Methodologies Of Information System

Published Date: 02 Nov 2017

Verification involves accepting or denying the person's identity. The identification has to establish a person's identity. Biometric systems are divided based on the authentication medium used. They are widely divided as identification of Hand Geometry, Voice Pattern, DNA, Vein Pattern, Signature Dynamics, Iris Pattern and ,Face Detection Finger Prints. These methods are used based onthe scope of the testing medium, the accuracy required and speed required. Every medium of authentication has its own shortcomings and advantages. While the use of computers is increased as the vehicle of the information technology, it is necessary to restrict unauthorized access. Biometric techniques being able to augment this restriction are enjoying a renewed interest.

How it works:

Biometrics systems work by recording and comparing biometric characteristics. In many cases, characteristics are recorded as images, but for signature recognition time series data and for speaker recorded as a wave form. For efficiency reasons, instead of using recorded characteristics directly, it is better to extract identifying features from the samples and encode the features in the form that facilitates storage and comparison.

When the person first uses a biometric system, his identifying features are enrolled as a reference for future comparison. This reference may be stored in the database or on a card depending on the needs of the application (it also can be stored in both database and card).

When biometric recognition is required, the individual�s biometric characteristics are recorded again. This time however, the identifying features are compared by the system with the stored reference to determine if there is a close match.

Biometric Characteristics

Many of biometric characteristics available and are in use different applications, each biometric has its weaknesses and strengths. No single biometric system is expected to meet the requirements of all the applications. In other words, no biometric is optimal. Any human behavioral and/or physiological characteristic can be used as a biometric characteristic if it satisfies the following requirements:

1. Universality: every person around the world should have a biometric characteristic based on which he/she could be represented, for example everyone has a different thumb impression which is unique.

2. Distinctiveness: every two persons should be different in terms of a biometric characteristic measure.

3. Collect-ability: the biometric characteristic can be measured quantitatively with an ease. For example taking a snap of face is typically easy with a camera, where as it is difficult to take a retina sample of a person, but it is possible.

4. Permanence: the characteristic should be sufficiently invariant with the respect to the matching criterion, over a period of time. Taking an example of face recognition as a person get old in age his or her face changes over a time.

5. Performance: That refers to the achievable recognition accuracy and speed, the resources required to achieve the desired recognition accuracy and speed, as well as the operational and environmental factors that affect the accuracy and speed.

6. Acceptability: That indicates the extent to which the people are going to accept the using of a particular biometric identifier in the daily lives.

7. Circumvention: That reflects how easily the system can be fooled using fraudulent methods or could be avoided. For example in voice recognition any person�s voice can be recorded and the biometric system can easily be fooled.

Applications:

While biometric systems, especially automatic fingerprint identification systems, has been widely used in forensic identification for criminals, progress in biometric censors and matching algorithms have led to the deployment the authentication of biometric in a huge number of government and civilian applications. Biometrics is being used for computer log-in, physical access control, international border crossing welfare disbursement, and national ID cards. It can be used to verify the customers during transactions conducted via telephone and Internet. In automobiles, biometrics is being taken to replace keys for keyless entry and keyless ignition. Due to increased security threats, the International Civil Aviation Organization (ICAO) has approved the use of electronic passports.

Benefits of using Biometric systems:

For employers

1. Reduced costs of password maintenance

2. No shared passwords

3. Deter and detect fraudulent account access

4. No badge sharing in secure areas

5. familiarity with advanced technology

For employees

1. No passwords to remember or reset

2. Faster login

3. Confidential files can be stored securely

4. Biometrically transactions difficult to refute

For consumers

1. No passwords to remember or reset

2. Personal files, including emails, can be secured

3. Online purchases safer when enabled by biometric

4. Ability to transact anonymously

For retailers (online and point-of-sale)

1. Biometric users less likely to commit fraud

2. First to offer secure transaction method

3. Account access much more secure than via password

For public sector usage

1. Strongest way to detect and deter benefits fraud

2. Reduced entitlement abuse

Types of Biometric systems:

There are many types of biometric systems. For example:

1. Fingerprint

2. Facial recognition

3. Voice recognition

4. Iris-scan

5. Hand geometry

Fingerprint:

It is useful biometrics technology because fingerprints have long been recognized as a accurate and primary identification method.

Advantages:

1. Fingerprint is practical in forensic investigation.

2. Collecting samples is easy using low technology that means there is a continuous decline in the price and size of fingerprint readers. The conversion of fingerprints into digital images is being easier, cheaper and better.

3. There are large legacy databases of fingerprints in existence.

Disadvantages:

1. There is a large difference in the quality of the fingerprint over the population. The appearance of a person�s fingerprint depends on age, grease, and cut or worn fingers.

2. Elastic distortion of the finger due to touch potential problems and sensing methods with cleanness of the sensor and public hygiene.

3. There are some people without fingers (very rare cases), or without a full set of fingers. Obviously, these individuals cannot be fingerprinted.

Factors Cause Fingerprint to Fail:

� Cold finger

� Dry/oily finger

� High or low humidity

� Angle of placement

� Pressure of placement

� Location of finger on platen (poorly placed core)

� Cuts to fingerprint

� Manual activity that would mar or affect fingerprints (construction, gardening)

Facial recognition:

Uses image or series of images, either from a photograph or a camera to recognize a person.

Advantages:

1. Facial recognition systems are the least intrusive from a biometric sampling point of view because they neither require contact nor the awareness of the subject.

2. The biometric works with legacy photograph databases, video tape and other image sources.

3. It is a fairly good biometric identifier for small-scale verification application.

Disadvantages:

1. A face needs to be well lit by controlling light sources in the automated face authentication systems.

2. Face is better in verification and poor for use in a pure identification protocol.

3. There are some criminal associations with the face identifiers since this biometric has been used for long time by law enforcement agencies.

Factors Cause Facial Recognition to Fail:

� Change in facial hair

� Change in hairstyle

� Lighting conditions

� Adding/removing hat

� Adding/removing glasses

� Change in weight

� Change in facial aspect

� Too much or too little movement

� Quality of capture device

� Change between enrollment and verification cameras (quality and placement)

� �Loud� clothing that can distract face location

Voice recognition:

It is speaker recognition, considered both behavioral and physiological, and it�s low-cost and popular, but less accurate and sometimes lengthy enrollment

Advantages:

1. Voice is a natural biometric (one that people use instinctively to identify each other) under certain circumstances (phone) and machine decisions can be verified by relatively unskilled operators.

2. The voice biometric requires only inexpensive hardware and is easily deployable over existing, ubiquitous communications infrastructure (the telephone system). Voice is therefore very suitable for pervasive security management.

3. Voice allows incremental authentication protocols. For example, the protocol prescribes waiting for more voice data when a higher degree of recognition confidence is needed.

Disadvantages:

1. With the improvement of text-to-speech technology improving, it becomes possible to create non-existent identities with machine voices (when enrolment and authentication are remote) and trainable speech synthesis may make it possible to create an automatic system that can imitate a given person saying anything.

2. Voice recognition is dependent on the quality of the captured audio signal. Speaker identification systems are susceptible to background noise, channel noise (from phone lines, wireless transmission, or severe compression) and unknown channel or microphone characteristics.

3. Speech characteristics can drift away from models with age

Factors Cause Voice recognition to Fail:

� Cold or illness that affects voice

� Different enrollment and verification capture devices

� Different enrollment and verification environments (inside vs. outside)

� Speaking softly

� Variation in background noise

� Poor placement of microphone / capture device

� Quality of capture device

Iris-scan:

Analyzing the iris of the eye, this is the colored ring of the tissue that surrounds the pupil of the eye

Advantages:

1. Iris recognition has the smallest outlier (those who cannot use/enroll) group of all biometric technologies.

2. Template longevity is a key advantage of this technique, a single enrolment can last a lifetime.

3. The iris has a fine texture that is determined in random during embryonic gestation. Even genetically identical individuals have completely independent iris textures.

Disadvantages:

1. There are few legacy databases.

2. The small size of the iris makes sampling of the iris pattern require a great deal of user cooperation or complex, expensive input devices.

3. The performance of iris authentication may be weakling by glasses, sunglasses and contact lenses.

4. The iris biometric is not left as evidence on crime scene so it is not useful for forensic applications.

Factors Cause Iris-scan to Fail:

� Too much movement of head or eye

� Glasses

� Colored contacts

Hand geometry:

Hand geometry systems are generally available in two main forms. Full hand geometry systems take an image of the full hand while Two Finger readers image only two fingers of the hand.

Advantages:

1. Simple, relatively easy to use and inexpensive

2. Easier data collecting in Hand geometry, unlike the fingerprints

3. Environmental factors like dry weather will cause the drying of the skin is not an issue

4. Usually considered less intrusive than fingerprints, retinal, etc

Disadvantages:

1. The hand geometry is not unique and cannot be used in identification systems

2. Not ideal for growing children

3. Jewelry like rings, limited dexterity may pose a challenge in extracting the hand information

4. The data size of hand geometry biometrics is huge and is not ideal for using it in embedded systems

Factors Cause Hand geometry to Fail:

� Jewelry

� Change in weight

� Bandages

� Swelling of joints

Attacks against biometric systems:

There are six major types of threats that may attack a biometric system:

1. Circumvention: An attacker gets access to the system protected by biometric authentication

� Privacy attack: Attacker accesses the data that is not authorized to access (e.g., accessing the medical records of another user)

� Subversive attack: Attacker manipulates the system (e.g., submitting bogus insurance claims)

2. Repudiation: An attacker denies accessing to the system

� A bank clerk modified the financial records and then claims that his biometric data is stolen and denies that he is responsible

3. Contamination (covert acquisition): An attacker illegally gets biometric data of genuine users and try it to access the system

� Lifting a latent fingerprint and constructing a synthetic finger

4. Collusion: A user with wide super user privileges like system administrator, illegally modifies the system

5. Coercion: An attacker forces an authorized user to access the system

6. Denial of Service: An attacker corrupts the biometric system so that authorized users cannot use it

Figure 2 (Points of attack for a generic biometric system)

Some attacks:

� Attack 1: A fake biometric is presented at the sensor.

� Attack 2: Illegally intercepted data is resubmitted.

� Attack 3: Feature detector is replaced by Trojan horse software.

It produces sets of feature chosen by the attacker.

� Attack 4: Legitimate features are replaced with a synthetic set of feature.

� Attack 5: Matcher is replaced by Trojan horse software.

It produces the scores chosen by the attacker.

� Attack 6: Templates in the database are removed, modified or new templates are added

� Attack 7: The transformed template information is modified in the communication channel.

� Attack 8: The matching result is overridden.

Attack Examples:

Synthetic Biometric Submission:

� No detailed system knowledge or access privileges are necessary.

� Digital protection mechanisms are not applicable.

Some fingerprint verification systems attacked by the dummy finger.

1. Dummy finger is created with cooperation of the user within few hours with liquid silicon rubber.

2. Dummy finger is created from a lifted impression of the finger without cooperation of the user within eight hours with silicon cement.

Some fingerprint verification systems attacked by Gelatin fingers.

1. With cooperation like finger pressed to plastic mold.

2. Without cooperation like residual fingerprint lifted from a glass.

Bypass Sensor:

1. Hill-climbing attack for a simple image recognition system:

a. Matching: Template images create correlation filters, these filters are then used with input images.

b. Attack: Synthetic images are input to the system:

i. At each iteration, randomly alter the gray level (8 bits) of 64 pixels: if matching score improves, keep the new image

ii. Continue till the system is compromised

2. Hill-climbing attack for three well known commercial face recognition systems.

a. Attack:

i. Select an initial image from a local database, based on the highest matching score

ii. At each iteration, successively add an eigenface multiplied with 6 constants (-3c, -2c, -c, c, 2c, 3c) to the current synthetic image: keep the change that results in the best matching score improvement

iii. Crop the gray scale values if they are outside the image capacity (8 bit ??0-255 values are allowed)

iv. Continue till the system is compromised

Bypass Feature Extractor:

1. Hill climbing-based attacker for a minutiae-based fingerprint authentication system.

2. Location (r,c)and orientation ?of the minutiae used

3. Template information unknown to the attacker, This is the secret data that attacker wants to mimic.

4. A synthetic minutia sets input to the matcher.

5. Attacker tries to generate a minutia set that result in a high matching score using the scores returned by the matcher.

Generate Biometric from Template Data:

1. Synthetic images generated from reverse engineered minutiae template data from a commercial fingerprint authentication system:

a. Author accessed unencrypted template data from a computer hard drive

b. The format of the accessed template discovered by trial/error and by introducing controlled changes in input images. For each minutiae, its 2D location, angle and ridge curvature was found

c. Orientation field of the target image estimated based on core and delta point locations.

d. Lines starting at minutiae points are drawn, by taking into account the orientation field

e. Synthetic images are not very realistic, but still they were accepted as genuine template images

2. Synthetic images are generated from minutiae location and angle:

a. Use minutia triplet sand estimate orientation fields inside the triangles using minutiae angles at 3 vertices

b. A neural network is used to estimate the fingerprint class from features of minutiae pairs

c. Estimated orientation fields are used as inputs to Gabor-like filters to generate synthetic images

Face identification system:

Criminal record contains personal information about specific person along with his photograph. To identify any criminal you need some identifications regarding person, which are provided by eyewitnesses. In most cases the resolution and quality of the recorded image-segments is poor and hard to identify a face. To overcome this problem we are developing software. We can do identification in many ways like fingerprint, eyes, hand, etc. One of the applications is face identification.

User Interface:

An interface is a collection of operations that specify component or a service of a class. An interface describes the externally visible behavior of that element.

Graphical User Interface:

1. Main Form:

The system main form present an over view to face database, and the basic function (Traverse, Add, Remove, Print), the user can use the muse and double click on any record, to retrieve back the selected data in popup form, where the user can modify the record, change the image, and print the record (as Shown Below).

2. Add New Data Record:

This form allows the user to add a new record; it appears if the user clicks on the (Add function from the Main form (as Shown Below).

3. Read Data:

The system needs to read the face database to create and save the eigenvector & the Eigen face. This process requires expert supervision; it must be done manually, whenever there are some updates in the database content. Otherwise, the system will keep the reading of the old saved features. Is possible to read the data automatically, but it involves overhead, and time consuming (as Shown Below).

4. Find Match

To get the match, the user can add the image from the File drop menu, or by dragging and dropping the image in drag-drop box (as Shown Below).

Database:

Database ER-Diagram:

Database Tables:

FACE:

Field Name Data Type Size

Criminal_Id

First Name

Last Name

Birth_Date

Age

Gender

Address

City

Arrested_Date

Crime_Info

Photo Number

Varchar2

Varchar2

Date

Number

Varchar2

Varchar2

Varchar2

Date

Varchar2

Long 5 (Primary Key)

15

15

5

8

20

15

20

CRIMINAL_SUSPECTTABLE:

Field Name Data Type Size

Crime_id

Criminal_Id Number

Number 5 (Primary Key)

5

FACE_HEIGHTSTABLE:

Field Name Data Type Size

Criminal_Id

Hair

Forehead

Eyes

Nose

Lips

Chin Number

Number

Number

Number

Number

Number

Number 5 (Foreign Key)

5

5

5

5

5

5

FACE_SUSPECTPHOTO:

Field Name Data Type Size

Crime_id

Suspect_Photo

Suspect_Height Number

Varchar2

Varchar2 5 (Foreign Key)

15

15

������������������������������������������������������������������������������������������