space space space space space space space space space space space
logo
contact
   
HOME      PRODUCTS & SERVICES      TECHNOLOGY DIVISIONS      ABOUT QS      WORKING WITH QS      prototype

Press Release


Nov 1, 2006
Quantum Signal wins Dept of Defense SBIR Phase II contract to develop advanced robotics technology >>


Archived News >>



Upcoming Events

None

Past Events >>


Biometrics
Capabilities

Advanced Biometrics for Identity Verification

As security threats grow larger, particularly in the information technology arena, there is increasing need for fast, robust personnel authentication that is less vulnerable to attack than traditional technologies.

Biometrics, the science of verifying identity based upon a physical or behavioral characteristic, has been identified by technology and security professionals and governments around the world as a technology who’s time has come.

Biometrics technology based on face, speech, fingerprint, and other human traits are being deployed for a wide variety of situations and applications – from enhancing airport security to making cutting-edge video games.

From this project, Quantum has developed key technologies and software in face and speech, all of which are available for licensing for inclusion into your product or application.


Advanced Speaker Verification

“My Voice is My Password"
This is the concept behind speaker verification. Rather than using the picture of someone’s face or the scan of a fingerprint, speaker verification technology allows one to simply speak a word, phrase, or number and have the computer verify identity by comparing it against previously recorded examples using a sophisticated algorithm.

SPEAKER VERIFICATION TECHNOLOGY

 

While speaker verification technology has been around for many years, the biggest technical challenge has been to build a system that recognizes your voice regardless of the device it is spoken into or transmitted over: microphone on your desktop, telephone on a land-line, or cellular telephone. Many speaker verification products on the market have substantial troubles with the cross-channel problem, and this has disappointed customers and hindered deployments in the past.

Quantum Signal has developed a breakthrough speaker verification technology that overcomes the cross-channel problem, allowing the system to recognize the users voice no matter what recording device is used! Some of the highlights of this technology include:

• Very high security – less than 0.5% equal error rate

• Short, fast enrollment for users – under 3 minutes

• Challenge and decision in seconds

• Small computational footprint

• Tunable to a wide variety of applications and uses

• Extensively tested

• Useful for securing facilities, computers, automating password resets

• Growing technology developed by world-class experts

Quantum Signal is proud to offer it’s speaker verification technology for inclusion into your product. We offer a complete software development kit (SDK) along with consulting and software engineering services to help you get the best performance.

For more information, please contact us (biometrics@quantumsignal.com).


Face Processing Technologies
When most people think biometrics, the idea of using faces for verification and recognition springs immediately to mind. Face based recognition is a staple of human interaction, and the concept of computers accomplishing the same task has captured the public imagination for many years. Since 9/11, there has been tremendous interest, growth, and advancement in this area.

Quantum Signal has been developing technologies and software that better enable computers to detect, track, and identify faces. Some of the technologies include:

Face Detection
Processes still image or video data and detects/provides location of human faces in that data. Can include photographic data, stylized pictures, and more. Robust to lighting, angles, obfuscations (e.g. odd haircuts, etc), and more.


Face Tracking
Performs real-time tracking of faces in live or pre-recorded video feeds. Highly robust to capture angle, lighting, scale, and other factors. Tracks multiple faces simultaneously in video feed.



Face Extraction

Extracts face image from still images or video and removes background information, leaving only the face for further processing.



Feature Analysis & Pose Estimation
Real time tracking of gross facial features, as well as off-line detection and localization of features. Ability to predict pose (e.g. where person is facing) based on facial contours.



Identity Comparison
Ability to compare face images and generate statistical measurement of a “match.”

Like the speaker verification, Quantum Signal is proud there technologies for inclusion into your product. At the current time, a software development kit (SDK) is available for face detection, tracking, and extraction, with additional features being added in the second half of 2005. Quantum also offers consulting and software engineering services in this area, and can assist you in evaluating and testing face-related technologies and implementing a highly functional solution.

For more information, please contact us (biometrics@quantumsignal.com).