners will now seek international recognition via the International Standards Organization in June.

The 3-D standard amends the current 2-D standard and provides vendors with instructions for storing high-resolution facial images. One component of the 3-D biometrics included in the standard is facial-feature-point data, which allows forensic specialists and law enforcement to query databases using descriptions of facial features, such as nose size, eye placement, and facial contours.

Facial Recognition Biometric Systems at France's Airports

03/ 05 - Countries worldwide have identified biometric technology as a critical tool for preventing terrorism and sabotage to their transportation infrastructures. The French Civil Aviation Authority has been working since the beginning of the year at Lyon's airport with French IT-service provider Euxia SA and Belgian biometric-systems integrator BioWise NV to use A4Vision's 3-D facial-imaging and -recognition systems to create security badges containing facial-recognition data for 500 pilots, mechanics, and other employees with access to the airport's highly secure tarmac. The airport hopes by June to issue as many as 5,000 badges to its employees.

Integrators working in Lyon are writing the specifications needed to roll out 3-D biometric systems throughout France at additional airports in Bordeaux, Lille, Nice, Paris, and Toulouse. The Lyon test demonstrated that 3-D biometric technology is reliable, with only a small number of incidents where the system enrolled a subject who shouldn't have been allowed to enroll or failed to enroll an eligible employee.

France's airports must also weigh the costs of deploying wide-scale 3-D facial-recognition systems, which can cost about $500,000 for a deployment of about 160 machines.

FacePrint Global Solutions, Inc

FacePrint Global Solutions, Inc. is dedicated to developing a slate of biometric solutions for secure identification, facial imagery products and distance-learning training for government and law enforcement agencies. FacePrint Global Solutions, Inc. offers biometric products and solutions for secure identification, facial imagery and e-learning.

FGS' SmartCard can be a passport, an ID card, a voters card, a drivers license, a resident alien card, or all in one.

FacePrint Global Solutions, Inc. is the developer of e-DNA(R) BioPrint Coding, the ground-breaking core technology supporting detailed biometric scans that can be reduced to extremely compact codes. e-DNA(R) BioPrint Coding enables rapid image capture and creation, search, transmission and re-creation without loss of image quality.

Omron recently announced they had found a way to enable facial recognition in PDAs and mobile phones with embedded cameras. 

The "OKAO Vision Face Recognition Sensor" only requires a .1 MP camera and a small software application to work. This means it will work with almost any embedded camera currently on the market. Perhaps the best part is the speed, the application can identify the owner and unlock the device in under a second, with 99% accuracy.

A4Vision Inc. says its proposal to amend the existing U.S. national standard for Face Recognition Format and Data Interchange to include 3D facial-image data has been supported and co-sponsored by such groups as ID Technology Partners, Logitech, Oracle, Unisys and new A4Vision partner Motorola.

The proposed standard opens new fields of applications for vendors to incorporate 3D face-recognition technology in such new products and applications as electronic cards that could be used as e-passports along with any application where an accurate facial image needs to be stored in a small record. In part, the proposed standard provides vendors with instructions for storing a high-resolution 3D facial image in fairly large files record without any compression loss.

Web www.smartcardscanada.com