A SEMINAR REPORT ON BIO-METRICS TECHNOLOGY

(Worldwide interoperability for Bio-metrics)

available at www.mindstien.net

Contents:1. Brief on Security System 2. Introduction to Bio-metrics 3. A Few Definition 4. History of Bio-metrics 5. International Bio-metrics society 5.1 Purpose 5.2 Journal 5.3 Meetings 5.3 Biometrics Bulletin 6. Generic model of bio-metrics 6.1 Data Acquisition:6.2 Transmission: 6.3 Signal Processing:6.4 Decision Making:6.5 Data Storage:7. Phases of Bio-metrics 7.1 Input 7.2 Process 7.3 Output 8. Bio-metrics Identification 8.1 Voice Print 8.2 Finger Print 8.3 Iris Pattern 8.4 Face Print 8.5 Hand Geometry 8.6 Retinal Scan 8.7 Keystroke Dynamics 8.8 Facial Recognition 8.9 Signature Recognition 9. Captured Using Web Cam and Microphone

page-no:1 2 3 4 5

10

19
2

available at www.mindstien.net

10. Comparison of Bio-metrics technique 10.1 Uniqueness of biometric features 11. How well do Bio-metrics works? 11.1 False Accept Rates 11.2 False Reject Rates 11.3 Equal Error Rates 11.4 Validity of Test Data 12. Is DNA Bio-metrics ?

20

23

25

13. About Bio-metrics 26 13.1 Will Bio-metrics solve all security concerns? 13.2 Which is the best identification technique? 13.3 Is Bio-metrics more securing then password? 13.4 Applications 13.5 New Opportunities 13.6 Future Technology 13.7 Benefits 14. Conclusion 15. Reference 30 31

available at www.mindstien.net

* WHAT SECURITY?* & WHY SECURITY?*

From the beginning of the human culture, people are always trying to secure their properties. Because some people are always eager to snatch others properties. Security refers to prohibit some unauthorized persons from some important data or from some precious assets. To restrict this kind of theft human beings are making special systems that can prevent the thieves such systems are known as security systems. We can consider the starting of these kinds of systems through lock and key. A Frenchman Alexander Fish has made a strong, fire proof safe in 1844. And Linas Yel an American, has invented a Pin lock kind of lock. Both these equipments are used to provide security to money, important documents and jewellary. So this is the first step in the world of security system. At that point of time this arrangement for security was considered the best. But as time passes it was found that any unauthorized person could access these through duplicate keys or using any technique. So as the world progresses, the thieves become more intelligent to break the security. Thus, we always need a new and strong security system to protect our assets. So, security system is ever developing field. Today in the world of e-commerce and Internet this thing is becoming more and more important because today the world is like village, from anywhere, any time, any body can harm you. e.g. :-If you are purchasing something from web and because of lack of security someone come to know your bank account number. Then that person will have all the information of your bank account, which can perhaps put you in a big trouble!

Now a days, there are many security systems are available. But we shall concentrate on BIO-METRICS.

available at www.mindstien.net

*INTRODUCTION*
BIO=Pertaining to biology; MATRICS=Science and art of measurement. The term 'biometrics' is used to refer to any and all of a variety of identification techniques which are based on some physical and difficultto-alienate characteristic. Today, the science of biometric technology refers to the "automated" methods used to recognize a person based upon physiological or behavioral characteristics. Biometric technologies are becoming the foundation of an extensive array of highly secure identification and personal verification solutions.

Biometrics is a modern technological field that focuses on identifying an individual through his or her unique physical traits. A biometric is a measurable, physical characteristic or personal behavioral trait used to recognize the identity or verify the claimed identity of an enrolled user. Biometrics is automated methods of recognizing a person based on a physiological or behavioral characteristic.

Physiological techniques include fingerprint recognition, retinal and iris scanning, facial recognition, hand and finger geometry and DNA analysis. Behavioral techniques include handwriting recognition, voice or speech recognition, gait, and keystroke dynamics. In all automated systems, the fundamental operational steps are: 1. Capture: The biometric data is captured, digitized and entered into a database. 2. Extraction: A template is created using this measurable unique data. 3. Comparison: The template is compared with a new sample. 4. Match/Non-Match: The existing template matches the new sample or it does not. The goal of most automated biometric ID systems is one of two outcomes: 1. Verification: Is the person who the they claim to be? 2. Identification or recognition: Who is this? Is the person already known to the system under a different identity?

Only biometrics can identify you as you [Not Password]. 5

available at www.mindstien.net

*A FEW DEFINATIONS*
Biometrics Authentication
Person authentication or simply authentication is a task of confirming or denying a persons claimed identity. Biometrics is a measurement using a persons behavioural or physiological characteristics. Examples of these characteristics are fingerprints, face, voice and signature. Biometric authentication is a task of confirming or denying a persons claimed identity using his behavioural or physiological characteristics. Biometrics, Biometric or Biometry Biometry refers to a much broader field involving application from statistics to biology and medicine. Biometrics is a noun and biometric is an adjective. Note the distinction between the terms biometrics and biometry. Biometry refers to a much broader field involving application from statistics to biology and medicine. Biometrics is a noun and biometric is an adjective. Authentication versus Identification According to [Jain, 1999], associating an identity with an individual is called personal identification. This problem can be categorised into: (i) authentication and (ii) identification. Authentication refers to the problem of confirming or denying a persons claimed identity while identification refers to the problem of establishing a subjects identity. The differences between identification and authentication can be summarised in Table A.1. Table A.1 Identification versus authentication Identification It determines the identity of the person. Authentication

It determines whether the person is indeed who he claims to be. There is no identity claim from the user. The user makes an identity claim. By using The system has to search through its record the claim as a key, the system searches the to find the identity. This search is called a users record to compare. It is a one to one many-to-one mapping. The cost of mapping. The system uses the identity as a computation in the identification task key to call for the identitys record. The increases in proportion to the number of cost of computation is not dependant on record of users. the number of record of users. It is often assumed that a captured The captured biometric signature comes

available at www.mindstien.net

biometric signature comes from a set of known biometric feature stored in the system. This is referred to as closed-set identification.

from a large set of unknown users. This is referred to as open-set verification.

HISTORY OF BIO-METRICS*
Francis Gal ton is one of the founders of Bio-metrics. In 1892, Gal ton Invented the first system of fingerprinting. He has observed that No two persons have same finger print. Each and every person has got a unique finger print pattern. In 1968, Bio-metrics technique was implemented successfully when wellknown bank of New York has arranged the finger print scanning to access the currency Walt , first time in the world. But after 1990, when IT (Information Technology) comes into the picture, through its advancement and simplicity Bio-metrics technique becomes very popular. This new technology is very easy to understand compare to its heavy label. Because of user-friendly environment this technique is very well welcomed by various fields.

*THE INTERNATIONAL BIO-METRICS SOCIETY*

The International Biometric Society, founded in 1947, is an international organization for the advancement of the subject-matter sciences through the development of quantitative the orgies and the development, application, and dissemination of effective mathematical and statistical techniques.

5.1. Purpose
To this end, it welcomes to membership biologists, statisticians, and others interested in applying statistical techniques to research data. The International Biometric Society and its journal, Biometrics, serve as an invaluable means of communication between the subject-matter specialists and the statisticians.

available at www.mindstien.net

5.2. Journal
Biometrics is published quarterly. Its general objectives are to promote and extend the use of mathematical and statistical methods in the various subject-matter disciplines, by describing and exemplifying developments in these methods and their application in a form readily assimilable by experimenters and those concerned primarily with analysis of data. The journal is a ready medium for publication of papers by both the experimentalist and the statistician. The papers in the journal include statistical, authoritative expository or review articles, and analytical or methodological papers contributing to the planning or analysis of experiments and surveys, or the interpretation of data. Many of the papers in Biometrics contain actual worked examples of the statistical analyses proposed.

5.3. Meeting
Encouragements is given the subject-matter experimenter by holding symposia dealing with applied statistics as a part of annual meetings of the biological, chemical, engineering, and physical science societies. Such symposia afford an opportunity for subject-matter specialists and statisticians to discuss problems of mutual interest. At Regional meetings of the Society, a member is afforded an opportunity to present technical papers. This permits the exchange of ideas and information through formal presentation and informal contacts with fellow statisticians and experimentalists.

5.4. Biometric Bulletin

The International Biometric Society also publishes an informal newsletter containing news about membership activities, letters to the editor, and membership and editorial commentaries. It is a means for expressing ideas on a wide range of statistical and biometrical matters and is also published quarterly.

available at www.mindstien.net

*GENERIC MODEL OF BIO-METRICS*

6.1 Data Acquisition:Data collection involves use of sensors to detect & measure an individuals physiological or behavioral characteristics. The biometric feature must have the following characteristics:(a) Universality, which means that every person should have the characteristic, (b) Uniqueness, two persons should not have the same term or measurement of Characteristic, (c) Permanence, the characteristic should be invariant with time, (d) Measurability, the characteristic can be quantified that is the origin of the Cameras used in biometric systems are generally either CCD (charge couple device) or CMOS (combined metal oxide semiconductor) image sensors.CCD is comparatively more costly than CMOS. The cost of CCD is nearly 2500 Rs. And CMOS is about 800 to 1000 Rs. Price depends on the resolution which reflects the accuracy of biometric system.

available at www.mindstien.net

6.2 Transmission: Not all the biometric systems process and store data on the measuring device. Often measurement is made using relatively simple device to a computer or server for processing and/or storage. Depending on the system, the data may be relatively large and thus would need to be compressed for quick transfer. The compression algorithm needs to be selected carefully; otherwise it may introduce some artefacts that could impact the decision process. In any image scanning Biometric system, JPEG compression is preferred due to the block ness it produces at high compression ratios. The data can also be transmitted to the database for storage as raw data.

6.3 Signal Processing:The signal processing unit uses features extraction algorithms to extract true biometric information from the sample in presence of noise introduced during data collection and transmission. Additional measurements are made if any flaw or corruption is noted, to ensure good quality. Pattern matching involves comparing the feature sample to a stored sample. (The biometric data can be stored locally on the biometric data can be stored locally on the biometric device, some central database/server, or on a smart card issued to users.) The result of the comparison is sent to the decision system to determine the match.

6.4 Decision Making:The final step is the decision to accept or reject user, and is based on a security threshold. The threshold value is either a parameter of comparison process itself, or the system compares the resulting match value with the threshold value. If for example, in a system performing identity verification, the match value is equal to or higher than the threshold value, the user is accepted. In an identification system, acceptance might require a match value that is both higher than threshold value and higher than the second-best match by a specific amount.

6.5 Data Storage:After extracting the biometric features, the system stores and maintains the new master template. Choosing proper discriminating characteristics for categorizing records can facilitate future searches. The system stores the templates in one of four locations: a card, a servers central database, a workstation, or an authentication terminal. If privacy is a concern, a card is the only choice because sensitive biometric data should not be stored (and potentially misused) in a central database.

available at www.mindstien.net

10

* PHASE OF BIO-METRICS*
7.1 Input

Using appropriate input device it will collect the data of the persons biological features. These features are known as the master key for the computer. Let us put in a simple manner.

For recording and converting biometric traits to usable computer data, one needs an appropriate sensor (see table). Of course, costs can greatly vary for different sensors. However, we can't forget that many technical devices already have sensors built in, and therefore, offer possibilities to measure biometric features nearly free of cost. Biometric Trait Fingerprint (Minutia) Signature (dynamic) Facial Structure Iris pattern Retina Hand geometry Finger geometry Vein structure of the back of the Hand Ear form Voice (Timbre) DNA Odor Keyboard Strokes Comparison: Password Sensor capacitive, optic, thermal, acoustic, pressure sensitive Tablet Camera Camera Camera Camera Camera Camera Camera Microphone Chemical Lab Chemical sensors Keyboard Keyboard

7.2 Process

Processes the input data. Here, computer takes the measurement of inputted Biological features.

available at www.mindstien.net

11

No two persons have same finger print. So, computer joins the points of that finger print features and make the pattern of it and it will store that pattern into its memory. Then computer compares this pattern against the master keys. e.g.: - Putting the tracing paper on Indian map, if we join Banglor, Chennai, Delhi, Hyderabad, Mumbai, Nagpur, Ahmedabad, Jaipur and Lukhnow drawing a line then by chance the original map is lost. Now if we want to know the distance between Ahmedabad and Bhavnagar for that we require the original one, which is drawn with the same measurement. But In this case, actual map is lost so we are helpless. Similarly, the finger print used by the computer to make its pattern is deliberately hidden. Thus, the security system does not allow any unauthorized person.

7.3 Output

Testing of processed data and give the access rights to the person. Here, When any person scanned his/her finger print , computer will Store a Bio-logical featured pattern and compares it against the stored masterKey in its memory. And if it is matched, then permission granted else not.

BIO-METRICS IDENTIFICATION
CONVENTIONAL METHODS OF BIOMETRIC IDENTIFICATION ARE. 1 VOICE PRINT 2 FINGER PRINT 3 IRIS PATTERN 4 FACE PRINT 5 HAND GEOMETRY 6 RETINAL SCAN 7 KEYSTROKE DYNAMICS 8 FACIAL RECOGNITION 9 SIGNATURE RECOGNITION

8.1 VOICE PRINT:

In this world of bio-metrics the first invention is of voice print. Voice recognition is very different to speech recognition. Speech recognition detect words and sentences from an

available at www.mindstien.net

12

incoming audio signal whereas voice recognition tries to detect the speaker given a voice sample.

Here voice is recorded by microphones. But as each & every person has got different voice pattern thats why no two voice prints are matched. Voice print is determined by many different factors: size of the vocal cavities (throat, oral, nasal) as well as the characteristics of the vocal chords themselves. Your voice is further modified by the way you speak - the way your mouth, lips, tongue, jaw and teeth move (these are called articulators). Therefore, the chances of two people have the same vocal characteristics are minute. But this technique is not 100% authentic because of the following reasons. Each and every persons voice changes as per the mood. All microphones, which are used to fetch the voice, may not work properly. During the recording of voice if there is any background sound or noise then it will not be able to give perfect result.

The computer converts the voice pattern into 0 & 1.After taking the input that input is also converted into 0 & 1.If the both code matches the person is authorized.

8.2 FINGER PRINT:

In this technique, no two persons have the same

available at www.mindstien.net

13

Finger print, Francis Galton proves this after 1982

a long observation in

No ID card is as best as finger print because the possibility of matching two persons finger print is 0%. Finger print is also called finger scanning. To scan the finger print, a special kind of scanner that can work with optical rays is used. When a person put his/her finger on the transparent glass of scanner, CCD chip scans the biological features of finger. It fetches the diagram of finger print & delivers it to the computer where a special programme measures the distance between the special features & that print. Here, those special features are core, split, island, end & delta. The new sophisticated software of bio metrics scans the finger as well as palm. In this case, it takes the 90 types of measurement of palm & matches that new digit against the stored digits in the computers memory as well as finger print is matched & then it gives the access rights. Hand technology is similar but surprisingly uses smaller amount information. Hand technology is also susceptible to injuries to hand and can be fairly expensive to install. Most fingerprint data takes up around 250Kb. Fingerprinting is a widely used and accepted technology with very good accuracy rates. Thus this technique is more authentic than the voice print. The computer converts the finger pattern into 0 & 1.After taking the input that input is also converted into 0 & 1.If the both code matches the person is authorized.

available at www.mindstien.net

14

8.3 PATTERN OF IRIS:

an iris has a mesh-like texture to it, with numerous overlays and patterns. The iris is located behind the cornea of the eye, but in front of the lens. Its only physiological purpose is to control the amount of light that enters the eye through the pupil, but its construction from elastic connective tissue gives it a complex, fibrillous pattern.

Iris recognition is the epitome of biometric identification - the

Entire planet could be enrolled into an iris database and there would still be a minute chance of false identification or rejection. In finger print technique there are only 90 special features, but in iris pattern there are 266 special features. In 1998, Dr. John Daugman. -inventor of Cambridge University of Britain has compared the approx 3, 00, 00,000 the iris but he couldnt find at least one similarity.

available at www.mindstien.net

15

In this technique CCD camera, which looks like small camcorders, scans the eyes iris pattern. The iris pattern of a persons left and right eyes are different. Here, CCD scan either left or right eyes iris pattern. After scanning that pattern is delivered to the computers processing unit. Computer only considers the data of iris and removes other data and analyzes the total 266 features and assigns a particular value to each feature. The pattern of iris is converted into the total 4,096 bits of 0 & 1. Then computer compares the new data of 0 & 1 to the iris print stored as master key and if it is matches then gives the access rights. Iris has stable and distinctive features for personal identification. That is because every iris has fine and unique patterns and does not change over time since two or three years after the birth, so it might be called as a kind of optical finger print The Iris Code is so specialized that the chances of two irises having the same code are 1 in 1052 !

SOME CURRENT AND FUTURE APPLICATIONS OF IRIS RECOGNITION:

Secure accesses to bank cash machine accounts: The banks of United, Diebold and Sensor have applied it. After enrolling once (a 30second process), the customers need only approach the ATM, follow the instruction to look at the camera, and be recognized within 2-4 seconds. The ultimate aim is to provide safe and secure transactions. Ticket less, document-free air travel: Passengers and airline employees will store digital images of their irises on a database. After the image of your iris is on the file, a video camera will be able to instantly verify your identity and clear you to board the aircraft.

available at www.mindstien.net

16

Computer login: the iris an living password. National border controls: the iris as a living passport. Premises access control (homes, office, and laboratory). Credit card authentication. Secure financial transactions. Internet security.

8.4 FACE PRINT:

Facial biometrics use various features of the face to recognize or verify a user. Fourth type of biometrics identification is of face print. To capture face print of any person special kind of camera is used. This vide camera delivers the capture image to compute Computer makes a pattern of that image using software. To make this pattern it will use approx 50 features of face like the distance between two eyes, breath of nose, cheeks, area of forehead, jaws etc. These features are converted into digital form. Though it is not necessary that a high tech and sophisticated software identify proper person. The computer converts the face pattern into 0 & 1.After taking the input that input is also converted into 0 & 1.If the both code matches the person is authorized. Here are some possibilities. To create a pattern of a persons face, the face of the person must be within the range of 35 degrees. A camera cannot recognize a person wearing goggles. e.g.: During a testing at Palm Beach, America, the camera has recognized simultaneously two persons as culprit. The face of one of them is really stored into the computer

available at www.mindstien.net

17

software as culprit. For that person Bio-metrics technology has been proved 100% true. While according to plan other person is eating a Jellybeans candy and he acts as if he is blowing. By doing this face was looking quite similar to the first one. So, camera recognizes him as a culprit by ringing an alarm. Because of continuous movement of head camera cannot recognize the face. In short, in compare to finger print and iris pattern -face print is less authentic. In some circumstances this is very useful like when we caught the culprit and make him/her to stand against the camera. A software FaceIt can compare 6, 00, 00,000 photos against the face/minute. In this process there is no chance of any error. But we cannot relay on it completely. Because of Bio-metrics the human body is serving s mobile lock and key.

available at www.mindstien.net

18

8.5 Hand Geometry: Hand geometry is by far the most widely used biometric system. The ID3D Hand key system from Recognition Systems, Inc. is the most popular. Various methods are used to measure the hand; these methods generally fall into one of two categories: 1. Mechanical 2. Image-edge detection Either method produces estimates of certain person. Key measurements of the hand (length of fingers and thumb, widths, etc.); this data are used to "categorize" a person.

It works as follows: The user first enters a PIN number on a keypad, and then positions their hand on a plate using a set of guidance pins which ensures that the hand will be in generally the same position for every measurement. Then a digital camera mounted above the plate, with the aid of a mirror, takes a picture of the top and side views of the hand. The dimensions of the hand, such as finger length, width and area, are extrapolated from the image and the magnification of the camera. This system, however, is subject to an attack using a fake hand modeled after that of an authorized user.

8.6 Retinal Scan: Retinal Scan technology is based on the blood vessel pattern in the retina of the eye. A retinal scan can produce almost the same volume of data as a fingerprint image analysis. It works as follows: The user positions their head against a support, and a low power infrared light is directed against the back of the retina. The image of the pattern of veins is reflected back to a camera.

available at www.mindstien.net

19

8.7 Keystroke Dynamics: This method analyzes the way in which a user types at a computer keyboard. The input is monitored thousands of times a second, and the durations of keystrokes and the latencies between them are recorded. The goal for keyboard dynamics is continual authentication of the user while at a computer; so that if an intruder user had access to the users session while they were away, the machine would eventually be able to recognize the discrepancy. Even though typing patterns are behavioral characteristics, they are very hard to mimic.

8.8 Facial Recognition: Facial recognition technology has recently developed into two areas of study: 1. Facial metrics 2. Eigenfaces. Facial metrics technology relies on the measurement of specific facial features (e.g., the distance between the inside corners of the eyes, the distance between the outside corners of the eyes and the outside corners of the mouth, etc.) and the relationship between these measurements. Within the past two years, an investigation has been made into categorizing faces according to the degree of fit with a set of "eigenfaces". It has been postulated that every face can be assigned a "degree of fit" to each of 150 eigenfaces; further, only the template eigenfaces with the 40 highest "degree of fit" scores are necessary to reconstruct a face with over 99% accuracy. The difference between the eigenface method of facial categorization and the police artist method of building a face from template parts is that the eigenface method is based upon an actual photo of the individual and the "eigenface" information is derived from a computer-based analysis of the digital image of the photo. Eigenfaces are (reportedly) highly repeatable and are not affected by human subjectivity. Eigenface technology has some promise, but it is a technique that is just in the infancy stage of development. Very little data regarding eigenface error rates (false negative, false positive) exists at this point.

8.9 Signature Recognition: Signature recognition is based on the dynamics of making the signature, i.e., acceleration rates, directions, pressure, stroke length, etc., rather than a direct comparison of the signature after it has been written. The problems with signature recognition lie in the means of obtaining the measurements used in the recognition process and the repeatability of the signature. The instrumentation 20

available at www.mindstien.net

cannot consistently measure the dynamics of the signature. Also, a person does not make a signature in a fixed manner; therefore, the data obtained from any one signature from an individual has to allow for a range of possibilities. Signature recognition has the same problem with match discrimination (i.e., finding a match in a large database) as does hand geometry.

*Biometrics Captured Using Web Cam and Microphone*

Figure 1 is made up of images of 320240 pixels. Figure 1(a) is the typical upright front image that we captured before processing it. At 50mm, the shortest depth-of-field, we can capture eye at a resolution of 320240, as shown in Figure 1(b). By visual judgement, this is quite promising for the authentication purpose, when lighting and reflection can be controlled. Since all eyes are round, colour analysis becomes an important property for extraction.

(a)

(b)

(c)

(d)

(e) Figure 1: Biometrics captured by the Webcam in this project Figure 1(c) shows an ear. An ear has more structure thus creates different shading. Segmenting ear can be very challenging. However, if ear can be segmented reliably and represented in a canonical image space, it is possible to extract ear shape information. The same goes to fist as shown in Figure 1(d). We tried to capture thumb too in the hope that fingerprints might be observed. However, at such resolution, fingerprints are partially captured and are not clear.

available at www.mindstien.net

21

By visual judgement, we have decided to use face biometrics, over eye biometrics, ear and fist in decreasing order of priority.

*COMPARISON OF BIO-METRICS TECHNIQUE*

10.1 Uniqueness of biometric features

Table shows the cross accuracy of different biometrics. Table: Comparison of Biometrics Techniques [Ruggles, 1998]

Biometrics Retinal Scan Iris Scan Fingerprints Hand Geometry Signature Dynamiques Voice Dynamiques

Crossover Accuracy 1:10,000,000+ 1:131,000 1:500 1:500 1:50 1:50

Crossover accuracy is the ratio of the crossover frequency as compared to the whole population of test data. The smaller the crossover accuracy, the more unique the biometrics is. It should be noted that comparison across biometrics is by no means measurable. Even the quality of same biometrics is difficult to judge unless the results are tested on the same data set. [OGorman, 1999]. Therefore, sufficient bias is expected from the readers as well.
From Table 2, Retinal scan, after [Daugman, 1995], appears to have the highest crossover accuracy. Even though Iris scan has high cross over accuracy, its user acceptability is low. Fingerprints and hand geometries are equally unique.

Signature dynamics and voice dynamics have the lowest accuracy rates according to [Ruggles, 1998]. In addition, these two techniques rely on behavioural measurements as opposed to physical measurements. In general, behavioural biometrics is less reliable than physical biometrics.

available at www.mindstien.net

22

Retinal scan has high accuracy but also has a high data collection error rate and low user acceptability. For this reason, retinal scan exists only in science fiction movies but not in real life applications! The fingerprint biometric has a low data collection error rate and high user acceptability. Further, fingerprint technology has had the most research and development money applied to both identification and authentication problem. Finally, fingerprint biometrics has the highest acceptance in the identification community and virtually every large biometrics system in operation today uses fingerprint biometrics. Notwithstanding its association with "criminal" applications, fingerprint biometrics is generally accepted by clients. The chosen biometrics in this project, i.e., face and voice, are based on user acceptability and another important factor, the availability of resources in the laboratory. Once the system is established, moving into other biometrics would be easier. Table 3: Comparison of biometrics technologies based on perception of three biometrics experts according to [Jain, 1999]

available at www.mindstien.net

23

*HOW WELL DO BIOMETRICS WORK?*

11.1 11.2 11.3 11.4

FALSE ACCEPT RATES FALSE REJECT RATES EQUAL ERROR RATES VALIDITY OF TEST DATA

Biometric devices can be adjusted to favor security or user convenience. How well do Biometrics work? Is depending on four points.

11.1 False Accept Rates

The probability that a biometric device will allow an unauthorized person is called the "False Accept Rate. False Accept Rates claimed for today's biometric access systems range from 0.0001% to 0.1%. It's important to remember that the only way an unauthorized person can get access is if a unauthorized person tries. Thus, the False Accept Rate must be multiplied by the number of

available at www.mindstien.net

24

attempts by unauthorized person to determine the number of possible occurrences.

11.2 False Reject Rates

The probability that a biometric device won't recognize an authorized person is called the "False Reject Rate. The False Reject Rates quoted for current biometric systems range from 0.00066% to 1.0%. A low False Reject Rate is very important for most applications, since users will become extremely frustrated if they're denied access by a device that has previously recognized them.

11.3 Equal Error Rates

The point where false accept and false reject curves cross is called the "Equal Error Rate. Error curves give a patternical representation of a biometric device's "personality." The Equal Error Rate provides a good indicator of the unit's performance. The smaller the Equal Error Rate, the better.

11.4 Validity of Test Data

Here, it checks for validity of the processed data and decides whether the person is authorized or not. Testing biometrics is difficult, because of the extremely low error rates involved. To attain any confidence in the statistical results, thousands of transactions must be examined. It's important to remember that error rates are statistical: they are derived from a series of transactions by a population of users. In general, the larger the population and the greater the number of transactions, the greater the confidence level in the accuracy of the results.

available at www.mindstien.net

25

If the reported error rate is 1:10, then a sample of 100 transactions may provide a sufficient confidence level. Conversely, a 100-transaction sample would be too small if the error rate was reported as 1:100,000.

Biometric devices are extremely secure, thanks to the combination of low False Accept Rates at moderate sensitivity settings, combined with a short user keypad code. At the same time, biometrics is extremely convenient and errorfree, thanks to low False Reject Rates.

*IS DNA A BIOMETRIC?*

DNA differs from standard biometrics in several ways: DNA requires a tangible physical. DNA matching is not done in real-time, and currently not all stages of comparison are automated. DNA matching does not employ templates or feature extraction, but rather represents the comparison of actual samples. Regardless of these basic differences, DNA is a type of biometric inasmuch as it is the use of a physiological characteristic to verify or determine identity.

*ABOUT BIO-METRICS*
13.1 13.2 13.3 13.4 13.5 13.6 13.7 WILL BIO-METRICS SOLVE ALL SECURITY CONCERNS? WHICH IS THE BEST IDENTIFICATION TECHNIQUE? IS BIO-METRICS MORE SECURE THEN PASSWORD? APPLICATION NEW OPPORTUNITIES FUTURE TECHNOLOGY BENEFITS

available at www.mindstien.net

26

13.1 Will Biometric Solve All Security Concerns?

No System Can be 100% Safe Biometric Measures Can be Very Successful in Fighting Today's Security Concerns Major Drawback in Adopting Biometric Systems is that Concerned person Need to be Physically Present

13.2 Which is the best technique among above?

Accordingly, the pattern of iris is the best one to recognize a person. Because the possibility of two persons having the same iris pattern is 1 per 1000 million, the population of earth is 6 million so eyes are 12 million and that is why probability is not at least 1%.

13.3 Is Bio-metrics more "secure" than passwords?

This question at least poses two problems: biometrics is not equal to biometrics, and the term "secure" is in fact commonly used, but it is not exactly defined. However, we can try to collect pros and cons in order to find at least an intuitive answer. It is a matter of fact that the security of password protected values in particular depends on the user. If the user has to memorize too many passwords, he will to use the same passwords for as many applications as possible. If this is not possible, he will go to construct very simple passwords. If this will also fail (e.g., if the construction rules are too complex), the next fallback stage is to notify the password on paper. This would transform "secret knowledge" into "personal possession". Of course, not every user will react this way. Rather the personal motivation plays an important role: is he aware of the potential loss caused by careless handling of the password? It is easy if the user is the owner. But often foreign possession (e.g., that of the employer) has to be guarded, whose value one often can hardly estimate. If motivation is missing, any password primarily tends to be felt bothersome. In this case, and that seems to be the normal case, it is assumed that biometrics has considerable advantages. Contrariwise, passwords feature unbeatable theoretic protection ability: an eight-digit password which is allowed to contain any symbol from an 8-bit alphabet offers 1020 possible combinations! This is a real challenge for any biometric feature. The requirements are obvious: such a password is maximally difficult to learn, it must not be written down, it must not be passed to 27

available at www.mindstien.net

anyone, the input must take place absolutely secret, it must not be extorted, and the technical implementations must be perfect. This leads us to the practical aspects: the implementation must be protected against replay attacks, keyboard dummies (e.g., false ATMs), wiretapping etc. Even biometric features have to cope with such problems. However, it can be assumed that the protection of biometric feature acquisition is not easier than the acquisition of the password, provided the implementation expense is comparable!

13.4 Where BIO-METRICS is used?

For Security Protect Sensitive Data High degree of identity certainty in transactions Create databases with singular identities

For Accountability Improve auditing / reporting / record keeping Time keeping

For Efficiency Reduce password-related problems It is also used in following areas.

Enterprise-wide network security infrastructures Secure electronic banking, investing and transactions Retail sales Law enforcement Health & Social services

other

financial

13.5 The New Opportunities for Biometrics

The unfortunate events of September 11, 2001 have placed the recent focus on creating new systems and installations employing biometrics. Examples include facial recognition at airports and fingerprint and iris scanning solutions at new government facilities.

available at www.mindstien.net

28

 Combination of Biometrics with smart cards Smart cards, when combined with biometrics, offer a number of benefits. o Smart cards provide a portable storage mechanism for the biometric template. This means template management is eliminated across the biometric reader network. Enrolled users present their smart card to the biometric reader at any location where the card is valid. The biometric template stored on the card (which is usually encrypted) is compared to the live biometric. If the two match, the system grants the user access. o Another benefit of combining biometrics with smart cards pertains to dual-technology cards that are embedded with a smart chip. These combination cards function as both smart cards and proximity cards. This grouping permits organizations currently using proximity-based access control to incorporate biometric security and limit card issuance to users who are permitted access to the biometrically secured area. Lastly, since the biometric templates are stored on individual cards, the number of users for any particular biometric reader is unlimited. This scenario differs from biometric readers that store o the template, thereby limiting the template capacity to the reader's storage capacity.

13.6 Near Future Technologies

In near future every person having a Bio-chip in the body. This Bio-chip concept is recently used in an African Safaris to trap the activities and diseases of animals. For this every animal has a Bio-chip in its body. For Bio-chip in human we will store identification information. So, in future it may not require to ask What is your name? just make a shake hand and allow the Bio-chips to interact and will know the persons Identification.

available at www.mindstien.net

29

We dont know, But Bio-metrics identification may include brain Chemistry because every one has got different brain chemistries. Nowadays, a Bio-Processor-chip is available which can recognize the bacteria that can cause diseases- spread through blood, saliva And urine in few minutes instead of hours.

13.7 Benefits of Adopting Biometrics

No Need to Remember Passwords: As it uses our biometry we are not supposed to give any password. Because where ever we go our finger print, iris pattern, our face, voice definitely with us! Unauthorized Access to Personal Data Can be Prevented: It is a one type of security system that no one can hack it because they do not have our fingerprint, iris, voice, face etc. So, it prevents the data from an unauthorized person. Fraudulent Use of ATMs, Credit Cards Can be Prevented: None can use our ATM just by knowing our PIN. Because no number or password is there.

available at www.mindstien.net

30