DOC PREVIEW
SJSU CS 265 - FACIAL RECOGNITION TECHNOLOGY

This preview shows page 1-2 out of 5 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

II. HOW FACIAL RECOGNITION SYSTEMS WORKa. Basic steps of a facial identification processIII. RELIABILITY OF FACIAL RECOGNITION SYSTEMS AND THEIR LIMITATIONSIV. USES OF FACIAL RECOGNITION SYTEMSV. PRIVACY ISSUESVI. CONCLUTIONFACIAL RECOGNITION TECHNOLOGYBy Joanne voI. HISTORY OF FACIAL RECOGNITION TECHNOLOGYAmong one of the 4 best established biometric techniques that fall under the umbrella of biometrics, which identify people by measuring some aspect of individual anatomy or physiology,or some deeply ingrained skill or behavioral characteristic, facial recognition is the fastest growing biometric technology and will probably be accepted more widely in the future because it’s simpler for set up and use and also getting less expensive.As a brief history about this technology, according to the article “Facial-recognition tech has people pegged” by Emelie Rutherford, facial recognition technology has come to existence over adecade ago when U.S Defense Department was trying to find a technology that can spot criminalsat border crossing; since then, “university scientist have been working on facial recognition with the financial support from the U.S Defense Department.” Eventually, companies began commercializing the technology in the mid of 1990s. This technology made headlines in February 2001 immediately after it was first used in public at Super Bowl XXXV in Tampa by theauthorities to search for felons and terrorists among the crowd of 100,000 spectators. Soon after that, the systems are also installed and used in Tampa’s Ybor City, a coastal village, and some other places such as in the city of Virginia Beach to look for criminal suspects and missing children…etcAlthough this technology is the fastest growing biometric technology today, it is also the most controversial of all biometrics. While many people support the use of this new technology, especially after September 11, many others start concerning about the “possibility of identity theftand privacy infringement” from it’s widespread use. No matter which side we’re on, we should still learn to know about this technology and how the facial recognition systems work.II. HOW FACIAL RECOGNITION SYSTEMS WORKA facial recognition system consist of cameras that capture images of people who pose or simply walk by, and a software that matches those pictures. Following are the details of how these systems work, in general.a. Basic steps of a facial identification processAlthough facial recognition methods may vary, but they generally involve a series of steps that serve to capture, analyze and compare your face to a database of stored images. These basic stepsare:1) Detection: When the system is attached to a video surveillance system, the recognition software keeps searching the field of view of a video camera for faces. For each head-likeshape it detects, the system processes it.2) Alignment: Once a face is detected, the system determines the head's position, size and pose. A face needs to be turned at least 35 degrees toward the camera for the system to register it.3) Normalization: The image of the head is scaled and rotated so that it can be registered and mapped into an appropriate size and pose. Normalization is performed regardless of the head's location and distance from the camera. Light does not impact the normalizationprocess.4) Representation: The system translates the facial data into a unique code. This coding process allows for easier comparison of the newly acquired facial data to stored facial data. 5) Matching: The newly acquired facial data is compared to the stored data and (ideally) linked to at least one stored facial representation. b. What facial data the system takes?Assuming that the system has detected a head and has done with the alignment and normalization processes, what facial data does it take into consideration before it translates that data into a unique code? Different systems may analyze a face that they have captured differently. Let’s look at Visonics’s way because it’s one of the leading companies specialized in developing facial recognition systems. Visionics’ facial recognition system, FaceIt, measures a face according to itspeaks and valleys, which are called nodal points. The tip of the nose, the depth of the eye sockets, distance between eyes, width of nose, cheekbones, jaw line, and chin are some examples of nodal points, and “there are about 80 nodal points on a human face”. The software concentrates on the inner region of the face called “golden triangle”, which runs from temple to temple and just over the lip; this is the most stable area because even if you grow beard, put on classes, put on weight or age, that region tends not to be affected. FaceIt only needs from 14 to 22nodal points to complete a recognition process, according to Kevin Bonsor’s article “How FacialRecognition Systems work.” This software takes the measurements and creates a numerical code called a faceprint, which is made up of a string of numbers and it represents a face (as step #4 above). c. How is the matching process done?The faceprints stored in the database represents the people (or criminal) we’re looking for and will be compared against the ones of the people passing in front of the cameras to find a match. “The FaceIt system can match multiple faceprints at a rate of 60 million per minute from memoryor 15 million per minute from hard disk.” As comparisons are made, the system assigns a value to the comparison using a scale of one to 10. If a score is above predetermined threshold, a matchis declared. The operator then views the two photos that have been declared a match to be certainthat the computer is accurate.III. RELIABILITY OF FACIAL RECOGNITION SYSTEMS AND THEIR LIMITATIONSFacial recognition software is proven not yet as accurate as the other types of biometrics such as fingerprints or iris scanning. Although the marketing people of those companies selling facial recognition systems claim that their systems can recognize a person even if he/she tries to disguise himself / herself, the real study showed that “a disguise or even a minor change in appearance, like wearing sunglasses or wearing or growing a mustache can often fool the system. Even an unusual facial expression can confuse the software.” Lighting and the angle at which the photos are taken are some other factors that can affect the image quality and system


View Full Document

SJSU CS 265 - FACIAL RECOGNITION TECHNOLOGY

Documents in this Course
Stem

Stem

9 pages

WinZip

WinZip

6 pages

Rsync

Rsync

7 pages

Hunter

Hunter

11 pages

SSH

SSH

16 pages

RSA

RSA

7 pages

Akenti

Akenti

17 pages

Blunders

Blunders

51 pages

Captcha

Captcha

6 pages

Radius

Radius

8 pages

Firewall

Firewall

10 pages

SAP

SAP

6 pages

SECURITY

SECURITY

19 pages

Rsync

Rsync

18 pages

MDSD

MDSD

9 pages

honeypots

honeypots

15 pages

VPN

VPN

6 pages

Wang

Wang

18 pages

TKIP

TKIP

6 pages

ESP

ESP

6 pages

Dai

Dai

5 pages

Load more
Download FACIAL RECOGNITION TECHNOLOGY
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view FACIAL RECOGNITION TECHNOLOGY and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view FACIAL RECOGNITION TECHNOLOGY 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?