0% found this document useful (0 votes)
42 views15 pages

Lecture : Biometrics

The document discusses biometrics and key elements of biometric systems. It explains that enrollment is the process of collecting biometric samples from a person to create a template, which is a data representation of distinguishing features extracted from the samples. Matching is the process of comparing a submitted biometric sample against one or many templates in the system database. The document also compares different types of biometrics and their attributes.

Uploaded by

Basant
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
42 views15 pages

Lecture : Biometrics

The document discusses biometrics and key elements of biometric systems. It explains that enrollment is the process of collecting biometric samples from a person to create a template, which is a data representation of distinguishing features extracted from the samples. Matching is the process of comparing a submitted biometric sample against one or many templates in the system database. The document also compares different types of biometrics and their attributes.

Uploaded by

Basant
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

Biometrics

Lecture (3)

➢ Biometric authentication technologies are used in two ways:

▪ To prove who you are or who you claim you are.


▪ To prove who you are not (for example, to resolve a case of
mistaken identity).

‫ إلثبات هل انت هو الشخص وال ال‬: Biometric authentication‫ يستخدم ال‬-

➢ KEY ELEMENTS OF ALL BIOMETRIC SYTSEMS :

1) Enrollment
2) Templates
3) Matching

➢ Enrollment and template :

- Enrollment : is the process of collecting biometric samples


from a person .
‫ من االشخاص‬biometric ‫ هى عمليه جمع ال‬enrollment ‫ ال‬-

- They are created by the biometric device, which uses an


algorithm to extract “features” appropriate to that technology
from the enrollee’s samples.

‫ الداخل فى عمليه ال‬samples ‫ المميزه من كل ال‬features ‫ نقوم بعمل عمليه استخراج لل‬-
enrollment

# Scope 8
- Templates : are the data representing the enrollee’s biometric.

‫ الخاص بكل‬features ‫ الذى يخزن فيه ال‬database ‫ فى ال‬record ‫ هى‬template ‫ ال‬-


‫شخص‬

- Because templates are only a record of distinguishing features


of a person’s biometric characteristic or trait (and not an image
or complete record of the actual fingerprint or voice), the template
is usually small and allows for the near-instantaneous processing
time characteristic of biometric authentication.

‫ فتستهلك مساحه‬, ‫ الداخل‬sample‫ المميز من ال‬features ‫ يخزن ال‬template ‫ ال‬-


template ‫ كامله ( مثل الصوره ) فى ال‬sample ‫ وال نحتفظ بال‬, ‫صغيره‬

- The small size of some templates allows for storage on bar


codes placed on plastic cards.

# Scope 8
- Example of the Formation of a Template for a Fingerprint :

- For any biometric technology, a small percentage of the


population will be unable to produce a usable template.

‫ بشكل‬sample ‫ بعض الناس ال تستطيع ان تدخل ال‬, enrollment ‫ وجدوا ان فى عمليه ال‬-
‫ غير سليم‬template ‫صحيح و بالتالى تكون الداتا الخاص بهم الموجوده فى ال‬

- FTA (Failure To Acquire) , FTE(Failure To Enroll) :

- This failure to enroll (or acquire) is the failure of the technology


to extract adequate distinguishing features appropriate to that
technology.

system ‫ الموجوده فى ال‬error ‫ هى نسبه ال‬: )Failure To Acquire( FTA ‫ ال‬-


, sample ‫ او حدوث خطا اثناء اخذ ال‬features ‫بسبب حدوث خطا عند عمليه استخراج ال‬
testing ‫ خطا فى اثناء عمليه ال‬biometric system ‫يؤدى لقرار ال‬

- The device takes three samples of the same biometric and then
averages them to produce an enrollment template.

# Scope 8
‫ هذه‬average ‫ ثم ناخذ‬, ‫ هى اخذ اكثر من صوره لنفس الشخص‬FTA ‫ حل مشكله ال‬-
template ‫الصور و نستخدمه ك‬

- For example, a small fraction of the population cannot be


fingerprinted either because their prints are not distinctive enough
(e.g., no bifurcations that can be picked up by the system) or
because of the individual’s age, which can alter distinguishing
features.

➢ Matching :

- Matching : is the process of comparing a submitted biometric


sample against one (verification) or many (identification)
templates in the system’s database.

‫ واحد فقط موجوده‬template ‫ الداخل و‬sample ‫ هى عمليه مقارنه بين ال‬matching ‫ ال‬-
‫ الداخل و كل ال‬sample ‫ او مقارنه بين ال‬, verification ‫ لو كان‬database ‫فى ال‬
identification ‫ لو كان‬database ‫ الموجوده فى ال‬templates
matched score ‫لينتج عن عملته المقارنه‬

!‫ ؟‬matched score ‫ ما هو ال‬-


‫ و الذى سوف‬, matching ‫ الناتج من عمليه ال‬score ‫ او ال‬output ‫✓ هو ال‬
reject ‫ او ال‬accept ‫ سؤاء بال‬decision ‫يستخدم الخذ‬

- In general, verification applications provide more security than


identification applications because a biometric and at least one
other piece of input (e.g., PIN, password, token, user name) are
required to match a template.

# Scope 8
‫ من ال‬security ‫ اعلى فى ال‬verification ‫ ان ال‬, ‫ زى ما قلنا فى المحاضره السابق‬-
, ‫ واحد‬template ‫ ب‬sample ‫ بنقارن كل‬verification ‫ الننا فى ال‬, identification
‫ فمن الممكن ان يتشابه الداتا الخاص‬template ‫ بنقارن بكل ال‬identification ‫لكن فى ال‬
system ‫ اى شخص على ال‬template ‫بالشخص المزور مع‬

- Because the search seeks only a match against one template in


the database, verification applications require less processing
time, less memory, and less cost than identification applications.

‫ فلذلك تستهلك ال‬, ‫ واحد‬template ‫ مع‬match ‫ نقوم بعمل‬verification ‫ الن فى ال‬-


identification ‫ اقل من المستهلك فى ال‬memory ‫ وقت و‬verification

- All biometric systems face a common issue:

- The template database must be stored somewhere.

- Biometric templates must be protected to maintain user privacy.

- Possible solutions include storage on the biometric device itself, a


central computer that is remotely accessed, a plastic card
with a bar code, and optical memory cards.

‫ و ذلك عن طريق وضع ال‬, ‫ الخاص بالداتا‬privacy ‫ ان احنا يجب ان نحافظ على ال‬-
‫ خاص بيه او نقوم‬database ‫ فى‬verification ‫ المستخدم فى عمليه ال‬password
‫بتشفيرها‬

➢ Compare between types of biometrics (‫ )مهم‬:

# Scope 8
‫‪: Barriers to universality )1‬‬
‫‪ -‬يعنى ان تكون متوفره عند كل االشخاص ‪ ,‬هذه الخاصيه غير متوفره فى ال ‪face‬‬
‫‪ recognition‬و ال ‪ fingerprint‬الن ممكن ان يحدث قطع فى ال ‪( finger‬من المفروض‬
‫معالجه كل هذه المشاكل فى ال ‪ algorithm‬الذى سنستخدمه)‬

‫‪: Distinctiveness )2‬‬


‫‪ -‬هى ان تكون الصفه مختلف بين االشخاص و مميزه ‪ ,‬القيم ‪ low‬فى ال ‪face recognition‬‬
‫النه ممكن ياخذ من ال ‪ face‬بعض ال ‪ features‬الغير مميزه ‪ ,‬و القيم ‪ low‬فى ال ‪voice‬‬
‫الن الصوت ممكن يتقلد بسهوله ‪ ,‬و القيمه ‪ medium‬فى ال ‪ hand geometry‬الننا‬
‫ناخذ مساحه ال ‪ palm‬و شكل اليد فمن الممكن ان تتشابه القيم بين االشخاص ‪ ,‬لكن ال‬
‫‪ fingerprint‬و ال ‪ iris‬ال تتكرر بين االشخاص لذلك اخذوا قيم ‪high‬‬

‫(ملحوظه ‪ :‬اى نوع من ال ‪ biometrics‬تكون قيمته رقم فيكون من السهل ان يتكرر بين‬
‫االفراد)‬

‫‪# Scope 8‬‬


: Permanence )3
‫ ال تتغير على مدار العمر‬iris ‫ و ال‬fingerprint ‫ ال‬, ‫ ثبات الصفه على مدار فتره من الزمن‬-
‫ قابل للتغير كل لما االنسان يتقدم فى العمر‬hand ‫ و ال‬face ‫ لكن ال‬, high ‫لذلك اخذوا قيم‬
‫ النه متغير دائما‬low ‫ قيم‬voice ‫ و اخذ ال‬, medium ‫لذلك اخذوا قيم‬

: collectability )4
‫ امام ال‬eyes ‫ او ال‬finger ‫ يجب وضع ال‬iris ‫ و ال‬finger ‫ فى ال‬, ‫ القدره على قياسها‬-
‫ و ال‬hand palm ‫ و فى ال‬, medium ‫ صغير بوضع معين لذلك اخذوا القيم‬sensor
high ‫ ياخذ الصوره كامل فلذلك اخذوا قيم‬face

: performance )5
‫ انه عالى الكفائه‬-

: acceptability )6
biometric ‫ ان االشخاص الذين سيستخدموا متقبلين استخدام هذا النوع من ال‬-

: Circumvention )7
‫ غير قابل للتزوير بسهوله‬-

➢ Biometrics Attributes :

• Distinctiveness
- Biometric features allow high levels of discrimination
between individual.

• Stability
- Age, accident or disease, change biometrics over . Preserve
enough features so changes have minimal effect on
system’s ability to discriminate.

‫ انه ثابت وال يتاثر بالتغيرات التى تحدث‬-

# Scope 8
• Scalability
- Biometric capable of being processed efficiently, both at
acquisition time and when searching in a database.
‫ تعنى ان اقدر اعدل فيه بسهوله‬-

• Inclusiveness
- high proportion of population, biometric which excludes
users causes additional complexities in managing security
biometrics ‫ تعنى ان المستخدمين قابلين باستخدام هذا النوع من ال‬-

• Insensitivity
- Changes in the external environment (e.g. lighting,
temperature) within reasonable boundaries should not cause
system failures.
‫ تعنى انها ال تتاثر بالعوامل الخارجيه زى الضوء‬-

• Vulnerability(spoofing)
- Difficult to create a fake biometric (known as spoofing).
‫ صعبه التزوير‬biometric ‫ تعنى ان تكون‬-

• Privacy
- permission of biometric owner before acquisition. Data
should be stored encrypted.

• Maintenance

• Health
- pain is avoided during biometric acquisition
‫ ان تكون غير مؤلم‬-
• Quality
- good quality sample should be easy for user.

# Scope 8
• Integration
- Biometric capable of being used with other authentication
mechanisms, such as smart-cards or passwords.

• Cost
- Cost of system should be in proportion to benefit.

➢ Biometric Data:

- Biometric data is represented or stored in one of these forms:

: ‫ و هى‬biometric data ‫ انواع من ال‬4 ‫ هناك‬-


1) Raw
2) Token
3) Template
4) Metadata

1. Raw Data :

- Raw biometric information (known as the biometric sample) is


data gathered directly from the sensor before any processing
has been carried out.

‫ و هى الداتا(سؤاء كانت صوره او صوت) الداخل من‬sample ‫ تعرف باسم ال‬raw ‫ ال‬-
processing ‫ بدون اى نوع من ال‬sensors

- Biometric acquisition techniques :


camera images, infrared images, sound recordings, full motion
video recordings, keystroke logs, pen motion (signature)
samples ‫ هذه هى بعض االشكال المستخدمه لجمع ال‬-

# Scope 8
2. Token Data:

- Token data is raw data with minimal amount of processing


applied.
raw ‫ قريب من ال‬, ‫ مع القليل من المعالجه‬raw or sample ‫ هى‬token ‫ ال‬-

- Reason for using a token is that advances in algorithms may


discover new ways of extracting distinctive features from the
original biometric sample.

‫ فسوف‬, ‫ ؟! الن ممكن اكتشف مشكله موجوده فى الصور العندنا‬token ‫ ليه استخدمت ال‬-
cost ‫ مره اخرى و هكذا نقلل من ال‬acquisition ‫ بدل من عمل‬token data ‫نستخدم ال‬
‫ المميزه‬features ‫ او الن اكتشفت طريق افضل الستخراج ال‬, ‫المستهلك‬

3. Template Data :

- Template is the refined, processed and stored representation of


distinguishing characteristics of individual.

‫ فى ال‬template ‫ لتوضع فى‬extract features ‫ الناتج من عمليه‬features ‫ هى ال‬-


database

- Template is the data that gets stored during an enrollment and


later used for matching.

‫ و تستخدم بعد ذلك فى‬, enrollment ‫ هى الداتا التى تخزن فى مرحله ال‬template ‫ ال‬-
‫ (بعد ان تمر بنفس ال‬user ‫ الداخل من ال‬sample ‫ لمقارنه ال‬matching ‫عمليه ال‬
template ‫ ) مع ال‬template ‫ الذى تم على الداتا التى حفظت فى ال‬process

4. Metadata:

# Scope 8
- Data that describes attributes of either the biometric sample
(e.g. wearing glasses), the capture process (e.g. time acquired)
or the demographics of the person (e.g. gender and age).

‫ ال‬:‫ الخاص بالداتا زى‬description ‫ او هى ال‬data about data ‫ هى‬metadata ‫ ال‬-


‫ مثال يتغير و الشخص يرتدى‬face ‫ الخاص بكل شخص الن لما نتقدم فى العمر شكل ال‬age
‫ و كان مرتدى خاتم وال ال‬, ‫ و وقت التقاط الصوره‬, ‫نظاره ام ال‬

‫ كى ال يتسبب فى‬preprocessing ‫من المفترض معالجه كل هذه الظروف فى مرحله ال‬


‫مشكله‬

- Metadata is important for testing and evaluation, as strong


correlations between demographic characteristics and
matching performance often are found.

‫ الن ممكن شخص و هو‬testing ‫ ؟! مهم فى عمليه ال‬metadata ‫ طيب ليه ممكن نلجا لل‬-
‫ فبنرجع‬, accept ‫ و هو المفروض انه يكون‬reject ‫ حصل له‬system ‫داخل يسجل على ال‬
‫ و نشوف هل هو كان مثال مرتدى نظاره اثناء التقاط الصوره‬metadata ‫لل‬

➢ Biometric System General Model :

- Biometric System Subsystems :

1) Biometric Capture.
2) Transmission and Processing to enhance biometric features.
3) Storage of biometric information.
4) Biometric comparison.
5) Decision Subsystem.

‫ السابق‬subsystems ‫ من مجموعه من ال‬biometric system ‫ يتكون اى‬-

# Scope 8
▪ Enrollment, verification, and identification transactions all
share related paths through the subsystems.

‫ هناك فى الرسم ثالث اشكال من الخطوط لكل عمليه من‬, ‫ الرسم السابق مهم جدا‬-
‫ حيث ان يوضح‬, ‫ شكل معبر عنها‬Enrollment, verification, and identification
‫مسار كل عمليه منهم‬

1. Data Capture :

- Firstly, the capture of raw biometric information from real


person, which is translated into a digital sample.

# Scope 8
Enrollment, verification, ‫ حيث انها المرحله التى تمر بها كل من‬, ‫ اول مرحله‬-
‫ من االشخاص باستخدام ال‬biometric ‫ و هى عمليه جمع ال‬, and identification
‫ المميز‬features ‫ ثم نستخرج ال‬, sensor

- Common Data Capture Equipment :

- digital cameras (for face recognition), fingerprint scanners


(fingerprint), and microphones (speaker verification).

- Each transformation involves addition of noise.

2. Storage Subsystem :

- Digital Biometric Sample is transformed using signal processing


techniques into reference features that are used to distinguish the
individual. This involves processing the sample to remove noise
extracting features.

‫ المميز‬features ‫ و نستخرج ال‬, ‫ اذا لزم االمر‬preprocessing ‫ نقوم بعمل‬-

- After features have been extracted and are of sufficient


quality, reference template is created.

reference ‫ او‬template ‫ نقوم باالحتفاظ بيها فى‬, features ‫ بعد عمل استخراج لل‬-

- At enrollment, a template is created and then data is stored in a


database or on a device such as a smart-card.

‫ الموجود‬reference ‫ او ال‬template ‫ و سناخذ ال‬, enrollment ‫ هيكمل معانا خط ال‬-


data storage subsystem ‫ او ال‬database ‫ و نحفظه فى ال‬features ‫به ال‬

reference ‫ هو ال‬template ‫ ال‬: ‫ ملحوظ‬-

# Scope 8
3. Comparison Subsystem :

- Biometric Algorithm takes features from the stored reference


template, with features extracted from presentation sample, and
compare them to generate a score which indicates the likelihood
that both are from the same person.

‫ حيث ان هنعمل مقارنه بين‬, identification ‫ و ال‬verification ‫ هيكمل معانا ال‬-


‫ (بعد ما تمر على نفس ال‬user ‫ و الداتا الداخل من ال‬template ‫الداتا الموجوده فى ال‬
‫ حيث ان فى ال‬,) template ‫ المحفوظه فى ال‬sample ‫ الذى مر بيه ال‬algorithm
‫ لكن فى ال‬, database ‫ الموجوده فى ال‬template ‫ يقارن بكل ال‬identification
‫ واحد‬template ‫ نقارن ب‬verification

score ‫و يكون ناتج المقارنه‬

- Output Comparison Score come in a variety of forms, such as


from zero to one, the closer the score is to zero, the more likely
the match.

‫ النه مر‬one ‫ و ال‬zero ‫ قيمته تكون بين ال‬, score ‫ هو‬comparison ‫ ناتج عمليه ال‬-
normalization ‫بعمليه‬

4. Decision Subsystem :

- For verification, comparison score is used to make a decision


about accepting the person as genuine or rejecting them as an
impostor.

- During identification the match is conducted against two or more


enrolled people to produce a candidate list of possible genuine
matches.

# Scope 8
‫ ) او‬genuine == accept( ‫ يتخذ قرار اما انه‬, ‫ الناتج‬score ‫ بناء على قيمه ال‬-
) impostor == reject(

5. Transport subsystem :

- Mechanism by which data is moved between different


subsystems.

‫ من مكان الخر‬system ‫ هى الخطوط التى تنقلك فى ال‬transport ‫ ال‬-

# Scope 8

You might also like