The Innovation In Competitive Environments

Published Date: 02 Nov 2017

Biometric technologies (face-recognition cameras, digitalised fingerprints, voice prints, retina and iris scanners, etc.) are becoming unseparatable part of nowadays world. Such technologies can bolster online commerce, locate a missing child, and transmit medical information to doctors (Crews, 2002). With a biometric technologies it is very difficult, if not impossible, for any individual to disassociate oneself from one’s biometric – in a sense you are your biometric (Van der Ploeg, 2002). Evidently, biometric surveillance is a powerful technology for social control (Graham and Wood, 2003).

Fingerprinting is widely acceptable method of biometric surveillance. Such technique has been used for the identification of individuals since the late 19th century and this is the oldest biometric known. Compared to other biometrics, fingerprints are relatively inexpensive to capture. What is more, fingerprints remain relatively constant over time, with the exception of injury (Komarinski, 2005). In general, fingerprints are one of the most important tools for surveillance.

The purpose of this paper is to discuss the Automated Fingerprint Identification System, its evolution, development, strengths and weaknesses. Additional attention will be given to the analysis of the leading company in fingerprint biometrics and its current issues.

AFIS (Automated Fingerprint Identification System) is a type of biometric system that uses digital imaging to capture a fingerprint, which then can be compared to a database of fingerprint records to help determine the identity of an individual. Peter Komarinski (Komarinski, 2005), a criminal justice policy analyst, briefly explains the four components of AFIS name:

"The automation (A) process has eliminated the need for a print classifier to locate fingerprint cards from a file and compare two physical cards. The searchable database is composed of fingerprint (F) images collected from individuals either by using fingerprint cards or by electronic capture using a device similar to a scanner. The identification (I) aspect occurs when the person is fingerprinted, and the resulting images are searched against the database of fingerprint images on a local, state, or national database. It is considered a system (S) because it uses computers and software and can interact with subsystems and other identification systems, including other AFIS systems".

AFIS systems is an identification tool for almost every law enforcement agency in all over the world. The system is very flexible: at the same time it can comprise information about e.g. one city or country and huge numbers of records (e.g. 46 million records held by the FBI) as well. Another important feature is that AFIS system can be linked to other databases (Komarinski, 2005). It allows to create larger networks of the technology and applications of AFIS systems, to develop more accurate programs or even new products related to fingerprinting.

AFIS systems search databases for candidates based on image characteristics. The characteristics include the points where ridges end, the points where they split, the directions that ridges appear to flow, and even dots. The AFIS system translates what a human sees as a picture, selects key features, searches these features against a database, and produces the best match from that database (Komarinski, 2005).

Probably the largest AFIS system is the Integrated Automated Fingerprint Identification System (IAFIS), operated by the Criminal Justice Information Services (CJIS) division of the FBI in the USA. However, the development of AFIS system is not restricted to the USA only. Several countries in Central America, the Middle East, Asia, and Africa require that all their adult citizens be fingerprinted. For that reason, AFIS systems are used to confirm these identifications. In these countries, AFIS can be used in determining eligibility for government benefits, to ensure that people do not exceed their social services benefits or services such as voting.

The Genesis of AFIS

Looking to the genesis of DNA fingerprinting, identification systems did not originate with AFIS systems. Before AFIS fingerprints were classified by hundreds of staff, who were entrusted with the responsibility of confirming, based on fingerprint images, that a subject did or did not have a criminal record. While the card was being classified, other technicians looked through their records to see if the subject’s name was already on the fingerprint files. (Komarinski, 2005) It was a slow, time-consuming and labor-intensive process undertaken by hundreds of specially trained examiners. Naturally, a new automated approach was needed to extract each fingerprint image from a tenprint card, process each of these images to produce a reduced-size template of characteristic information, and search a database to automatically produce a highly reduced list of probable candidate matches (Cole, 2001, pp 251–252).

In the early 1960s, the FBI in the United States, the Home Office in the United Kingdom, Paris Police in France, and the Japanese National Police initiated projects to develop automated fingerprint identification systems. (Mosses, 2010) In 1963 special agent Carl Voelker initiated attempts to automate the manual searching of the criminal files. He sought the help of engineers Raymond Moore and Joe Wegstein of the National Institute of Standards and Technology (NIST) (Mosses, 2010). They believed that a computerized solution to automatically match fingerprints could be developed that would operate in a manner similar to the techniques used by human examiners. But to achieve this goal, three major tasks would have to be accomplished. First, a scanner had to be developed that could automatically read and electronically capture the inked fingerprint image. Second, it was necessary to accurately and consistently detect and identify minutiae [1] existing in the captured image. Finally, a method had to be developed to compare two lists of minutiae descriptors to determine whether they both most likely came from the same finger of the same individual. (Mosses, 2010)

To address the first two of the three tasks, on December 16, 1966, the FBI issued a Request for Quotation (RFQ) "for developing, demonstrating, and testing a device for reading certain fingerprint minutiae" (FBI, 1966). The 14 proposals received in response to this request. After the evaluation of proposals, two separate projects were funded to provide a basic model for reading fingerprint images. Both proposed to use a flying spot scanner [2] for capturing the image. But each offered a different approach for processing the captured image data, and both seemed promising. One contract was awarded to Cornell Aeronautical Labs, Inc., which proposed using a general-purpose digital computer to process binary pixels and develop programs for detecting and providing measurement parameters for each identified minutiae. The second contract was awarded to North American Aviation, Inc., which proposed using a special-purpose digital process to compare fixed logical marks to the image for identifying, detecting, and encoding each minutiae (Mosses, 2010).

The third task of comparing two minutiae lists to determine a candidate match was addressed by Joe Wegstein (Wegstein, 1970). He developed the algorithms for determining fingerprint matches based on the processing and comparison of two lists describing minutiae both location and orientation.

After a year’s experience with the prototype system, the FBI issued a new request containing additional requirements such as a high-speed card-handling subsystem. In 1974, Rockwell International, Inc., was awarded a contract to build five production model automatic fingerprint reader systems. This revolutionary system was called Finder (Mosses, 2010).

AFIS Today

After the events of September 11, 2001 the biometric surveillance has moved to the new phase. Increasing concern about safety has become a reason of new successful attempts to improve technology of surveillance, including DNA fingerprinting and AFIS. According to the International Biometric Industry Association, the biometrics industry has grown from 6400 devices shipped in 1995 to 400 000 in 2001, worth $168 million (The Business of Federal Technology, 2001).

As the importance of AFIS system is increasing nowadays, it is essential to identify the general strengths, weaknesses and future opportunities of AFIS systems. The same important is to overview the leading companies of biometric surveillance and its products. The main attention will be paid to the NEC Corp. and its current issues. In addition, the analysis of the case of the Indian Unique ID Project will be given.

Strengths and weaknesses

Strengths

Automated Fingerprint Identification Systems have become a widely accepted method of identifying persons in a one-to-many (1 : N) search that provides with accuracy and quick response (Komarinski, 2005). During 50 years of AFIS development, the following strengths of the AFIS systems can be emphasised today:

The systems are more reliable now than they were in the past. Plus, the hardware and operating software can be more easily repaired and replaced because of the standartization.

The computer equipment and software now in use are very reliable.

New computers require less energy, generate less heat, last longer, and are less likely to break down before scheduled maintenance. They are also less expensive.

The emergence of AFIS technology into other market areas has brought indirect benefits to the forensic AFIS community since it means more customers for AFIS products.

The identification of latent prints has increased geometrically with AFIS. AFIS systems can compare the image characteristics of a latent print against all the millions of records in the AFIS database and produce a list of candidates based on matching score.

The current methods for developing latent fingerprints are all used successfully in forensic investigations. Moreover, the possibility that a fingerprint can provide more information about a person than just identity is particularly exciting. For example, information about whether a person has taken narcotic drugs or has been in contact with explosive materials has recently been reported by using methods to develop latent fingerprints.

Weaknesses

Despite the fact that significant results have been achieved in DNA fingerprinting, there are still some issues to deal with. The most significant ones are the following:

Lack of interoperability. For example, in the case ofthe USA the database of millions of AFIS records available in one state is not immediately available to another state for searching. Moreover, Each of the major AFIS vendors has developed its own application software, and all of the software does not work on a single platform. The implication of the lack of interoperability many latent print identifications that could be made are not. The record of a burglar, for example, may not be present on a local database because the burglar was never arrested in that locale (Komarinski, 2005).

Maintenance. AFIS are incredibly precise and new systems typically functions as expected. However, after hundreds of cycles devices may begin to show the signs of use. For example, the computers that run the coder and matcher software have limited life span and have to be replaced at scheduled intervals. Otherwise, monitors can develop quirks that can result in looseness. Hence, if proper maintenance is not performed, it can lead to equipment failures that are expensive and timeconsuming to fix.

Training. The success of AFIS systems greatly depends on the staff. It is neccesary for the staff to know both the characteristics of the system and ways how to exploit the existing performance. The training process need to be the unseparatable part when working with AFIS systems.

Challenges and Future Opportunities of AFIS

Probably the most important issue is related to the interoperability of AFIS. Despite the fact that much good work has been done in recent years (e.g. quicker and more efficient searches, better use of law enforcement resources, etc.) there is still enough space for the future improvements.

"The technical challenges to AFIS interoperability involve both those that are encountered and addressed by the information technology community in other disciplines (such as data sharing and algorithmic performance) and those that are specific to AFIS and the sharing of fingerprint information (e.g., feature identification, reliability of latent print comparisons)" (National Research Council, 2009).

Therefore, the following innovative improvements can be done in future years:

To create solid technical standarts to eliminate the conflicts between different systems and software. It is very important because even different versions of similar systems from the same vendor sometimes cannot share fingerprint data with one another. This task is greatly challenging since AFIS system encompass a diverse array of stakeholders, proprietary systems, and ever-advancing technological capabilities. HoHhhHowever, some examples, like banking system or the Interner, are the proofs that it is possible to create a number of finely crafted and agreed standards and protocols that allows greater collaboration and sharing of information (National Research Council, 2009).

Improvement in quality and accuracy in latent print analysis. Today‘s system still often require substancial input from fingerprint examiners (e.g. " If the ridge count between two clusters of points in a latent is unclear, IAFIS may fail to retrieve the true source of the print. [3] " Clearly, there is significant room for improvement. National Research Council (USA) suggests that an ideal, comprehensive AFIS would be capable of automated: reading of latent prints; encoding of most features of usable quality (fingerprint classes such as whorl, arch, pores, cuts, and ridge paths), together with a provision for including other features that could be defined by the vendor/user; recognizing absent, blurred, double/multioverlap, poor-quality sections of an observed print and encoding the system to downweight, or omit entirely, during the search process (National Research Council, 2009).

The development and use of a secure Web interface (or an analogous system) that would permit authorized latent print examiners to submit queries to databases.

In addition, to achieve better results in AFIS interoperability, AFIS equipment and service vendors should cooperate with each others. However, in a technology sector in which product differentiation and the maintenance of competitive advantages are prime concerns – vendors have little incentive to design their systems to enable them to share information with competitors’ systems (National Research Council, 2009). For that reason the possibility of cooperaion remains for the future.

The Leading Companies of Biometric Surveillance and Fingerpriting

"The technology and applications of AFIS systems are just beginning to emerge from initial development. The scope of this technology has moved from a select few uses to everyday uses. The core of AFIS technology, the computer and related software, progresses on an almost daily basis. In particular, the software that runs AFIS systems improves constantly as companies develop faster, more accurate programs. New markets have emerged in AFIS-related applications as manufacturers carve out niche products", - says P.Komarinksi (Komarinksi, 2005).

Today there are many companies which specialize in biometric technologies including fingerprint biometrics.

Company

Country

Products

NEC Corp.

Japan

IT Solutions Business, Carrier Network, Social Infrastructure, Personal Solutions

Morpho

France

Wide variety of security solutions and products

Suprema

Korea

Fingerprint technologies for PC and embedded applications

S.I.C. Biometrics

Canada

Mobile Biometrics

Digital Persona

USA

Authentication Software, Fingerprint Biometrics

Advanced Optical Systems, Inc.

USA

Biometric products, processing cards, Autonomous Positioning Systems

Lockheed Martin

USA

Aerospace and Defense, Information Technology, Space (satellites, etc.)

Table 1. The Leading Compoanies in the Fingerprint Biometrics

Source: Authors‘ Construction. The data was taken from the official pages of the companies

However, in this paper the NEC Corporation, its products, competitiveness and challenges will be discussed.

NEC Corporation

NEC Corporation is the world‘s leading company in IT solutions, carrier network and social infrastructure (including surveillance systems). Founded in 1899, the company achieved remarkable results. NEC has been taking part in the development of fingerprint matching technologies since the 1970's. In 1982, the company realized a first-of-its-kind matching algorithm using "relation" (ridge count between minutiae), which returned a remarkably high accuracy rate. In large-scale fingerprint-related technology evaluations sponsored by NIST, NEC has gained high marks for its technological strength as the company was ranked number one in tests such as the PFT Study in 2007; SlapSeg04 in 2004 and FpVTE2003 in 2003. As a result of these technological evaluations, NEC's fingerprint identification systems have been used by police forces, judicial branches and immigration offices in Japan as well as in 24 states, including California, in the United States and 22 additional countries.

In 2009 NEC Corporation and its latent fingerprint matching technology has been ranked number one in the world on the Evaluation of Latent Fingerprints Technologies (ELFT) test carried out by the National Institute of Standards and Technology (NIST), commissioned by the U.S Department of Homeland Security.

NEC‘S AFIS features are the following:

Fingerprint and palmprint identification

Positive Identification (PID)

Advanced processing for finger, latent, palm and slap matching

Enhanced Search Sending to Other AFIS

Seamless interface to live scan units, mugshot systems, criminal history systems, etc.

The prime advantages that separate NEC from its competitors are:

The NEC AFIS is not only a tenprint registration system but also a fingerprint repository for the identification of dangerous criminals;

Latent identification accuracy and speed;

NEC AFIS is the first AFIS that uses ridge counts and relationships between minutiae in its matching algorithm.

The Case of The Indian Unique ID Project

Talking about latter challenges that NEC Corporation is facing, the most important is The Indian Unique ID (UID) is an extremely-large-scale system that attempts to identify India’s 1.2 billion people, which is about 1/6th of the world population, using biometric authentication.

The Indian UID is a program conducted by the Indian electronic government. The program will be applied in a wide range of authentication operations, like tax payment, welfare and banking. The project was begun in 2009 and the registration/matching of 200 million citizens was completed in the end of 2012. Thereafter, the ID issuing speed will be accelerated in order to complete registration of 600 million citizens by the end of 2014.

The implementation of this system is a serious challenge for the NEC Corporation. This is an unprecedented extremely-large-scale system aiming at authenticating the 1.2 billion people of India. For that reason the accuracy and technical precision are essential. Naturally, the NEC Corporation faces a number of issues. NEC has identified the following four issues through analysis and study of the system (Su, Sakamoto 2012):

Issue 1: Multimodal authentication accuracy. The most important thing is to to collect biometric information of a quality, suitable for authentication. However, the population of India is the second largest in the world. For that reason, the is a need to have qualified and experienced teams that can properly collect biometric data from citizens. Another problem is the quality dependency on differences in the living environments of people and their occupations. For example, the fingerprints of workers in the agricultural sector are often damaged. Hence, the NEC Corporation need to consider all quality factors and to develop matching algorithms.

Issue 2: Optimization of multimodal authentication processing. In the field of information processing, the search index is developed in advance from the registered population. In biometrics, information taken at different times may not strictly match with each other due to aging and conditions prevailing at the time of collection. As a result, it is generally difficult to develop the index effectively. For example, in order to confirm that a person requesting registration is not duplicated, all that can be done is to match the data with all of the registered data. Therefore, to enable duplication checks on 600 million people in three years, NEC has to implement a system capable of more than 1.9 billion matches per second without interruption.

Issue 3: Scalability. Even when individual biometric authentication operations can be optimized, a scalability issue will be posed if the system is incapable of an efficient response to matching requests from the extremely-large-scale biometric database.

Issue 4: High-availability system designing. as this system is used as a government authentication platform, it is subjected to severe fault tolerant performance requirements. The main problems in this stage are:

No biometric information should be lost, even in case of a fault.

All operations should be able to continue, if the fault is in a single component only.

Multiple data centers should be operated as a countermeasure against disasters as well as for load distribution.

Despite the number of issues, the NEC Corporation has found solutions for the issues 1-3, while the 4th issue is related to actual implementations.

Conclusion

Fingerprinting is a refined methodology that is proven in practice and accepted in courts of law. The AFIS was originally used by the U.S. Federal Bureau of Investigation (FBI) in criminal cases. Lately, it has gained favor for general identification and fraud prevention. AFIS itself has been around for more than 50 years.

Before AFIS, fingerprinting and identification process were implemented only by the hundreds of staff. The process was time-consuming, labour-intensive and very slow. Therefore, the need to create automated system arose.

The most significant researches were undertaken in the USA because the initiative to create the AFIS system was generally conducted by FBI.

The new approach to the innovations in fingerprinting has arisen after the terror attacks in the USA in 2001. The concern about safety and the need of surveillance have grown over the period of time. Naturally, the need to improve the AFIS systems and search for new possibilities and opportunities has developed.

The AFIS system today presents a number of its strengths: reliability, flexibility, energy saving, the emergence of technology into other market areas that brings new opportunities, etc.

However, there is a place for further improvements. The main weaknesses of AFIS are: lack of interoperability, permanent need for maintenance of system and staff training. These weaknesses, at the same time, are identified as a future opportunities for improvements.

There are many companies that specialize in biometric technologies including fingerprint biometrics. However, in this paper the main attention was paid to the NEC Corporation (Japan).

The NEC Corporation has gained recognition as a leading company in fingerprint biometrics.

The Case of The Indian Unique ID Project shows us that bionetric technologies are moving forward. The scope of biometric surveillance technologies is growing year by year. However, new challenges arise together with new opportunities.