NYC Surveillance Camera Project*

Video Tracking Technologies and Identity Systems

A Mediaeater Report Nov. 2001



Ver. 0.7 Last Rev: Wednesday, November 21, 2001 - 6:21:36 PM ================================================== Index Definition Industry Players News Items Related Research Benchmarks and Tests Papers Laws Usage by Sector 1999 Biometrics Revenue by technology Glossary Problems With.. Recommendations ============== Facial Recognition Face recognition analyzes facial characteristics / facial image of a person for authentication. See: DOD - Facial Recognition 101 (pdf) ==============

Industry Players

(alpha sort) > Company -- BioID AG -- Raleigh, NC -- -- BioID AG develops, produces and markets biometric personal identification systems > Company -- Biometrica Systems, Inc -- Mont Vernon, NH -- -- The Visual Casino Product suite is a set of products and services that provide casinos with identification tools for use in surveillance and security. Facial Recognition, Database, Image, and Network technologies. Clients include Foxwoods, Trump, Stratosphere.. > Company -- eTrue -- Southborough, MA -- Managed service provider of biometric authentication for trusted services. TrueFace Engine Core software module for locating, verifying and/or identifyng people's faces. This face recognition software engine can be customized to any face recognition application from secure financial transactions to database searches to surveillance identification. All the interface hooks are in place so that integration is a snap. TrueFace ID Complete solution for identifying a person's face in a database of people's faces from either a surveillance video or image files. > Company -- FaceKey Corp -- San Antonio, TX -- -- > Company -- Image Metrics -- Stockport, Cheshire, UK -- -- Image Metrics have a model-based computer vision platform called Optasia™, which automates the extraction of information from images and video. a "Image Understanding platform" > Company -- Imagis Technologies Inc. (OTCBB: IGSTF; CDNX: NAB; GERMANY: IGY) -- Vancouver, B.C. Canada -- -- Imagis Technologies is a Canadian developer of image identification software with a focus on biometric facial recognition. Imagis markets its technology as both modules, for integration into existing applications, and as complete products for law enforcement departments, security agencies, and casinos, who all have a clear need for facial recognition capabilities. Systems and installations can be found in numerous sites as diverse as international airports, casinos, government offices, correctional institutions, and police departments around the world. Imagis Technologies - Products - Airport Security - ID-2000 > Company -- Lau Technologies / Viisage Inc. -- Littleton, MA -- -- -- Viisage's face-recognition technology uses "eigenfaces," which map characteristics of a person's face into a multi-dimensional face space. The unknowing usage of their product at the Super Bowl did not play well in early 2001 > Company -- MetaSignal, Inc. -- Newton, MA -- -- MetaSignal, Inc. is developing a system that offers theme park guests a high quality digital video souvenir of their entire day at the park without having to carry a camcorder. Video clips of a specific individual or group taken throughout a park by well positioned cameras are automatically assembled to create a personalized DVD movie. Video Information System that manages MPEG-2 footage from multiple live camera sources. Video streams are indexed via precision metadata (tracking or other information) and can be automatically and simultaneously, queried, retrieved, and assembled by many clients. > Company -- Neurodynamics Limited -- Cambridge UK -- -- Tridentity is Neurodynamics biometric identification poduct focused on three-dimensional facial recognition. > Company -- Printrak -- Anaheim, California -- -- Imagetrak™ 5.0 Intelligent Image Management System > Company -- SafeTzone Technologies Corp. -- Laguna Hills, CA -- -- Clients: Wild Rivers water park in Irvine, Calif. SafeTzone is a guest tracking system that quickly locates their child on a map of the park. SafeTzone products are used at Waterparks/Theme Amusement Parks, Casinos/Hotels, Cruise Ships. SafeTzone System consists of SafeTzone Tags, (about the size of a wrist watch) which are worn by each member of the family or party, and the SafeTzone software and hardware that is installed in the environment. By making an inquiry at any LocationStationtm within the environment, any parent or child may locate the other. Using radio signals, SafeTzone can read Tags at long distances (no line of sight is required) and track their movement within a defined perimeter. The data generated by the hardware is tied to SafeTzone’s software infrastructure that transforms, distributes and displays it at the LocationStations using the environment’s own Ethernet backbone. > Company -- TSSI -- Mandrake Facial Recognition System -- Swindon, United Kingdom -- email: -- Mandrake Facial Recognition Mandrake Facial Recognition System provides automatic monitoring for known faces, accepting video input from Wavelength CCTV or other non-photographic sources. Images are compared with those on the database to check for a match. Customised extensions to the system enable integration with electronic photo-fit systems and automatic number plate recognition. TSSI, a Scipher company make automated systems for the authentication of people, objects and documents. Biometrics technologies, structured magnetics and identity systems for integration into new or existing security systems Quote (product information sheet.) "The Case for Face Recognition ...Covert - the subject does not need not know of its operation." > Company -- Visionsphere Technologies -- Ottawa, Ontario, Canada -- Company has developed proprietary face recognition software and hardware products. UnMask Plus is a software AI (artificial intelligence) system for identification and removal of duplicate or multiple images from large photo-ID databases. It's Me -- an advanced automatic computer logon authentication system. The system hardware and software components can not only be used to verify the identity of a network or workstation user at logon time, they can also be used as part of e-Business verification, video conferencing, etc. FaceCam -- Visionsphere's layered, biometric, user-verification terminal, designed and built specially for secure access applications, such as physical access control, time and attendance and registration systems. > Company -- Visionics Corporation (Nasdaq: VSNX) -- Minnetonka, MN -- Visionics Facial Recognition -- Company is the maker of identification technologies and systems. Products like FaceIt, live scan and IBIS which are biometric identification systems that use facial recognition technology. TENPRINTER and FingerPrinter CMS are live scan systems used by government agencies, law enforcement, airports, banks and other commercial institutions. IBIS is a remote identification system that uses a capture and transmission method that alows access into law enforcement and other legacy databases. ==============


ResearchIndex (NEC Research Institute ) On-line publications (page is maintained by Peter Kruizinga) Biometrics: Uses and Abuses Face Recognition and Tracking from Video. Eye Identification for Face Recognition with Neural Networks Face Recognition by Elastic Bunch Graph Matching Fabrizio Smeraldi's Papers Automatic Detection and Tracking of Faces and Facial Features in Video Sequences Public Video Surveillance: Is It An Effective Crime Prevention Tool? ==============

Related Research

The Face Recognition Home Page Video Surveillance and Monitoring Homepage Biometric Consortium Feature Detection Facial Animation Facial Analysis Privacy International Video Surveillance The Face Detection Homepage. By Robert W. Frischholz Ear Identification
Image Understanding ==============

Benchmarks and Tests

Facial Recognition Vendor Test (FRVT) 2000 Results FERET Database Demo FRVT_2000 size 6410942 bytes - 11/14/2001 Facial Recognition Vendor Test 2000 Final Report (.pdf) size 409552 bytes =============


Flordia v Thomas Kyllo v. United States (99-8508) Katz v. United States, 389 U.S. 347 (1967) (USSC+) ==============

Useage by Sector

City/State Virginia Beach VA Tampa FL Casino industry (matched against a database of scam artists) Airports (matched against a database of terrorists) Logan Airport (2) Iceland's Keflavik Airport Toronto's Pearson Airport San Francisco International Airport (tbc) Oakland International Airport (tbc) McGhee Tyson Airport Knoxville, Tenn.(tbc) ============== 1999 Biometrics Revenue by technology Fingers 34% Hand 26% Face Scanning 15% Voice 11% Signature 3% Retina 2% Iris 9% Source: Internataional Biometrics Group ==============


A Statistical Learning/Pattern Recognition Glossary EER Equal Error Rate FAR False Acceptance (or Alarm) Rate FERET A comprehensive U.S. government face recognition test, used for personnel security. Authentication: Any process that validates who you are Biometric: A unique, machine measurable biological trait or characteristic that can be used to verify identity. Common biometrics are fingerprint, hand geometry, retina or iris patterns and face. Enrollment: The process of generating a biometric template for a user. It is used during the comparison cycle in biometric technology for authentication. False Acceptance: When a system accepts someone that is not legitimate and allows access. Sometimes called a false positive and also known as a Type II error. False Rejection: When a system fails to recognize or verify a person and therefore rejects a legitimate user. Also called a Type I error. Minutiae: Unique characteristics of a fingerprint linked via algorithms. This linking forms a unique pattern called a minutiae template. Appearance (supported by still images) e.g. descriptions used in passports, such as height, weight, colour of skin, hair and eyes, visible physical markings, gender, race, facial hair, wearing of glasses natural physiography e.g. skull measurements, teeth and skeletal injuries, thumbprint, fingerprint sets, handprints, retinal scans, earlobe capillary patterns, hand geometry, DNA-patterns bio-dynamics e.g. the manner in which one's signature is written, statistically-analysed voice characteristics, keystroke dynamics, particularly login-id and password social behaviour (supported by video-film) e.g. habituated body-signals, general voice characteristics, style of speech, visible handicaps imposed physical characteristics e.g. dog-tags, collars, bracelets and anklets, bar-codes and other kinds of brands, embedded micro-chips and transponders ============== Problems With..
  • Lighting
  • Age
  • Glasses
  • Hair
  • Hats
============== Recommendations (in part)
  • Safeguards be adopted to ensure that biometric data is not misused to compromise any information, or released without personal consent or the authority of law; e.g. Prohibiting collected face scan data from being sold to or shared with third parties without a warrant, civil subpoena or the informed consent of the person;
  • Private sector implement policies that clearly set forth how biometric data will be collected, stored, accessed, and used, and that preserve the rights of individuals to limit the distribution of the data beyond stated purposes;
  • The government sector adopt clear legal standards that carefully define and limit the conditions under which agencies of national security and law enforcement may acquire, access, store, and use biometric data; and seeing that government agencies need to get a warrant before using face scan systems
  • In both the public and private sectors, managerial and technical controls be put in place to protect the confidentiality and integrity of databases containing biometric data.
  • Requiring data that doesn’t match that of a criminal or a suspected criminal to be destroyed. (culled from SB 169 and the International Biometric Industry Association)