Prior to the revolution and the mass migrations to the cities, people lived mostly in rural areas where everyone knew everyone else by name and there was little or no need for any other form of identification. In the golden days, there were no police force, no prison and a very low crime rate. As the birth rate has increased, cities have become crowded, crime rates have soar and criminals have flourished within a world of anonymity.
When the provision of identity became important, in increasing number of situations identity confirmation was carried out by checking a verbal statement of identity against the stored details in the database. Slowly as technology developed, new forms of identification came to use such as identification card (ID card), pin codes to allow access to a building and biometric identification such as using fingerprints to allow access to a secure place or to log into your PC.
Today in Afghanistan Biometric technology has become a hot topic, where recognizing people based on their physiological or behavioral characteristics is the main focus of the science of biometrics. Biometric technology is a fascinating topic and heralded as the new way forward for the public sector in both developed and developing countries and since the event of September 11 2001, we are seeing an increasing interest and practical deployment of biometric systems around the world.
As much as modern societies are increasingly dependent on systems to provide secure environments and services to people, it also becomes dominant to ensure the security of a system through different means to identify individuals requesting access to it. This is usually carried out by extracting some form of information from the individual in order to check against information held by the system to identify that individual. This paper encapsulates Biometric matching features, classification and storage on the one hand, and on the other it looks at Biometrics as a security technology to identify individuals in Afghanistan considering the associated challenges.
Biometric recognition can be described as an automated method to accurately recognize individuals based on distinguishing physical, behavioral and biological traits. It is a subset of the broader field of the science of human identification. Current technologies used in biometrics include recognition of fingerprints, face, vein patterns, iris, voice and keystroke patterns1, and there are many more in the pipeline. Following the events of September 11 2001, various biometric technologies were quickly endorsed with what was called the ‘new treat of global terrorism’3. Fingerprints have long been used for recognition of individuals due to their uniqueness and immutability4, as everybody holds a distinctive pattern of fingerprints and it offers user convenience, increased security, cost-effectiveness and non-repudiation2.
Since 2001 Afghanistan has made a remarkable progress in laying the minimum basis for the country’s recovery. But in spite of significant change in economy, higher education, social life, technology and security, the country’s recovery is still fragile and cannot be sustained without the prolonged international assistance. Mainly security in Afghanistan is becoming increasingly elusive and of prime importance. However question will be asked on how long are we going to hold hands of the international community? Is their last decade of assistance, support and sacrifice not enough?
Even though there are increasingly difficult and lots of ethical problems such as resistance from the general public, technological security will and is currently playing a major role towards maintaining the security in Afghanistan.
3. Problem Statement
Since September 2011, Afghanistan has become the only country in the world to fingerprint and photograph all travelers who pass through Kabul International Airport, arriving and departing10. Biometric have become a useful tool in Afghanistan but it is currently facing social, legal, and execution challenges.
The U.S. Military in Afghanistan is working with other authorities to record information of citizens such as their eye scans, fingerprints and facial images8. Formerly established in 2009, the FBI has also collected thousands of latent prints from the battlefield in Afghanistan9 with the help of the Biometrics Department of the Ministry of Interior; apart from this the U.S army and the afghan government have also built a database of digital records15. The problem that currently exists with the various accessible systems is that none of the systems are integrated where they can share the same Biometric and demographic data. In this paper we try to minimize these problems if not completely eradicate it.
4. Literature Review
Before a digital form of fingerprint was introduced and came to use, fingerprint were captured and matching was done manually where hundreds and thousands record of fingerprints were stored in paper-based form and accumulated in manual files. Whenever a fingerprint match was required, a huge amount of paper records were scanned manually and a match was found, which was most likely to be inaccurate and inconsistent. Apart from the inaccuracy of the fingerprint match the fingerprint match was very costly due its labor cost and search being time consuming.
Fingerprints are the type of Biometrics that is inexpensive, easy to capture, store and it does not require sophisticated hardware. Each person has ten fingers, ten unique tokens tied to his or her identity. Fingers may be scarred or cut, but can still contain enough information to link the image with the owner. Farmers can have scarred or cut fingerprints because of the work carried out but it will still contain enough information in order to identify the individual. Apart from being non-transferable among individuals, fingerprints do not provide data about the person; but rather, information of the person11. No two fingerprints have ever been found to be identical even identical twins22 although they share the same DNA but not the same fingerprints5.
The use of fingerprints for criminal verification and access control is becoming very popular19. The relevance of fingerprints in modern society has been reinforced by the demand for large-scale identity management systems whose functionality relies on accurately determining an individual’s identity. As Identity theft is becoming the fastest growing crime in the world of technology, the approach to prepare for control of crimes in Afghanistan is as essential. Fingerprint identification and verification system is the easiest way to take control of such crimes
Thousands of people travel in and out of Afghanistan everyday through the borders to Pakistan, Iran, Uzbekistan and Tajikistan. No security procedures have been considered for the road passers in and out of Afghanistan not by the Ministry of Interior or any other relevant authority. This is creating the major security risks, people flying in and out of the country are being registered demographically including being fingerprinted but the treats of people travelling by road are not considered risky.
Humans have used fingerprints as a reliable personal identification method for a very long time18. Fingerprints are one of the most mature biometric technologies and are considered legitimate proof of evidence in courts of law all over the world. Fingerprints are therefore, used in forensic division worldwide for criminal investigations. More recently, an increasing number of civilian and commercial applications are using fingerprint-based identification because of their reliability and matching performance12.
Fingerprints usually appears as a series of dark lines that represents the high, peaking portion of the friction ridge skin, while the valleys between the ridges appear as white space and are the low, shallow portion of the friction ridge skin5. Fingerprints are the unique patterns of ridges and valleys present on the surface of fingertip. These ridges mostly run in parallel direction sometimes they bifurcate and sometimes terminate. The part of area fingerprint with different patterns of curvature, termination and bifurcation is called singular region. The singular points are either delta point or core point depending upon its position and type7.
A core point is a point at which the orientation in the small area surrounding this point represents a semi circular shape and delta point is a point where a small area around the point forms three sectors with hyperbolic tendency as shown in figure 2 17. These core and delta points are used for fingerprint classification.
A fingerprint is an impression of the friction ridges found on the inner surface of a finger or a thumb. The ridges are also referred to as friction ridges. They provide a relatively rough surface area, making it possible to grasp and hold on to objects with ease5.
5. Fingerprint Matching
As mentioned above for years fingerprint matching has been done manually and today the technology has made it possible to access vast databases of fingerprint and identify possible matches within a matter of seconds. Fingerprint matching is required to associate and compare a live fingerprint with copies of fingerprints that are already stored within a database.
Fingerprint matching is one of the most popular biometric techniques used in automated personal identification20. One of the world’s largest fingerprint recognition systems is the Integrated Automated Fingerprint Identification System (IAFIS), maintained by the FBI in the US since 1999. The IAFIS currently contains fingerprints of more than 60 million individuals, with corresponding demo¬graphic information, providing both latent-print search for crime scene investigation and 10-print ID for sus¬pect identification and general-population background checks. In 2008, the FBI began updating the IAFIS to the Next Generation Identification (NGI) system, which will support other biometric traits such as palm print, iris, and face23.
There are two basic types of fingerprint matching techniques: graph based and minutiae based21. For modern embedded fingerprint recognition systems, the minutiae-based matching is popular because the minutiae of the fingerprint are widely believed to be the most discriminating and reliable features whilst on the other hand the template size of the biometric information based on minutiae is much smaller and the processing speed is higher than that of graph-based fingerprint matching21. Automatic classification can be also used as a preprocessing step for fingerprint matching, reducing matching time and complexity by narrowing the search space to a subset of a typically huge database 22.
A fingerprint matching system can be used for both verification and identification. In verification, the system compares an input fingerprint to the “enrolled” fingerprint of a specific user to determine if they are from the same finger (1:1 match). In identification, the system compares an input fingerprint with the prints of all enrolled users in the database to determine if the person is already known under a duplicate or false iden¬tity (1:N match)23.
6. Fingerprint Classification Method
Fingerprint Classification or Automatic fingerprint classification is required to reduce and narrow the search complexity by classifying fingerprints into different classes according to their relevant features. Generally, the advantage of classification is that it provides an indexing mechanism and facilities the matching process over large databases. There are many methods of fingerprint classification that will help reduce the search time and increase the capacity of the matching servers.
For a reliable and efficient representation of fingerprint images18, Discrete Fourier Transform method has been utilized which provides an efficient means for detecting directionality and periodicity in the frequency domain and removes noise by deleting high frequency coefficients, while16 is insistent on using Slap Fingerprint Segmentation which allows slaps or four fingers to be taken simultaneously onto a scanner, once captured slaps will be divided into four images of individual fingers in order to store them individually so it will take less time to search.
Other forms of fingerprint classification are Singular Point Detection14 and Orientation field image13, but in this paper, we recommend Sir Henry’s fingerprint classification approach22 who established the famous “Henry System” in 1899. This approach classifies fingerprints into 5 major classes depending on their characteristics, according to Henry’s classification scheme the five classes are Left Loop, Right Loop, Whorl, Arch and Tented Arch6. Every individual belongs to a specific fingerprint class depending on the features and characteristics of their fingerprints. Once the individuals are enrolled, if then they are classified into the specific class of fingerprint to which they belong the search problem is solved easily taking less time.
7. Problem Solution
As stated in the problem statement section, FBI, MOI and the U.S military are all working together as well as independently to collect fingerprints and demographics of individuals in Afghanistan. The problem that is creating a chaos is the non-existence of a harmony between the current systems. These systems are all independent and not associated with one another in a single way.
Afghanistan should work towards creating a National Automated Fingerprint Identification System (NAFIS) where the data held in any of the systems whether MoI, FBI or the U.S military should be linked with each other, because stand alone systems with no effective communication amongst each other will collapse.
A NAFIS will allow all the relevant authorities to access, securely use and perform a real-time “one-to-one” verification through their live fingerprints, identification of incapable people due to their mental illness, identification of criminals collected from the crime scenes.
As more agencies begin working together, the number of AFIS systems is likely to grow, which will lead to more effective, reliable, and an integrated system. This will save time, money and will lead us to a safe and a secure system that will reduce the security risks that we are currently facing.
Other solution is to start implementing Biometric capturing systems in all borders to capture demographics and biometrics of individuals and record them in the NAFIS and retrieve it when required, this will bring stability in place that will control travelers in and out of the country, and it is for their own safety as well for the security of the government.
8. Conclusion and Future Work
With the current rapid growth in multimedia technology, there is an imminent need for efficient techniques to search and query large image databases. Therefore, in this article, we have proposed an efficient method which classifies fingerprints into the standard classes. The basic idea of the proposed method is dividing fingerprint images into pre-specified classes to improve search efficiency and accuracy.
Currently, we are in the process of building a national electronic identification system for which we required to fingerprint and record details of all the population of Afghanistan in the next 7 years. In the feature extraction side we are planning to get the best out of every fingerprint possible. A good quality fingerprint will always lead us towards success. Often, poor quality images lower the system accuracy. Hence, an evaluation of image quality at the input stage to accept or reject an input is being considered. More complex and robust matching methods are also being researched.
 John D. Woodward, J., Orlans, N. M., & Higgins, P. T. (2003). Biometrics (pp. 1–463).
 Visiongain, 2008. Biometrics for Defense, [online] [Accessed 02 February 2013].
 Magnet, Shoshana & Amielle (2011) When Biometrics Fail: Gender, Race, and the Technology of Identity. Durham: Duke University Press.
 Bhanu, B., Tan, X., & Member, S. (2003). Fingerprint Indexing Based on Novel Features on Minutiae Triplets, 25(5), 616–622.
 Komarinski, P., (2005) Automated Fingerprint Identification Systems (AFIS). California: Elsevier Inc.
 E.R. Henry. Classification and Uses of Finger Prints. London: Routledge, 1900.
 Tariq, A., Akram, M. U., & Khan, S. A. (2011). An Automated System for Fingerprint Classification using Singular Points for Biometric Security, (December), 11–14.
 U.S. News, 2011. Biometric System working in Afghanistan, [online] [Accessed 22 January 2013].
 The FBI Federal Bureau of Investigation, 2011. Mission Afghanistan: Biometrics. [Online] [Accessed 09 February 2013].
 The New York Times, 2011. Afghanistan Has Bing Plans for Biometric Data, [online] [Accessed 13 February 2013].
 Al-Khouri, A. M., & J.Bal. (2007). Digital Identities and the Promise of the Technology Trio : PKI , Smart Cards , and Biometrics, 3(5), 361–367.
 Frost, A., Ayoub, R., & Rodriguez, C. (n.d.). A Best Practices Guide to Fingerprint Biometrics Ensuring a Successful Biometrics Implementation.
 N. K. Ratha, J. H. Connell, and R. M. Bolle, "Enhancing security and privacy in biometrics-based authentication systems," IBM systems Journal, vol. 40, pp. 614-634, 2001.
 S., R. Bhagtani and H. Chheda, 2005. “Biometrics: A further echelon of security”, Biological and Medical Physics, pp. 27-30 March 2005.
 Nextgov, 2012. Biometrics Blossom in Afghanistan, [online] [Accessed 12 February 2013].
 Ulery, B., Hicklin, A., Watson, C., Indovina, M., & Kwong, K. (2005). Slap Fingerprint Segmentation Evaluation 2004 Analysis Report Executive Summary, 1–8.
 R. Anwar, M. U. Akram, R. Arshad, “A Modified Singular Point Detection Algorithm”, International Conference on Image Analysis And Recognition (ICIAR), Springerlink, LNCS5112, pp. 905-914, Portugal, June 2008.
 Ý, C. H. P., & Þ, H. P. (n.d.). Fingerprint Classification using Fast Fourier Transform and Nonlinear Discriminant Analysis , 1–24.
 A.K. Jain, A. Ross, “Introduction to Biometrics”, A Handbook of Biometrics, Springer, pp.1-22, 2008.
 B. Miller, ”Vital Signs of Identity,” IEEE Spectrum, vol. 31, no. 2, pp. 22-30, Feb. 1994.
 Yang, S., & Verbauwhede, I. M. (2003). A Secure Fingerprint Matching Technique.
 H.C. Lee and R.E. Gaensslen. Advances in Fingerprint Technology. Elsevier, 1991.
 Jain, A. K., Faulds, H., Galton, F., & Henry, E. (2010). Biometrics: Fingerprint Matching, 36–44.
 Galton, S. F., Points, G., Points, G., Bureau, F., Fbi, T., & Bureau, N. (2006). Fingerprint Recognition, (August), 1–13.