An introduction to biometric recognition
Summary (4 min read)
Introduction
- The characteristic should be sufficiently invariant (with respect to the matching criterion) over a period of time, also known as Permanence.
- A practical biometric system should meet the specified recognition accuracy, speed, and resource requirements, be harmless to the users, be accepted by the intended population, and be sufficiently robust to various fraudulent methods and attacks to the system.
II. BIOMETRIC SYSTEMS
- A biometric system is essentially a pattern recognition system that operates by acquiring biometric data from an individual, extracting a feature set from the acquired data, and comparing this feature set against the template set in the database.
- In the verification mode, the system validates a person’s identity by comparing the captured biometric data with her own biometric template(s) stored in the system database.
- Identity verification is typically used for positive recognition, where the aim is to prevent multiple people from using the same identity [26].
- Thus if otherwise where is the function that measures the similarity between feature vectors and , and is a predefined threshold.
- In the matching module of a fingerprint-based biometric system, the number of matching minutiae between the input and the template fingerprint images is determined and a matching score is reported.
III. BIOMETRIC SYSTEM ERRORS
- These two types of errors are often termed as false accept and false reject, respectively.
- There is a tradeoff between false match rate (FMR) and false nonmatch rate (FNMR) in every biometric system.
- Also, this expression is only an approximation since it does not consider the probability of falsely matching an incorrect template before the right one is retrieved [28].
- In applications like bank ATM card verification, a false match means a loss of several hundred dollars while a high FNMR may lead to a potential loss of a valued customer.
IV. COMPARISON OF VARIOUS BIOMETRICS
- A number of biometric characteristics exist and are in use in various applications (see Fig. 3).
- Commercial hand geometry-based verification systems have been installed in hundreds of locations around the world.
- The complex iris texture carries very distinctive information useful for personal recognition.
- Signatures are a behavioral biometric that change over a period of time and are influenced by physical and emotional conditions of the signatories.
- The applicability of a specific biometric technique depends heavily on the requirements of the application domain.
V. APPLICATIONS OF BIOMETRIC SYSTEMS
- The applications of biometrics can be divided into the following three main groups.
- HIGH, MEDIUM, AND LOW ARE DENOTED BY H, M, AND L, RESPECTIVELY access, ATM, credit card, physical access control, cellular phone, PDA, medical records management, and distance learning.
- Government applications such as national ID card, correctional facility, driver’s license, social security, welfaredisbursement, border control, and passport control.
- The Schiphol Privium scheme at the Amsterdam airport, for example, employs iris scan cards to speed up the passport and visa control procedures [4].
- Thus, biometric systems can be used to enhance user convenience while improving security.
A. Positive Recognition in Commercial Applications
- The traditional technologies available to achieve a positive recognition include knowledge-based methods (e.g., PINs and passwords) and token-based methods (e.g., keys and cards).
- Strong passwords are difficult to remember and result in more help desk calls for forgotten or expired passwords.
- Further, a hacker needs to break only one password among all the employees to gain access to a company’s Intranet and thus, a single weak password compromises the overall security of every system that the user has access to.
- Biometrics introduces incredible convenience for the users (as users are no longer required to remember multiple, long and complex frequently changing passwords) while maintaining a sufficiently high degree of security.
- Forrester Research states that the average help desk labor cost for a single password reset is about US $38.
B. Negative Recognition in Government and Forensic Applications
- In negative recognition applications such as employee background checking and preventing terrorists from boarding airplanes, the personal recognition is required to be performed in the identification mode.
- While the identification will still be 1%, the identification will be .
- While biometric systems may not yet be extremely accurate to support large-scale identification applications, they are the only choice for negative recognition applications.
- The authors need to understand that, in such semi-automatic applications, the biometric system only generates an alarm that calls for a closer examination of the individual and an alarm does not directly translate into catching a terrorist.
- Appropriate legislation is required to protect the abuse of such systems.
VII. LIMITATIONS OF (UNIMODAL) BIOMETRIC SYSTEMS
- The successful installation of biometric systems in various civilian applications does not imply that biometrics is a fully solved problem.
- Noisy data could also be the result of defective or improperly maintained sensors (e.g., accumulation of dirt on a fingerprint sensor) or unfavorable ambient conditions (e.g., poor illumination of a user’s face in a face recognition system).
- As another example, the varying psychological makeup of an individual might result in vastly different behavioral traits at various time instances.
- This limitation restricts the discriminability provided by the biometric trait.
- Golfarelli et al. [29] have shown that the information content (number of distinguishable patterns) in two of the most commonly used representations of hand geometry and face are only of the order of and , respectively.
VIII. MULTIMODAL BIOMETRIC SYSTEMS
- Such systems, known as multimodal biometric systems [12], are expected to be more reliable due to the presence of multiple, independent pieces of evidence [14].
- These systems are also able to meet the stringent performance requirements imposed by various applications [13].
- Multimodal biometric systems address the problem of nonuniversality, since multiple traits ensure sufficient population coverage.
- Further, multimodal biometric systems provide antispoofing measures by making it difficult for an intruder to simultaneously spoof the multiple biometric traits of a legitimate user.
- By asking the user to present a random subset of biometric traits (e.g., right index and right middle fingers, in that order), the system ensures that a “live” user is indeed present at the point of data acquisition.
A. Modes of Operation
- A multimodal biometric system can operate in one of three different modes: serial mode, parallel mode, or hierarchical mode.
- In the serial mode of operation, the output of one biometric trait is typically used to narrow down the number of possible identities before the next trait is used.
- Further, a decision could be arrived at without acquiring all the traits.
- In the hierarchical scheme, individual classifiers are combined in a treelike structure.
B. Levels of Fusion
- Multimodal biometric systems integrate information presented by multiple biometric indicators.
- The data obtained from each biometric modality is used to compute a feature vector.
- Techniques such as weighted averaging may be used to combine the matching scores reported by the multiple matchers.
- An integration at the feature level typically results in a better improvement than at the matching score level.
- It is more difficult to perform a combination at the feature level because the relationship between the feature spaces of different biometric systems may not be known and the feature representations may not be compatible.
C. What to Integrate?
- Multimodal biometric systems can be designed to operate in one of the following five scenarios (see Fig. 9).
- The information obtained from different sensors for the same biometric are combined, also known as 1) Multiple sensors.
- Second, an identification system may use such a combination scheme for indexing.
- In their opinion, scenarios 4 and 5 combine strongly correlated measurements and are expected to result in a smaller improvement in recognition accuracy than scenarios 2 and 3, but they are more cost effective than scenario 2 and more convenient than scenario 3.
- As a result, the overall response time of the system is limited by the slowest individual feature extractor and/or matcher.
D. Examples of Multimodal Biometric Systems
- Multimodal biometric systems have received much attention in recent literature.
- Brunelli et al. [16] describe a multimodal biometric system that uses the face and voice traits of an individual for identification.
- Kumar et al. combined hand geometry and palmprint biometrics in a verification system [33].
- General strategies for combining multiple classifiers have been suggested in [19] and [20].
- On the other hand, such a combination requires the users to provide multiple identity cues, which may cause inconvenience.
IX. SOCIAL ACCEPTANCE AND PRIVACY ISSUES
- Human factors dictate the success of a biometric-based identification system to a large extent.
- More importantly, people fear that biometric identifiers could be used for linking personal information across different systems or databases.
- On the positive side, biometrics can be used as one of the most effective means for protecting individual privacy.
- Legislation is necessary to ensure that such information remains private and that its misuse is appropriately punished.
- First, the actual physical characteristic cannot be recovered from the digital template thus ensuring privacy.
X. SUMMARY
- Reliable personal recognition is critical to many business processes.
- It is thus obvious that any system assuring reliable personal recognition must necessarily involve a biometric component.
- This is not, however, to state that biometrics alone can deliver reliable personal recognition component.
- The security level of a system depends on the requirements (threat model) of an application and the cost-benefit analysis.
- As biometric technology matures, there will be an increasing interaction among the market, technology, and the applications.
Did you find this useful? Give us your feedback
Citations
2,021 citations
Cites background from "An introduction to biometric recogn..."
...Multimodal biometric systems consolidate the evidence presented by multiple biometric sources and typically provide better recognition performance compared to systems based on a single biometric modality....
[...]
...(1) is zero is not valid in most practical biometric systems....
[...]
1,067 citations
Cites background from "An introduction to biometric recogn..."
...Biometric authentication or, simply biometrics [5], [6], [69], refers to establishing identity based on the physical and behavioral characteristics (also...
[...]
705 citations
695 citations
Cites background from "An introduction to biometric recogn..."
...to either validate or determine an identity [1]....
[...]
679 citations
References
5,670 citations
"An introduction to biometric recogn..." refers background in this paper
...General strategies for combining multiple classifiers have been suggested in [19] and [20]....
[...]
3,821 citations
"An introduction to biometric recogn..." refers background in this paper
...1 Some portions of this article have previously appeared in [1] and [2] ©IEEE, ©Springer....
[...]
1,845 citations
1,703 citations
"An introduction to biometric recogn..." refers background or methods in this paper
...All the approaches presented in [19] (the highest rank method, the Borda count method and logistic regression) attempt to reduce or re-rank a given set of classes....
[...]
...General strategies for combining multiple classifiers have been suggested in [19] and [20]....
[...]
974 citations