The rise of biometrics as a security system for the next generation
ALTHOUGH THE idea of using the human body as a security system might still seem far from reality, it has recently emerged as the most promising technological innovation. With the rapid development of Information and Communication Technology (ICT) in the past few years, biometrics is being hailed as potentially the most suitable security system for the next generation. Biometrics, which is a pattern recognition system that extracts physiological or behavioral characteristics from the human body, has garnered such acclaim because it is a relatively safe and convenient way for user authentication. Every individual has his or her own unique physical characteristics that can be neither duplicated nor modified. Once these characteristics are recorded on the system, the user can access the stored information conveniently without worrying about losing keys, forgetting passwords or getting them hacked. The biggest problem of the conventional password system – the loss or leakage of private information – can be addressed by adapting biometrics to security systems.
Simple steps, complex math: how biometrics works
The process of identifying the user is divided into three steps: enrollment, storage and comparison. First is to enroll the name and the biological traits of the user. In this step, a sensor is used to detect distinct behavioral or physical characteristics of a person. In the second step, a computer stores these detected traits, which are then analyzed and converted into a recognizable data, such as codes or graphs. For further convenience, there are also more advanced systems that record this data on a portable smart card so that the user can carry it with him or her. The last step is comparison. When the user accesses the system later on, the software compares the user’s biological characteristics to the stored information. If the two correspond, then the system finally allows the user to access it.
Although these three steps seem quite simple, a great deal of math and technology are used to encode, store and identify the users’ physical characteristics. Take fingerprints for example. With only the naked eyes, people cannot figure out exactly how similar fingerprints are. However, a math algorithm in the system precisely differentiates between the authentic user of the system and any unregistered users who might attempt to access the system. For example, the algorithm can draw out a precise picture of the fingerprints with geometry and accurately compare the immediate user’s fingerprints with the stored fingerprints to make an accurate judgment by using probability. In order to encode this information into legible data for further usage, the system represents a fingerprint’s patterns and characteristics as sequences by using the number theory. All these principles can be applied to biometrics of other human body parts beyond fingerprints, including eyes and faces.
Various types of biometrics
Although finger scans are most common in biometrics, the world of biometrics is bigger than it appears to be. It is very surprising that many parts of our body, from fingerprints, to veins and even, incredible as it may seem, one’s private parts, can be used within a security system. Among the diverse types of biometrics that have been recently developed, fingerprint and iris scans are the most commonly used.
A fingerprint scan is currently the singular most popular type of biometrics. Fingerprints, which are the tiny friction ridges on the ends of our fingers, are formed in the womb before people are born. Fingerprints of all people, even those of twins, are different because they are randomly created under the control of each individual’s distinct DNA code. Fingerprints rarely change because they are created in the corium, which is the permanent skin layer beneath the external skin. However, there is still a drawback. Because it is the most frequently used human body part, people who have a faint fingerprint due to hard labor or those who easily sweat might have trouble using it to access a security system.
Turing to the eyes, the iris scan is using the iris pattern that is distinct from each individual. It is a much more developed security system compared to fingerprint or retinal scans in terms of its security, accuracy, convenience, and speed. The iris pattern is formed within one to two years after birth, and it can be identified in a few seconds by converting iris wrinkles into the form of frequency. Unlike fingerprint scans, people who use iris scans do not have to directly touch the system device because the iris can be identified from a distance of 8 to 25 centimeters. However, there are still serious weaknesses to iris biometrics. Although an iris itself is an ideal body part for biometrics because it is permanent, consistent and distinctive, current technology is not advanced enough to make an accurate authentication. For example, the eyelid and eyelash that partly cover the iris might lead to an incorrect judgment on the part of the system. There should be an algorithm that detects any extraneous body parts and excludes them to prevent false recognition. In addition, the current iris identification system is easily deluded by the high-definition photo, which can be printed on the contact lens. A technology that can reject fake photos should be further installed.
Unlock your bank account with your body – Fintech
Although biometrics techniques are being used in a wide variety of industries, they are hailed the most by the finance industry. With the rise of Fintech*, which is an innovative financial service that actively utilizes ICT technology, security issues regarding user identification and privacy protection have become increasingly important. This is because internet or smartphone banking, typical examples of Fintech, are highly susceptible to phishing scams. The rise of biometrics is also a part of this trend. It has been widely acclaimed as the best security system for internet-based transaction platforms, especially those involving smartphones. Fingerprints are most generally used for Fintech in other fields as well. However, numerous companies and developers are constantly releasingnew products that utilize different body traits.
Among new technologies being introduced to the Fintech market, a face recognition system that was introduced in March 2015 is best known. Alibaba, the world’s biggest e-commerce company based in China, came up with *Smile To Pay*. This face recognition S Using facial recognition is a groundbreaking strategy in the mobile payment platform, because its rivals, Samsung and Apple, rely on conventional fingerprint scanners in mobile payment services. Given the fingerprint’s susceptibility to spoofing and Alibaba’s predominance in Chinese market, facial recognition is a very promising technology. After months of testing, facial recognition became available as a login option on December 2015 for the Apple iOS 9 and certain android smartphones. It is expected to expand soon to other phones and other Alibaba services.
* * *
Both convenient and safe, biometrics has become irrefutably one of the most powerful security systems that will lead future technology. Although it is only in its nascent stage, a considerable amount of relevant research and developmen has already emerged. The biometrics market is expected to expand to worldwide to a 17 trillion dollar business in 2019, which is almost twice the market size as of 2016. In order to take a leadership role in this future industry, Korea should invest heavily in biometrics research.
* Fintech: A neologism that combines Finance and Technology
Jeong Da-eun
clairejeong96@yonsei.ac.kr