Biometrics: A Technology Assessment
By: Alex English, Christina Means, Kris Gordon, Kevin Goetz


Biometrics: A Technology Assessment
By: Alex English, Christina Means, Kris Gordon, Kevin Goetz


Introduction


The events of September 11, 2001 have caused an increase in security concerns across the world. The U. S Government has moved towards using biometric technology as a means for preventing any further terrorist attacks and ceasing the ever increasing identity related thefts. Biometrics is the science of using a persons biological properties to verify identification such as, fingerprints, voice recognition, iris recognition, hand geometry, and signature verification.

The purpose of this paper is to determine which biometric device is most reliable and cost effective to meet the needs of society in deterring identity theft and safeguarding the security of student confidentiality in education.


TOP

Chapter 1: Fingerprinting


Overview

Fingerprinting is the oldest and most widely used form of identification. Law enforcement agencies have used this method to find and convict criminals for several years. This method can not be forged since each person has their own unique skin characteristics. However, the means for capturing fingerprints is not full proof. The contact-based devices over time reduce the quality of the image resulting in errors. (Biometric Technology, Inc. 2002)

The current fingerprinting systems offer a wide variety of technologies in capturing, storing and verifying searches by the use of large databases. Fingerprinting will remain the leading biometric technology through 2008. (International Biometric Group, 2004) This biometric system is projected to grow from $198 million in 2003 to $1.493 billion in 2008 due to the versatility of fingerprinting. (International Biometric Group, 2004)

Fingerprints

Fingerprinting is the oldest method of successfully matching an identity. A person’s fingerprints are a complex combination of patterns known as lines, arches, loops and whorls. (Biometric Technology, Inc. 2002) The most distinctive characteristics are the minutiae, the smallest details found in the ridge endings. There are several factors in favor of using fingerprinting for the purpose of an identification system. Fingerprints can not be forged and every individual has a unique print. Due to the several layers of skin that makes up the fingerprint, chemicals can not erase the uniqueness of the ridges. Although some prints can be altered due to finger surgery or hand injury. (Biometric Technology, Inc 2002)

Obtaining a high quality image where the fingerprint ridges and minutiae are recognizable is a complex procedure. The area from which to take measurements is extremely small, thus sophisticated devices have been developed to capture an image with significant detail. (International Biometric Group, 2004) A fingerprint capture system accomplishes the task in a variety of ways. The following devices are capture methods used today: reflected light optical, solid-state capacity inductance scanners, ultrasound scanners, micropad pressure sensors, and polarized multi-frequency infrared illumination. These capture methods differ in terms of convenience, cost and accuracy. (Biometric Technology, Inc. 2002)

Advantages:(Biometric Technology, Inc. 2002)

  • Prints remain the same throughout a person’s lifetime.
  • Fingerprinting is neither frightening nor emotionally disturbing.
  • People’s prints are unique.

Disadvantages:(Biometric Technology, Inc. 2002)

  • Searching through a huge database can be rather slow
  • Dirt on the finger or injury can blur the print.
  • A fingerprint template is rather large compared to other biometric devices.

The fact that fingerprint files are extreme large (250 Kb) compared to other biometrics, has created a problem for installing fingerprint biometric data on portable ID cards. The users must have databases at each verification site or create a way to download data to a central site of identity verification, which increases cost and slows down the matching process. (Biometric Technology, Inc. 2002)

Summary

Fingerprint verification is seen as the most trusted and convenient method. The cost for such technology is becoming easily affordable as more industries enter the marketplace. (Frost & Sullivan, 2003) Even with the prices of fingerprinting systems falling, it is difficult to forecast the overall cost and cost-effectiveness of the biometric security system. (Ulfelder, 2004)


TOP

Chapter 2: Signature Verification

Overview

Signature verification is the process used to recognize an individual’s hand-written signature. (Topick Z Inc. 2004) In order to have a full understanding of signature verification and its possible uses, it is important to look at the different types of signature verification.


Signature Verification Type: Capture

Capture signature systems are those systems that allow a person’s signature to be “captured” electronically for the purpose of a transaction validation and forwarded to remote locations. Most of us who use credit cards have experienced signature capture when signing for purchases on a small electronic tablet device. The signature is captured with the receipt and can be forwarded anywhere via. electronic mail. This type of device can only serve to validate a transaction and does not serve as a means to identify the signer. Another means of identification would be needed to verify that the person signing is actually that person. Many banks use a type of capture signature verification, which allows them to quickly compare signed checks against stored signature cards. This system uses signature comparisons and has helped banks cut down on forgery but has limited uses for means of identification.

Signature Verification Type: Dynamic

The second type of signature verification can also be called dynamic signature verification. There is an important distinction between simple signature comparison and dynamics signature verification. Both can be computerized, but a simple comparison only takes into account what the signature looks like. Dynamic signature verification takes into account how the signature was made. With dynamic signature verification it is not the shape or look of the signature that is meaningful; it is the changes in speed, pressure and timing that occur during the act of signing. Only the original signer can recreate the changes in timing and X, Y, and Z (pressure). (Topick Z Inc. 2004) This unique means of analyzing a signature makes it virtually impossible for another person to duplicate the timing changes in X, Y, and Z. A pasted bitmap, a copy machine or an expert forger may be able to duplicate what a signature looks like, but the natural motion of the original signer would be required to complete the signature. (Topick Z Inc. 2004)

Signature Verification: Costs

The costs of implementing a dynamic signature verification system, it appears to be on the low end as compared with other biometric systems. Software, user license, and a graphic tablet or PDA is all that is required to operate a signature verification system. Cyber Sign charges $1,500 per 10-user agreement, which breaks down to about $150.00 per user. Several other companies were contacted but at the time of this writing, have not yet responded.

Although signature verification appears to be within a reasonable price range, it would still be too expensive for most public schools across the country. Federal cutbacks and state mandates have most school budgets at their limits. Public opinion does not appear to be on the side of spending thousands of dollars to implement these new systems for the increased security they offer.

Advantages

One of the advantages of this type of biometric system is the fact that signatures have been an accepted means of identity verification for centuries. This promotes a biometric technology that is easy for organizations and consumers to accept and to trust. “Signatures are considered to be the least intrusive form of authentication and enjoy a high degree of social acceptability. Thus, a move toward adopting signature verification biometrics for electronic transactions is expected to be smooth”. (Business Wire 2004) This acceptance seems to be supported by the United States Government who in the year 2000 passed the U.S. eSign Bill and related legislation. This legislation extended the legal status of handwritten signatures to the electronic equivalent, thereby promoting the growth of e-transactions nation wide. (Business Wire 2004) Therefore the need to draft new legislation concerning signature verification is not necessary at this time.

Another advantage of a signature verification system would be in the replacement of passwords, PINs, or keycards. Identification that can be stolen, lost or forgotten is eliminated and replaced with a simple signature. (Cybersign 2004) Logging on to a secure computer system from a remote location could be done with a graphics tablet and a signature. Other advantages are as follows:

  • Low total error rate (about 1.5% per session)
  • Forgery is detected even when the forger has managed to get a copy of the authentic signature
  • Possible detection of inconsistent user during enrollment stage
  • Fast and simple training
  • Cheap hardware (almost any tablet device is allowable)
  • Little storage requirements
  • Fast response (about 1 sec per signature on the “old” 486DX-33 computer)
  • The results do not depend on the native language of the user
  • You can use any kind of information as your signature: name, second name, or even nice curves
  • Very high compression rate (100-150 bytes are needed to keep the shape of the signature)
  • The system represents a natural way to prove authenticity. (IBM 2004)

Disadvantages

Having presented many advantages for the use of signature verification, disadvantages will also be presented so that a fair analysis can be made. Signature verification is designed to verify subjects based on the traits of their unique signature. As a result, individuals who do not sign their names in a consistent manner may have difficulty enrolling and verifying in signature verification. Individuals with muscular illness and people who sometimes sign with only their initials might result in a higher False Rejection Rate (FRR), which measures the likelihood that a system will incorrectly reject an authorized user. (International Biometric Group 2004)

Signatures can also be affected by behavioral factors; stress or distractions could cause a person to vary from their normal signature sequence thereby generating a rejection. (GSA/FTS Network Services Conference 2003) There has been presented information regarding the frequency of use factor. It has been stated that a person’s signature could vary over a period of years. Therefore the person not using a signature verification system for an extended period of time may develop different characteristics in their signature thereby causing the signature to be rejected.

Summary

In an effort to address the questions concerning the use of biometrics in a school, university, or an on-line class situation, the conclusions drawn are as follows:

Signature verification would be an excellent way of securing computers and computer labs. Having students sign-in would mean that the student was actually there, and that it was not in fact another student who had the correct password. Student sign-in could also be useful in other areas of every day operations, such as signing in and out of a secure building, or verifying identity in the cafeteria or book store.

As far as an on-line class situation, there seems to be no easy solution, short of surveillance cameras. The point being made is that anyone could sit down at the computer once the correct student has logged in. The bottom line is that there is nothing available at this time that can positively guarantee that the person operating a computer is who he/she says they are. The conclusions then must be that there are some situations that this type of biometrics may be useful, while in other areas biometrics are not foolproof.

One last point to consider, although there has been a major focus into the recent intrusions into our constitutional rights due to the heighten security issues here in the United States; signature verification in its simplest form is older than our constitution. For centuries man has been legally identified by his signature or mark. It has been an accepted means both here and around the world and as stated earlier, signature verification is considered to be one of the least intrusive forms of authentication and enjoys a high degree of social acceptability. Being able to verify a person’s identity is rapidly becoming an important part of our modern society. In these times of ongoing terrorist threats, we all may have to sacrifice some part of our constitutional rights for the protection of our country, our families, and our way of life.


TOP

Chapter 3: Voice Authentication

Overview

The human voice is unique because of the “physiological and behavioral aspects of speech production”. (Speaker Verification, 2004) The shape of the vocal tract in humans is what makes the voice unique. The vocal tract’s location is depicted by the shaded area in the image below.


Vocal Tract
Speaker Verification, 2004

There are two models used to make a voiceprint that can be stored in a database to verify at a later time. The two models are as follows: stochastic model and template model. (Speaker Verification, 2004) In short the stochastic model takes random words and samples the speech to determine parameters and the template model takes samples from the same word spoken multiple times. (Speaker Verification, 2004) Both of these models are much more complex and for the purposes of this document it is not necessary to understand in detail how these models work. However it is important to know that different voice authentication software solutions use one of these two models.

Finally a voice authentication system matches the pattern of a voice that was obtained using one of the above mentioned models and compares it to a database of voiceprints. The image below shows this process.


Process Image
Speaker Verification, 2004

Voice Authentication: Security and Costs

Like fingerprints the human voice is specific to a single person. Therefore according to Ron Croen, CEO of Nuance, no ones voice can be “altered, forged or stolen.” (Speech Technology Magazine, 2002) Aside from being very secure voice authentication is a very cost effective solution. The reason voice authentication is cost effective is because unlike other biometric solutions, voice authentication, does not require any special hardware to make it work. According to Claudia Graziano, a reporter for Wired News, “verification can be done using standard sound cards and microphones…software will cost $995 and up, depending on the number of biometric profiles to be stored.” (Wired News, 1998) Cost effectiveness is very important when considering a large deployment.

Another feature that makes voice authentication a viable biometric solution is its integrity. With the proper voice authentication technology the likeliness of using tape recorded voices to gain access to secure information/areas is greatly reduced. The proper technology will test for liveness. The way this technology works is by listening for acoustic patterns that can be found in a voice that has been tape recorded. (JMC Publications, 2004) This type of technology helps to prevent users that are not physically at a location from gaining access.

Areas of Opportunity: Education

The features mentioned above make voice authentication a viable solution in the education vertical. Voice authentication requires virtually no new hardware because most PCs built today already have the necessary sound cards and microphones built-in. This makes voice authentication a cost effective solution especially for educational institutions with limited budgets.

There is also very little time and effort required to record a persons voice to use for authentication. A user would simply need to speak into a microphone and they are done. Although there might be some additional information that would need to be entered (i.e. social security number, name, address). Another feature is that you do not have to remember your voice, unlike pass codes.

This technology could be used to log users on to computers, allow them access to rooms without the need for a physical key which could be lost or stolen. Voice authentication could also be used in environments where a large number of students are taking a test to verify who they are before they are allowed to receive the test.

Finally voice authentication could be used for distance learning situations. A company called SAFlink Corp. has developed a voice authentication tool that can be used over the World Wide Web to transmit a voiceprint back to a remote server for verification. (Wired News, 1998) By using this technology students taking an online course would be required to use their voice to verify who they are before they are allowed to participate in the class.

Summary

Voice authentication technology is a viable solution to securely and inexpensively authenticate users both at a physical location and remotely. By using technology that is already available in most PCs on the market today voice authentication limits costs for hardware. Even if new hardware is required it is relatively inexpensive. Voice authentication is secure because every human has a distinct and unique voice which is virtually impossible to duplicate even using tape recording devices.


TOP

Chapter 4: Iris Recognition

Overview

The iris is the plainly visible, colored ring that surrounds the pupil of one’s eye. It is a muscular structure that controls the amount of light entering the eye, with intricate details that can be measured, such as striations, pits, and furrows. The iris is not to be confused with the retina, which lines the inside of the back of the eye. (Iridian Technologies Incorporated, 2004) The iris recognition biometric technology uses the measurable features of the iris to create mathematical algorithms of the iris. The algorithms are then stored and later compared with new algorithms of irises presented to a capturing device for either identification or verification purposes.


Eye Diagram
This diagram is of the basic internal structure of the eye. The image was taken from www.iridiantech.com

Iris recognition technology was developed by Dr. John Daugman, and is patented solely by Iridian Technology Incorporated. (IrisGuard Incorporated, 2004) There are three basic steps to iris recognition. The first step involves capturing or acquiring an image of the iris. This step is generally fulfilled by a person standing in front of a camera. The camera then takes a picture of the iris using visible and/or infrared light. (Iridian Technologies Incorporated, 2004) The second step is that of converting the image to what is called an iriscode™. In this step, the digital image is filtered, by an algorithm, to map segments of the iris into hundreds of vectors, also known as phasors. The visible characteristics including the tribecular meshwork (the appearance of radial divisions in the iris), rings, furrows, freckles, and corona of the iris are mapped into the different phasors and stored as hexadecimal representations in a computer. (Iridian Technologies Incorporated, 2004) The third and final step of iris recognition is to search an already known database of Iriscode™ information for a match with the Iriscode™ of a new sample. (IrisGuard Incorporated, 2004)


EyeEye
The image on the left shows the visible characteristics of the iris (www.cl.cam.ac.uk/users/jgd1000), while the image on the right gives a "camera's eye" view of the subject (www.iridiantech.com)

Iris recognition in an opt-in technology, which means that the user must cooperate with the system for the technology to be used. Since precise measurements must be taken with the image, the subject must hold reasonably still in a specific location, even if momentarily, in order for the image to be taken. (International Biometrics Group, 2004) Systems in which the user is infrequent may encounter ease-of-use problems. When the biometric is used more frequently, the interaction becomes easier to use. Glasses and colored contact lenses can also alter the image of the iris. These items cause glare and color changes in the image, although the algorithms created by Dr. John Daugmen recognize and account for most problems when an iris region is obscured by eyelids, contains any eyelash occlusions, specular rejections, and boundary artifacts of hard contact lenses. (John Daugman, 2004)

The converted hexadecimal representation (Iriscode™) of the iris is stored into a 512 byte template. From the iris’ 11mm diameter, algorithms provide 3.4 bits of data per square mm. This density of information is such that each iris is said to have 266 ‘degrees of freedom’ instead of the 13-60 for most other biometric technologies. (International Biometrics Group, 2004) A key difference between iris recognition technologies and others is that its 512 byte templates facilitate extremely fast match speeds. On a 300 MHz CPU, such exhaustive searches are performed at a rate of about 100,000 irises per second. On a 2.2 GHz server, one million IrisCodes™ can be compared in 1.7 seconds. (International Biometrics Group, 2004) Iris recognition technology boasts an extremely low false recognition rate (FRR) of 1:1,200,000. (Panasonic Inc., 2004)

Areas of Opportunity

Iris recognition technology is currently used in many locations and for many reasons. In 1996, Lancaster County Prison in Pennsylvania became the first correctional facility to use iris scanning. Warden Vincent A. Guarini said the facility sometimes needs to release prisoners on short notice and can't wait for fingerprint tests. (Michael Meehan, 2001) In the largest national deployment of iris recognition to date, the United Arab Emirates (UAE) Ministry of Interior requires iris recognition tests on all passengers entering UAE from all 17 air, land, and sea ports. Their Iriscodes™ are then checked againt those of deportees and since its inception the program has caught 7,253 deportees returning to that country. (John Daugman, 2004) Most uses of iris recognition technology are for business purposes of access to offices, laboratories, computers, and bank accounts. The Venerable Bede School in the United Kingdom uses this technology with its 900 students to check-out library books and for cafeteria payments. (International Biometrics Group, 2004) Most recently iris recognition technology has been used in Pakistan to limit Afgan refugees to one cash grant each by the United Nations. (John Daugman, 2004)

Advantages

  • Highly protected, internal organ of the eye
  • Externally visible; patterns imaged from a distance
  • Iris patterns possess a high degree of randomness
    • variability: 266 degrees of freedom
    • entropy: 3.4bits per square-millimeter
  • Pre-natal morphogenesis (7th month of gestation) (John Daugman, 2004)
  • Limited genetic penetrance of iris patterns
  • Patterns apparently stable throughout life (John Daugman, 2004)
  • Search speed: 100,000 IrisCodes per second on 300MHz CPU

Disadvantages

  • Small target (1 cm) to acquire from a distance (1 m)
  • Moving target ...within another... on yet another
  • Located behind a curved, wet, reflecting surface
  • Obscured by eyelashes, lenses, reflections
  • Partially occluded by eyelids, often drooping
  • Deforms non-elastically as pupil changes size (John Daugman, 2004)
  • Illumination should not be visible or bright
  • Some negative (Orwellian) connotations (John Daugman, 2004)

Summary

Iris Recognition Technology is versatile. Systems can be relatively inexpensive at the cost of ~ $5,000 each depending on the application. (Michael Meehan, 2001) Successful previous applications for high security area’s such as prisons, US Congressional offices, US Department of Treasury, and US Vice President Cheney’s offices provide powerful information on the reliability and accuracy of this biometric. (John Daugman, 2004) Iris recognition is set to grow substantially from a $36 million dollar industry in 2003 to a $366 million dollar industry in 2008. (International Biometrics Group, 2004) Although, there could be several ease-of-use issues when administering this technology into a school system, it is a viable option in education. This biometric technology could be an option for educational uses such as library check-out; cafeteria payments; access to buildings for faculty, staff, students, and parents; parent verification for student pick-ups; access to computers; and attendance to name a few.


TOP

Conclusion

Everyone at one time or another has forgotten the password to their PC. This costs the company time and money. The solution is to install biometric systems in the work place. Now, with the threat of terrorist and identity fraud, biometric systems are being installed in everyday civilian life as well. The need for biometric systems is expected to grow significantly over the next few years, maybe even reaching over $1 billion. (Ganapathi, 2002)

These identification systems offer the public a promise in avoiding error or fraud. The desire for improved security has fueled the need for biometric security systems. Whether privacy advocates agree or not with the usage of biometrics still remains to be seen. However, with the increasing demand for homeland security, the use of biometrics will need to conform to privacy standards in order to appease the every day civilian.


TOP


References

  1. Biometric Technology, Inc. (2002). “Biometric Technical Assessment.” Retrieved April 4, 2004 from http://www.bio-tech-inc.com.
  2. Business Wire. (2004). “Signature Verification Biometrics Market to Overcome the Effect of Reduced IT Expenditures.” Retrieved April 16, 2004 from http://home.businesswire.com.
  3. Cybersign. (2004). “Biometric Signature Verification.” Retrieved April 16, 2004 from http://www.cybersign.com/index.html.
  4. Croen, Ron. (March/April 2002). “The Business Case for Voice Authentication.” Speech Technology Magazine Online. Retrieved April 15, 2004 from http://www.speechtechmag.com/issues/7_2/cover/586-1.html.
  5. Daugman, J. & Downing, C. (2001). “Epigenetic randomness, complexity, and singularity of human iris patterns.” Procedings of the Royal Society, B, 268, Biological Sciences, pp 1737-1740.
  6. Frost & Sullivan. (2003). “World Non-AFIS Fingerprint Biometrics Markets.” Retrieved April 2, 2004 from http://www.marketresearch.com/map/prod/935284.html.
  7. Ganapathi, S.K. (2002). “Fingerprint Authentication: Shifting the Electronic Security Paradigm.” Retrieved April 4, 2004 from http://www.fidelica.com/Documents/Feb_02_SC_Magazine.pdf.html.
  8. Graziano, Claudia. (September 1998). “Use Your Lungs to Log On.” Wired News. Retrieved April 17, 2004 from http://www.wired.com/news/technology/0,1282,14913,00.html.
  9. GSA/FTS Network Services Conference. (2003). “Managing the Future: Mastering the Maze, Dynamic signature verification.” Retrieved April 14, 2004 from http://www.fts.gsa.gov/2003.
  10. Kulkarni, Y. & Jain, A. (n.d.). “Speaker Verification.” Retrieved April 17, 2004 from http://biometrics.cse.msu.edu/speaker.html.
  11. Markowitz, J. (n.d.). “Frequently Asked Questions.” JMC Publications. Retrieved April 13, 2004 from http://www.jmarkowitz.com/ask.html.
  12. IBM. (2004). “Online Signature Verification.” Retrieved April 8, 2004 from http://www.haifa.il.ibm.com/projects/image/sv/.
  13. International Biometric Group. (2004). “International Biometric Group – Fingerprint Feature Extraction.” Retrieved April 17, 2004 from http://www.biometricgroup.com/about_us.html.
  14. International Biometrics Group. (2004). “Iris recognition: How it works.” Retrieved April 16, 2004 from http://www.biometricgroup.com/reports/public/reports/iris-scan_tech.html.
  15. International Biometrics Group. (2004). “Iris recognition issues.” Retrieved April 16, 2004 from http://www.biometricgroup.com/reports/public/reports/iris-scan_issues.html.
  16. International Biometric Group. (2004). “Signature Verification.” Retrieved April 10, 2004 from http://www.ibgweb.com/reports/signature-scan market.html.
  17. International Biometrics Group. (2004). “What factors cause biometric systems to fail?” Retrieved April 16, 2004 from http://www.biometricgroup.com/reports/public/reports/multiple_biometrics.html.
  18. Iridian Technologies Incorporated. (2004). “What is iris recognition?” Retrieved April 8, 2004 from http://www.iridiantech.com/basics.php.
  19. IrisGuard Incorporated. (2004). “Iris recognition technology.” Retrieved April 16, 2004 from http://www.irisguard.com/pages.php?menu_id=6.
  20. Meehan, Michael. (July 2001). “Iris scans take off at airports.” Retrieved April 18, 2004 from http://www.computerworld.com/securitytopics/security/story/0,10801,47202,00.html.
  21. Panasonic Inc. (2004). “Benefits of using iris technology.” Retrived April 20, 2004 from http://www.panasonic.com/cctv/products/bmet500_technology.asp.
  22. Topick Z Inc. (2004) “Find Biometrics, Understanding Signature Verification.” Retrieved April 16, 2004 from http://www.findbiometrics.com/Pages/signature%20articles/signature_1.html.
  23. Ulfelder, Steve. (2004). “Indelible Signatures.” Retrieved April 17, 2004 from http://www.microsoft.com/business/executivecircle/content/page.aspx?cID=122&subcatID=3.
  24. University of Cambridge Computer Laboratory. (2004). “Advantages and disadvantages of the iris for identification.” Retrieved April 18, 2004 from http://www.cl.cam.ac.uk/users/jgd1000/.


This page was created for ITEDU 510, Technology Use and Assessment, Ball State University.
Send comments to
Alex English

TOP

Biometrics: A Technology Assessment
awenglish@bsu.edu
http://www.bsu.edu/web/awenglish/SCHOOL/ITEDU_510/ta.html
Created by: A. English
Last Update: 4/22/04