Biometrics: Modern History
Now that we’ve explored the very early roots and discoveries of biometrics in , let’s dig into the most exciting part of the story. Starting thousands of years ago in concept, this amazing technology was slow to evolve – until the mid-20th century, that is. And that’s exactly where we’re going to begin, in the decade that changed everything from rock music to rocket ships – the 1960s.
- In 1960, Woodrow W. Bledsoe developed the first semi-automatic face recognition system. It may come as no surprise that this was via contract with the U.S. Government. His system required that the administrator locate features on a photograph (eyes, ears, nose, mouth). It then calculated ratios between these points that could be used to compare to available reference data.
- Also in 1960, a Swedish professor named Gunnar Fant created a model that explained the physiological components of acoustic speech based on the x-rays of people making phonic sounds. This model gave us a better understanding of the biological components of speech and became an integral part of speaker recognition.
- 1965 introduced the first automated signature recognition system based on research by North American Aviation.
- In 1969, the FBI began searching for a system that could automate its fingerprint identification process. The existing manual process required an inefficient number of man-hours. The National Institute of Standards and Technology (NIST) was contracted for this quest. They earmarked two crucial challenges to its solution:
- Ability to scan fingerprint cards and identify minutiae
- Then to compare a scanned fingerprint to lists of minutiae
- Starting off the disco era in 1970, three researchers, A.J. Goldstein, A.B. Lesk, and L.D. Harmon, drastically improved automated facial recognition. Locations and measurements still needed to be computed manually but they now included analysis of 21 specific markers, including lip thickness and hair color.
- Also in 1970, Dr. Joseph Perkell expanded upon the acoustic speech production model with the inclusion of the tongue and jaw in motion x-rays.
- In 1974, the University of Georgia deployed one of the first commercial systems for hand geometry recognition. There were three primary purposes for this technology: time and attendance, personal identification, and physical access control.
- In 1975, the FBI gave the NIST additional funds for the production of minutia scanners and extracting technology – leading to the development of a prototype reader. At that time, only a print’s minutiae was captured because digital storage came at a high cost. The NIST wanted to improve that, so over the next few decades they led developments in automatic digitizing of inked fingerprints. They also focused on image compression, quality, classification, data extraction, and matching. This all culminated in the development of M40, the FBI’s first operational matching algorithm.
- In 1976, Texas Instruments released the first prototype speaker recognition system. It was rigorously tested by the U.S. Air Force and the MITRE Corporation.
- In 1977 a patent was awarded to Veripen, Inc. for a Personal Identification Apparatus that was used to acquire dynamic signature information. The device could measure pressure and eventually led to the testing of automatic handwriting verification.
- Starting off the decade in 1980, the NIST Speech Group was developed to research speech processing techniques. The Group still operates to this day.
- In 1984, the U.S. Army started testing hand geometry for banking and financial transactions.
- In 1985, a patent for hand geometry identification was secured by David Sidlauskas.
- Another 1985 patent was awarded, this time to Joseph Rice for subcutaneous vascular pattern recognition. This technology uses the blood vessels just below the skin’s surface to recognize individuals.
- Also in 1985, Ophthalmologists Aran Safir and Leonard Flom proposed that no two irises are alike.
- In 1986, the National Institutes of Standards and Technology (NIST), together with the American National Standards Institute (ANSI), published a standard for the exchange of fingerprint minutiae data (ANSI/ NBS-I CST 1-1986). This was the first version of the standards used by law enforcement today.
- In 1986, Drs. Leonard Flom and Aran Safir were awarded a patent for using the iris for identification. Dr. Flom approached Dr. John Daugman to develop an algorithm to automate human iris identification.
- In 1988, the Lakewood division of the Los Angeles County’s Sheriff Department began using video images of suspects to search their digital mugshot database. Two researchers, Sirovich and Kirby, applied principle component analysis to facial recognition and proved that less than 100 values were needed to approximate an aligned, normalized facial image. This is known as the Eigenface technique which uses standard linear algebra.
- In 1991, while using the Eigenface technique, Turk and Pentland discovered that residual error could be used for facial detection in images, which led to the idea that real-time recognition could be possible despite the constraints of environmental influences.
- The NSA’s Biometric Consortium was formed in 1992 in order to expand biometrics efforts in testing, standards, and interoperability. While the Consortium remains active today as a liaison between government, industry, and academics, many of the original working groups have split into or been integrated with other organizations such as INCITS, ISO, and the NSTC Subcommittee on Biometrics.
- In 1993, the Defense Advanced Research Products Agency (DARPA) and the Department of Defense Counterdrug Technology Development Program Office sponsored FERET (Face Recognition Technology Evaluation), which ran until 1997. FERET assessed face recognition system prototypes and ushered in commercial market use.
- In 1994, a patent for iris recognition was awarded to Dr. John Daugman. Owned by Iridian Technologies, it is the basis of most commercial iris recognition products in use today.
- 1994 gave birth to the Integrated Automated Fingerprint Identification System (IAFIS) competition. Entrants were presented with three key challenges: acquisition of digital fingerprints, extraction of local ridge characteristics, and characteristic pattern matching. Lockheed Martin won the contract and was selected to build IAFIS for the FBI.
- In 1994, the INSPASS biometric implementation system was launched. It allowed travelers to skip immigration lines at certain U.S. airports using a card that encoded with their hand geometry data. Instead of being cleared by an Immigration Inspector, they simply presented this card and had their hands scanned by a biometric reading device. This program was discontinued in 2004.
- Also in 1994, RECOWARE Ltd., Hungarian company, created the RECOderm™ Palm System – the first automated fingerprint system that also supported palm prints.
- In 1995, RECOWARE Ltd. Was invited to the International Association for Identification (I AI) conference in Costa Mesa, California. RECOWARE was purchased by Lockheed Martin Information Systems just two years later.
- Also in 1995, the Iriscan prototype became commercially available as a joint project with the Defense Nuclear Agency.
- In 1996, the Atlanta Olympic Games used a hand geometry security system for access to the Olympic Village. The system processed more than one million transactions for more than 65,000 people in just four weeks.
- Also in 1996, for the National Institute of Standards and Technology received funding from the NSA to host annual evaluations for recognized industry speakers.
- In 1997, the NSA sponsored the Human Authentication API (HA-API) – the first standard for commercial, generic biometric interoperability, vendor independence, and interchangeability.
- CODIS was launched 1998, the Combined DNA Index System. This forensic database allowed the FBI to digitally store, search, and retrieve DNA markers for law enforcement. Lab sequencing could take anywhere from 40 minutes to several hours.
- In 1999, the International Civil Aviation Organization launched a study on the compatibility of biometric technology with Machine Readable Travel Documents (MRTD), and if a standard could be implemented internationally.
- Also in 1999, the FBI’s open-set fingerprint identification system was introduced. Before this point, a fingerprint captured on one system could not be matched against one gathered via another system.
We’ve seen biometrics grow by incredible leaps and bounds in the 21st century so far – and we’ve only just begun. Thanks to improved technology (which is now smaller, faster and more affordable), biometrics systems work more quickly and efficiently. Socially, the use of facial recognition started to become more acceptable and mobile biometric solutions have become common. From smartphone security to airport screening, it is now becoming a facet of everyday life. Here’s how we got there.
- In 2000, the first Face Recognition Vendor Test (FRVT) was held – it evaluated multiple commercially available biometric systems and large-scale databases. It has been since used to perform evaluations of fingerprints (2003) and iris recognition (2006) systems.
- Also in 2000, a Biometric Systems bachelor’s program (unaccredited) was developed by West Virginia University in collaboration with the FBI. It was the first degree program of its kind.
- In 2001, the Tampa Super Bowl used a facial recognition system to prohibit known criminals from entering the stadium. Unfortunately, there were about a dozen false positives (innocent sports fans) and zero true positives (no recognized criminals). Because of the media coverage this caused, it spurned Congressional inquiries into biometrics and its associated privacy concerns – and these thoughts became ingrained the consciousness of the general public.
- In 2002, the International Organization for Standardization (ISO) established the ISO/IEC JTC1 Subcommittee 37 (JTC1 /SC37) for the standardization of biometric solutions, data exchange, and interoperability across various systems.
- At the same time and along those same lines in 2002, the M1 Technical Committee on Biometrics was established by the U.S. Technical Advisory Group to develop standards across accredited ANSI member organizations.
- In April 2002, a staff paper on palm print technology was submitted to the Identification Services (IS) Subcommittee, Criminal Justice Information Services Division (CJIS), and the Advisory Policy Board (APB). The Joint Working Group called “for strong endorsement of the planning, costing, and development of an integrated latent print capability for palms at the CJIS Division of the FBI.” As a result, the FBI launched the Next Generation IAFIS initiative. It was tasked with developing and deploying an integrated National Palm Print Service and system.
- In 2003, the U.S. Government’s National Science & Technology Council initiated a Subcommittee on Biometrics tasked with coordinating research, development, policy, outreach, and international collaboration.
- On May 28, 2003, the International Civil Aviation Organization (ICAO) adopted an international blueprint for the integration of facial recognition biometric data into passports and other Machine Readable Travel Documents (MRTD).
- The European Biometrics Forum was also formed in 2003 – an independent organization supported by the European Commission that aims to establish the European Union as the World Leader in Biometrics Excellence. It focused on addressing adoption barriers and market fragmentation.
- In 2004, the United States Visitor and Immigrant Status Indication Technology (US-VISIT) security program was formed to confirm the identity of visa travelers, integrating digital photos and inkless fingerprints. For visa-waiver travelers, biometrics are captured at the port of entry. US-VISIT verifies whether the traveler has previously been determined inadmissible, is a known security risk (outstanding warrants), or has overstayed the terms of a former visa.
- The Department of Defense implemented the Automated Biometric Identification System (ABIS) in 2004. It identified and tracked people considered threats to national security by collecting iris images, ten rolled fingerprints, voice samples, DNA (oral swab), and up to five mugshots from persons of interest.
- In 2004, President Bush announced the Homeland Security Presidential Directive 12, which required all federal government employees and contractors to have mandatory personal identification cards containing stored minutia templates including two fingerprints.
- Also in 2004, California, Rhode Island, and Connecticut established palm print databases that allow law enforcement to submit unidentified prints and compare them to other databases of offenders.
- And finally in 2004, the U.S. Government sponsored the Face Recognition Grand Challenge (FRGC) develop algorithms to improve specific identified areas of interest in face recognition. Participating researchers were tasked with developing algorithms to improve specific areas of facial recognition.
- In 2005, the Sarnoff Corporation (now SRI International) demonstrated “Iris on the Move” (ability to collect iris images from individuals walking through a portal) at a Biometrics Consortium conference. It is the culmination of research and prototypes sponsored by the Intelligence Technology Innovation Center (ITIC) and the Defense Advanced Research Projects Agency (DARPA).
- In the summer of 2008, Google Maps enabled voice search on the BlackBerry and Nokia mobile phones. iPhone didn’t launch this ability until November 2008.
- Also in 2008, the U.S. Government’s Department of Defense and FBI began to develop more advanced databases for not just fingerprints but also palm, face and iris data. The Department of Homeland Security denied an individual entry into the U.S. after it was able to cross match biometric data with that of a suspected terrorist.
- In March 2010, a beta version of Google Voice Search was integrated into YouTube, providing automatic caption generation for hearing-impaired, English-speaking users.
- In a pivotal 2010 moment, U.S. National Security used biometrics to identify a terrorist (positive match with a GITMO detainee) who had been present at the 9/11 planning location.
- In 2011, that iPhone voice we know so well, Siri, was introduced, providing users with a voice-commanded digital assistant.
- Also in 2011, the CIA used DNA and facial recognition to confirm with 95% certainty that the remains of Osama Bin Laden were indeed his.
- In 2013, Apple released Touch ID, a fingerprint recognition feature, on their 5S, 6, and 6+ phones, the iPad Air 2, and the Mini 3. It not only allowed users to unlock their devices but also make purchases via Apple Pay and in the Apple iTunes Store, the App Store, and the iBookstore. Apple stated that fingerprint information is stored locally in a secure location on local device chip, not remotely on Apple servers or in iCloud, making it secure from external access.
- In 2015, Microsoft launched its Siri competitor, Cortana. It not only follows voice commands but also uses machine learning.
- 2018 saw a huge breakthrough in commercial applications when the MasterCard Biometric Card was released, combining chip with fingerprint technology to verify purchases.
- 2018 also introduced the Byton electric vehicle that integrated face biometrics to unlock its doors and load driver’s profile and settings (when they sit in front of the wheel). It also allowed drivers to control certain features using voice command and gestures.
Today, biometric technology is secure enough to support full-scale integration and the applications of it are about to charge ahead at full speed. Their measured traits are nearly impossible to counterfeit and hardware scanners are reporting over 99% effectiveness. As impressive as that percentage is, algorithms are getting even more accurate.
We are logging into our laptops, smart phones, even smart homes using biometrics. We are traveling more safely and efficiently using this technology. We’re essentially turning into a passwordless society (which is good for all of us who just can’t remember all our various passwords and their special characters). And the opportunities go even further, expanding beyond truly “bio” applications (like BagsID). That’s what we’ll be exploring in our next and final installment of this series, Part 3 – Biometrics: Future Vision.