Biometrics – Useful Resources The following resources contain additional information on Biometrics. Please use them to get more in-depth knowledge on this topic. Useful Links on Biometrics − This site details the basics of Biometrics − Useful Points of Biometrics Useful Books on Biometrics To enlist your site on this page, please drop an email to [email protected] Learning working make money
Discuss Biometrics This tutorial provides introductory knowledge on Biometrics. By accessing this tutorial, you would get sufficient information about the basics of biometrics and different biometric modalities such as physiological, behavioral, and combination of both modalities. This tutorial also provides a glimpse of various security issues related to biometric systems, and the comparison of various biometric systems. Learning working make money
Biometrics – Quick Guide Biometrics Overview The term Biometrics is composed of two words − Bio (Greek word for Life) and Metrics (Measurements). Biometrics is a branch of information technology that aims towards establishing one’s identity based on personal traits. Biometrics is presently a buzzword in the domain of information security as it provides high degree of accuracy in identifying an individual. What is Biometrics? Biometrics is a technology used to identify, analyze, and measure an individual’s physical and behavioral characteristics. Each human being is unique in terms of characteristics, which make him or her different from all others. The physical attributes such as finger prints, color of iris, color of hair, hand geometry, and behavioral characteristics such as tone and accent of speech, signature, or the way of typing keys of computer keyboard etc., make a person stand separate from the rest. This uniqueness of a person is then used by the biometric systems to − Identify and verify a person. Authenticate a person to give appropriate rights of system operations. Keep the system safe from unethical handling. What is a Biometric System? A biometric system is a technology which takes an individual’s physiological, behavioral, or both traits as input, analyzes it, and identifies the individual as a genuine or malicious user. Evolution of Biometrics The idea of biometrics was present since few years from now. In 14th century, China practiced taking finger prints of merchants and their children to separate them from all others. Fingerprinting is still used today. In the 19th century, an Anthropologist named Alphonse Bertillion developed a method (named Bertillionage) of taking body measurements of persons to identify them. He had realized that even if some features of human body are changed, such as length of hair, weight, etc., some physical traits of body remain unchanged, such as length of fingers. This method diminished quickly as it was found that the persons with same body measurements alone can be falsely taken as one. Subsequently, Richard Edward Henry from Scotland Yard developed a method for fingerprinting. The idea of retinal identification was conceived by Dr. Carleton Simon and Dr. Isadore Goldstein in 1935. In 1976, a research and development effort was put in at EyeDentify Inc. The first commercial retina scanning system was made available in 1981. Iris recognition was invented by John Daugman in 1993 at Cambridge University. In 2001, Biometrics Automated Toolset (BAT) was introduced in Kosovo, which provided a concrete identification means. Today, biometric has come up as an independent field of study with precise technologies of establishing personal identities. Why Biometrics is Required? With increasing use of Information Technology in the field of banking, science, medication, etc., there is an immense need to protect the systems and data from unauthorized users. Biometrics is used for authenticating and authorizing a person. Though these terms are often coupled; they mean different. Authentication (Identification) This process tries to find out answer of question, “Are you the same who you are claiming to be?”, or, “Do I know you?” This is one-to-many matching and comparison of a person’s biometrics with the whole database. Verification This is the one-to-one process of matching where live sample entered by the candidate is compared with a previously stored template in the database. If both are matching with more than 70% agreeable similarity, then the verification is successful. Authorization It is the process of assigning access rights to the authenticated or verified users. It tries to find out the answer for the question, “Are you eligible to have certain rights to access this resource?” Shortcomings of Conventional Security Aids The conventional methods of information system security used ID cards, passwords, Personal Identification Numbers (PINs), etc. They come with the following disadvantages − They all mean recognizing some code associated with the person rather than recognizing the person who actually produced it. They can be forgotten, lost, or stolen. They can be bypassed or easily compromised. They are not precise. In such cases, the security of the system is threatened. When the systems need high level of reliable protection, biometrics comes to help by binding the identity more oriented to individual. Basic Components of a Biometric System In general, a biometric system can be divided into four basic components. Let us see them briefly − Input Interface (Sensors) It is the sensing component of a biometrics system that converts human biological data into digital form. For example, A Metal Oxide Semiconductor (CMOS) imager or a Charge Coupled Device (CCD) in the case of face recognition, handprint recognition, or iris/retinal recognition systems. An optical sensor in case of fingerprint systems. A microphone in case of voice recognition systems. Processing Unit The processing component is a microprocessor, Digital Signal Processor (DSP), or computer that processes the data captured from the sensors. The processing of the biometric sample involves − Sample image enhancement Sample image normalization Feature extraction Comparison of the biometric sample with all stored samples in database. Database Store The database stores the enrolled sample, which is recalled to perform a match at the time of authentication. For identification, there can be any memory from Random Access Memory (RAM), flash EPROM, or a data server. For verification, a removable storage element like a contact or contactless smart card is used. Output Interface The output interface communicates the decision of the biometric system to enable the access to the user. This can be a simple serial communication protocol RS232, or the higher bandwidth USB protocol. It could also be TCP/IP protocol, Radio Frequency Identification (RFID), Bluetooth, or one of the many cellular protocols. General Working of a Biometric System There are four general steps a biometric system takes to perform identification and verification − Acquire live sample from candidate. (using sensors) Extract prominent features from sample. (using processing unit) Compare live sample with samples stored in database. (using algorithms) Present the decision. (Accept or reject the candidate.) The biometric sample is acquired from candidate user. The prominent features are extracted from the sample
Voice Recognition Voice recognition biometric modality is a combination of both physiological and behavioral modalities. Voice recognition is nothing but sound recognition. It relies on features influenced by − Physiological Component − Physical shape, size, and health of a person’s vocal cord, and lips, teeth, tongue, and mouth cavity. Behavioral Component − Emotional status of the person while speaking, accents, tone, pitch, pace of talking, mumbling, etc. Voice Recognition System Voice Recognition is also called Speaker Recognition. At the time of enrollment, the user needs to speak a word or phrase into a microphone. This is necessary to acquire speech sample of a candidate. The electrical signal from the microphone is converted into digital signal by an Analog to Digital (ADC) converter. It is recorded into the computer memory as a digitized sample. The computer then compares and attempts to match the input voice of candidate with the stored digitized voice sample and identifies the candidate. Voice Recognition Modalities There are two variants of voice recognition − speaker dependent and speaker independent. Speaker dependent voice recognition relies on the knowledge of candidate”s particular voice characteristics. This system learns those characteristics through voice training (or enrollment). The system needs to be trained on the users to accustom it to a particular accent and tone before employing to recognize what was said. It is a good option if there is only one user going to use the system. Speaker independent systems are able to recognize the speech from different users by restricting the contexts of the speech such as words and phrases. These systems are used for automated telephone interfaces. They do not require training the system on each individual user. They are a good choice to be used by different individuals where it is not required to recognize each candidate’s speech characteristics. Difference between Voice and Speech Recognition Speaker recognition and Speech recognition are mistakenly taken as same; but they are different technologies. Let us see, how − Speaker Recognition (Voice Recognition) Speech Recognition The objective of voice recognition is to recognize WHO is speaking. The speech recognition aims at understanding and comprehending WHAT was spoken. It is used to identify a person by analyzing its tone, voice pitch, and accent. It is used in hand-free computing, map, or menu navigation. Merits of Voice Recognition It is easy to implement. Demerits of Voice Recognition It is susceptible to quality of microphone and noise. The inability to control the factors affecting the input system can significantly decrease performance. Some speaker verification systems are also susceptible to spoofing attacks through recorded voice. Applications of Voice Recognition Performing telephone and internet transactions. Working with Interactive Voice Response (IRV)-based banking and health systems. Applying audio signatures for digital documents. In entertainment and emergency services. In online education systems. Learning working make money
Signal Processing and Biometrics There are various signals we can get in the real world such as sound, light, radio signals, biomedical signals from human body, etc. All these signals are in the form of a continuous stream of information, called analog signals. Human voice is a kind of signal we get from the real world and use as biometric input. What is a Signal? A signal is a measurable physical quantity containing some information, which can be conveyed, displayed, recorded, or modified. Signal Processing in Biometrics There are various reasons for processing signals. The biometric systems, require voice processing for various reasons − To extract meaningful information from the candidate’s sample. To remove noise from the sample. To make the sample transmittable. To remove distortion of sample. The analog signal processing module converts real world information such as sound wave in the form of 0s and 1s to make it understandable and usable by the contemporary digital systems such as biometric systems. The keystrokes, hand geometry, signature, and speech fall into the domains of signal processing and pattern recognition. Digital Signal Processing Systems (DSPs) There are two types of signals − analog and digital. The analog signals are uninterrupted, continuous stream of information whereas digital signal is a stream of 0s and 1s. DSP systems are one of the important components of biometric systems, which convert analog signals into a stream of discrete digital values by sampling and digitizing using an Analog-to-Digital Converter (ADC). DSPs are single-chip digital microcomputers, which process electrical signals generated by electronic sensors from cameras, fingerprint sensors, microphones, etc. DSP in Biometrics A DSP allows the biometric system to be small and easily portable, to perform efficiently and to be overall less costly. The DSP architecture is built to support complex mathematical algorithms that involve a significant amount of multiplication and addition. The DSP can execute multiply/add in a single cycle with the help of the multiply/accumulate (MAC) hardware inside its Arithmetic Logic Unit (ALU). It can also enhance the resolution of the captured image with the use of two-dimensional Fast Fourier Transforms (FFT) and finite IR filters. Learning working make money
Biometric Modality Selection To be able to select a proper biometric system, you need to compare them on various aspects. You need to assess the suitability of the systems to your requirements in terms of convenience, system specifications and performance, and your budget. You can select best suitable biometric system by studying various criteria for their effectiveness. Criteria for Effective Biometric System There are seven basic criteria for measuring effectiveness of a biometric system − Uniqueness − It determines how uniquely a biometric system can recognize a user from a group of users. It is a primary criterion. Universality − It indicates requirement for unique characteristics of each person in the world, which cannot be reproduced. It is a secondary criterion. Permanence − It indicates that a personal trait recorded needs to be constant in the database for a certain time period. Collectability − It is the ease at which a person’s trait can be acquired, measured, or processed further. Performance − It is the efficiency of system in terms of accuracy, speed, fault handling, and robustness. Acceptability − It is the user-friendliness, or how good the users accept the technology such that they are cooperative to let their biometric trait captured and assessed. Circumvention − It is the ease with which a trait is possibly imitated using an artifact or substitute. Comparison of Various Biometric Modalities Let us compare all the biometric system in the following terms − Biometric Characteristic Universality Uniqueness Permanence Collect-Ability Performance Accept-ability Circum-vention Finger Print Medium High High Medium High Medium High Face Recognition High Low Medium High Low High Low Hand Geometry Medium Medium Medium High Medium Medium Medium Iris Recognition High High High Medium High Low High Retinal Scan High High Medium Low High Low High DNA High High Medium High High Low Low Keystroke High Low Low High Medium High High Signature Low Low Low High Low High Low Voice Medium Low Low Medium Low High Low You can select an appropriate biometric system depending upon the criteria you need to deal with as shown in the table. Learning working make money
Pattern Recognition and Biometrics Pattern recognition deals with identifying a pattern and confirming it again. In general, a pattern can be a fingerprint image, a handwritten cursive word, a human face, a speech signal, a bar code, or a web page on the Internet. The individual patterns are often grouped into various categories based on their properties. When the patterns of same properties are grouped together, the resultant group is also a pattern, which is often called a pattern class. Pattern recognition is the science for observing, distinguishing the patterns of interest, and making correct decisions about the patterns or pattern classes. Thus, a biometric system applies pattern recognition to identify and classify the individuals, by comparing it with the stored templates. Pattern Recognition in Biometrics The pattern recognition technique conducts the following tasks − Classification − Identifying handwritten characters, CAPTCHAs, distinguishing humans from computers. Segmentation − Detecting text regions or face regions in images. Syntactic Pattern Recognition − Determining how a group of math symbols or operators are related, and how they form a meaningful expression. The following table highlights the role of pattern recognition in biometrics − Pattern Recognition Task Input Output Character Recognition (Signature Recognition) Optical signals or Strokes Name of the character Speaker Recognition Voice Identity of the speaker Fingerprint, Facial image, hand geometry image Image Identity of the user Components of Pattern Recognition Pattern recognition technique extracts a random pattern of human trait into a compact digital signature, which can serve as a biological identifier. The biometric systems use pattern recognition techniques to classify the users and identify them separately. The components of pattern recognition are as follows − Popular Algorithms in Pattern Recognition The most popular pattern generation algorithms are − Nearest Neighbor Algorithm You need to take the unknown individual’s vector and compute its distance from all the patterns in the database. The smallest distance gives the best match. Back-Propagation (Backprop) Algorithm It is a bit complex but very useful algorithm that involves a lot of mathematical computations. Learning working make money
Biometric System Performance Biometric system manufacturers claim high system performance which is practically difficult to achieve in actual operating environments. The possible reasons are, tests conducted in controlled environment setups, limitations on hardware, etc. For example, a voice recognition system can work efficiently only in quiet environment, a facial recognition system can work fine if lighting conditions are controlled, and candidates can be trained to clean and place their fingers properly on the fingerprint scanners. However, in practice, such ideal conditions may not be available in the target operating environment. Performance Measurements The performance measurements of a biometric system are closely tied to False Reject Rate (FRR) and False Accept Rate (FAR). FRR is also known as Type-I error or False Non Match Rate (FNMR) which states the likelihood of a legitimate user being rejected by the system. FAR is referred to as Type-II error or False Match Rate (FMR) which states the likelihood of a false identity claim being accepted by the system. An ideal biometric system is expected to produce zero value for both FAR and FRR. Means it should accept all genuine users and reject all fake identity claims, which is practically not achievable. FAR and FRR are inversely proportional to each other. If FAR is improved, then the FRR declines. A biometric system providing high FRR ensures high security. If the FRR is too high, then the system requires to enter the live sample a number of times, which makes it less efficient. The performance of current biometrics technologies is far from the ideal. Hence the system developers need to keep a good balance between these two factors depending on the security requirements. Learning working make money
Biometrics Modalities A biometric modality is nothing but a category of a biometric system depending upon the type of human trait it takes as input. The biometrics is largely statistical. The more the data available from sample, the more the system is likely to be unique and reliable. It can work on various modalities pertaining to measurements of individual’s body and features, and behavioral patterns. The modalities are classified based on the person’s biological traits. Types of Biometric Modalities There are various traits present in humans, which can be used as biometrics modalities. The biometric modalities fall under three types − 1. Physiological 2. Behavioral 3. Combination of physiological and behavioral modality The following table collects the points that differentiate these three modalities − Physiological Modality Behavioral Modality Combination of Both Modalities This modality pertains to the shape and size of the body. This modality is related to change in human behavior over time. This modality includes both traits, where the traits are depending upon physical as well as behavioral changes. For example − Fingerprint Recognition Hand Geometry Recognition system Facial Recognition System Iris Recognition System Hand Geometry Recognition System Retinal Scanning System DNA Recognition System For example − Gait (the way one walks) Rhythm of typing keys Signature For example − Voice Recognition It depends on health, size, and shape of vocal cord, nasal cavities, mouth cavity, shape of lips, etc., and the emotional status, age, illness (behavior) of a person. In the subsequent chapters, we will discuss each of these modalities in greater detail. Learning working make money
Multimodal Biometric Systems All the biometric systems we discussed till now were unimodal, which take single source of information for authentication. As the name depicts, multimodal biometric systems work on accepting information from two or more biometric inputs. A multimodal biometric system increases the scope and variety of input information the system takes from the users for authentication. Why Multimodal Biometrics is Required? The unimodal systems have to deal with various challenges such as lack of secrecy, non-universality of samples, extent of user’s comfort and freedom while dealing with the system, spoofing attacks on stored data, etc. Some of these challenges can be addressed by employing a multimodal biometric system. There are several more reasons for its requirement, such as − Availability of multiple traits makes the multimodal system more reliable. A multimodal biometric system increases security and secrecy of user data. A multimodal biometric system conducts fusion strategies to combine decisions from each subsystem and then comes up with a conclusion. This makes a multimodal system more accurate. If any of the identifiers fail to work for known or unknown reasons, the system still can provide security by employing the other identifier. Multimodal systems can provide knowledge about “liveliness” of the sample being entered by applying liveliness detection techniques. This makes them capable to detect and handle spoofing. Working of Multimodal Biometric System Multimodal biometric system has all the conventional modules a unimodal system has − Capturing module Feature extraction module Comparison module Decision making module In addition, it has a fusion technique to integrate the information from two different authentication systems. The fusion can be done at any of the following levels − During feature extraction. During comparison of live samples with stored biometric templates. During decision making. The multimodal biometric systems that integrate or fuse the information at initial stage are considered to be more effective than the systems those integrate the information at the later stages. The obvious reason to this is, the early stage contains more accurate information than the matching scores of the comparison modules. Fusion Scenarios in Multimodal Biometric System Within a multimodal biometric system, there can be variety in number of traits and components. They can be as follows − Single biometric trait, multiple sensors. Single biometric trait, multiple classifiers (say, minutiae-based matcher and texture-based matcher). Single biometric trait, multiple units (say, multiple fingers). Multiple biometric traits of an individual (say, iris, fingerprint, etc.). These traits are then operated upon to confirm user’s identity. Design Issues with Multimodal Biometric Systems You need to consider a number of factors while designing a multimodal biometric system − Level of security you need to bring in. The number of users who will use the system. Types of biometric traits you need to acquire. The number of biometric traits from the users. The level at which multiple biometric traits need integration. The technique to be adopted to integrate the information. The trade-off between development cost versus system performance. Learning working make money