SlideShare una empresa de Scribd logo
1 de 24
By :
Avin Jalal .
Supervisor: Ms.Shaheen Abdlkarim.
AI & PATTERN RECOGNITION
SUBJECTS:
 WHAT IS PATTERN?
 TYPE OF PATTERNS.
 IDEA OF PATTERN RECOGNITION.
 WHAT IS PATTERN RECOGNITION?
 PROCESS PHASE IN PR.
 DESIGN PATTERN RECOGNITION.
 PATTERN RECOGNITION PROCESS.
 WHAT IS PATTERN CLASS?
 APPROACHES FOR PATTERN RECOGNITION.
 PROBLEMS & APPLICATIONS.
WHAT IS PATTERN?
PATTERN : PATTERN IS A SET OF PHYSICAL
OBJECTS WHERE THE ELEMENTS OF THE SET
ARE SIMILAR TO ONE ANOTHER IN CERTAIN
WAYS.[1]
TYPE OF PATTERNS: [1]
SPATIAL PATTERNS:
TEMPORAL PATTERN:
ABSTRACT PATTERNS:
PATTERN RECOGNITION:
THE IDEA OF PATTERN RECOGNITION: [6]
HUMAN PERCEPTION: [2]
PATTERN RECOGNITION:
PATTERN RECOGNITION: Is the science for
how machines can: [6]
– Observing (sensing) the environment,
– Learning to distinguish patterns ,
– and making sound decisions about the patterns
or pattern classes.
RECOGNITION PATTERN WAYS: [9]
1. Statistical way.
2. Artificial Intelligence.
PROCESSING PHASES IN PR: [10]
1.Training/Learning
2.Detecting/Classifying
Input
processing
Feature
Measurement Classification
Pre-processing Feature extraction
selection
Learning
Classification
Train
Feedback
THE DESIGN OF A PATTERN RECOGNITION
SYSTEM : [6]
1. DATA ACQUISITION AND PREPROCESSING:
2. DATA REPRESENTATION:
3. TRAINING:
4. DECISION-MAKING:
PATTERN RECOGNITION PROCESS: [4]
PATTERN RECOGNITION : [9]
PATTERN RECOGNITION : [9]
PATTERN RECOGNITION:
WHAT IS A PATTERN CLASS (OR CATEGORY)?[10]
 SUPERVISED LEARNING: [4]
Examples: [5]
• The Perceptron learning algorithm,
 UNSUPERVISED LEARNING: [4]
Examples : [5]
• k-Means clustering algorithm,
CLASSIFICATION IN (PR):
• SUPERVISED TRAINING/LEARNING: [10]
CLASSIFICATION IN (PR):
• UNSUPERVISED TRAINING/LEARNING: [10]
APPROACHES FOR PATTERN RECOGNITION : [8]
TEMPLATE MATCHING:
STATISTICAL CLASSIFICATION:
SYNTACTIC OR STRUCTURAL MATCHING:
ARTIFICIAL NEURAL NETWORKS:
TEMPLATE MATCHING : [7]
Class Z
Class Y
Input pattern
Class X
Z
Y
Temp X
TEMPLATE MATCHING : [7]
TEMPLATE MATCHING : [7]
STATISTICAL CLASSIFICATION : [8]
Feature Extraction
Algorithm
X3
X2
X1
Y3
Y1
Y2
Feature Extraction
Algorithm
Set Of
Feature X
Set Of
Feature Y
...
STATISTICAL CLASSIFICATION : [8]
REFERENCES:
1. Dr. C. N. Ravi Kumar, Pattern Classification,(http://uni-mysore.ac)
2. Pradeep Mishra, Pattern Recognition, (http://http://www.slideshare.net)
3. Shyh-Kang Jeng, ―Introduction‖, Pattern recognition Course Website, 2009. [online]
Available: http://cc.ee.ntu.edu.tw/~skjeng/PatternRecognition2007.htm. [Accessed Sep.
30, 2009].
4. Talal A. Alsubaie, SFDA, Pattern Recognition,(http://http://www.slideshare.net)
5. M. Tim Jones, Artificial Intelligence A Systems Approach,page-257
6. Anil K. Jain, Robert P.W. Duin, Introduction to Pattern Recognition1,(citeseerx.ist.psu.edu)
7. , , -
,page 182-191
8. Pattern Recognition and Image Processing,
(http://lessons.bdr130.net/1586.html)
9. , Pattern recognition ”
2,1 ,(http://www.teedoz.com)
10. Luís Gustavo Martins, Introduction To Pattern Recognition,
(http://http://www.slideshare.net)
Question ?
THANK YOU ^_^

Más contenido relacionado

Similar a Avin jalal ai & pattern recognition1

Survey Research in Software Engineering
Survey Research in Software EngineeringSurvey Research in Software Engineering
Survey Research in Software EngineeringDaniel Mendez
 
From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...dgarijo
 
From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...dgarijo
 
cvpaper.challenge 研究効率化 Tips
cvpaper.challenge 研究効率化 Tipscvpaper.challenge 研究効率化 Tips
cvpaper.challenge 研究効率化 Tipscvpaper. challenge
 
Machine Learning for Everyone
Machine Learning for EveryoneMachine Learning for Everyone
Machine Learning for EveryoneAly Abdelkareem
 
Dr azimifar pattern recognition lect1
Dr azimifar pattern recognition lect1Dr azimifar pattern recognition lect1
Dr azimifar pattern recognition lect1Zahra Amini
 
Innofest2010
Innofest2010Innofest2010
Innofest2010Saine
 
Making Representations Matter: Understanding Practitioner Experience in Parti...
Making Representations Matter: Understanding Practitioner Experience in Parti...Making Representations Matter: Understanding Practitioner Experience in Parti...
Making Representations Matter: Understanding Practitioner Experience in Parti...alselvin
 
Data Structures and Algorithm - Week 6 - Red Black Trees
Data Structures and Algorithm - Week 6 - Red Black TreesData Structures and Algorithm - Week 6 - Red Black Trees
Data Structures and Algorithm - Week 6 - Red Black TreesFerdin Joe John Joseph PhD
 
RESEARCH in software engineering
RESEARCH in software engineeringRESEARCH in software engineering
RESEARCH in software engineeringIvano Malavolta
 
[2015/2016] RESEARCH in software engineering
[2015/2016] RESEARCH in software engineering[2015/2016] RESEARCH in software engineering
[2015/2016] RESEARCH in software engineeringIvano Malavolta
 
Arrangement of module in teaching methodology
Arrangement of module in teaching methodologyArrangement of module in teaching methodology
Arrangement of module in teaching methodologyJEWELGUINTO
 
Replicable Evaluation of Recommender Systems
Replicable Evaluation of Recommender SystemsReplicable Evaluation of Recommender Systems
Replicable Evaluation of Recommender SystemsAlejandro Bellogin
 
The Art of Intelligence – Introduction Machine Learning for Oracle profession...
The Art of Intelligence – Introduction Machine Learning for Oracle profession...The Art of Intelligence – Introduction Machine Learning for Oracle profession...
The Art of Intelligence – Introduction Machine Learning for Oracle profession...Lucas Jellema
 
A Self-Adaptive Evolutionary Negative Selection Approach for Anom
A Self-Adaptive Evolutionary Negative Selection Approach for AnomA Self-Adaptive Evolutionary Negative Selection Approach for Anom
A Self-Adaptive Evolutionary Negative Selection Approach for AnomLuis J. Gonzalez, PhD
 
Empirical Software Engineering
Empirical Software EngineeringEmpirical Software Engineering
Empirical Software EngineeringRahimLotfi
 

Similar a Avin jalal ai & pattern recognition1 (20)

Exposé Ontology
Exposé OntologyExposé Ontology
Exposé Ontology
 
Survey Research in Software Engineering
Survey Research in Software EngineeringSurvey Research in Software Engineering
Survey Research in Software Engineering
 
Xiangen Hu - WESST Keynote - Conversational Tutors and the Experience API
Xiangen Hu - WESST Keynote - Conversational Tutors and the Experience APIXiangen Hu - WESST Keynote - Conversational Tutors and the Experience API
Xiangen Hu - WESST Keynote - Conversational Tutors and the Experience API
 
From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...
 
From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...From Scientific Workflows to Research Objects: Publication and Abstraction of...
From Scientific Workflows to Research Objects: Publication and Abstraction of...
 
cvpaper.challenge 研究効率化 Tips
cvpaper.challenge 研究効率化 Tipscvpaper.challenge 研究効率化 Tips
cvpaper.challenge 研究効率化 Tips
 
Machine Learning for Everyone
Machine Learning for EveryoneMachine Learning for Everyone
Machine Learning for Everyone
 
NPS_TDA_forPDF_JPrendki
NPS_TDA_forPDF_JPrendkiNPS_TDA_forPDF_JPrendki
NPS_TDA_forPDF_JPrendki
 
Dr azimifar pattern recognition lect1
Dr azimifar pattern recognition lect1Dr azimifar pattern recognition lect1
Dr azimifar pattern recognition lect1
 
Innofest2010
Innofest2010Innofest2010
Innofest2010
 
Making Representations Matter: Understanding Practitioner Experience in Parti...
Making Representations Matter: Understanding Practitioner Experience in Parti...Making Representations Matter: Understanding Practitioner Experience in Parti...
Making Representations Matter: Understanding Practitioner Experience in Parti...
 
Data Structures and Algorithm - Week 6 - Red Black Trees
Data Structures and Algorithm - Week 6 - Red Black TreesData Structures and Algorithm - Week 6 - Red Black Trees
Data Structures and Algorithm - Week 6 - Red Black Trees
 
RESEARCH in software engineering
RESEARCH in software engineeringRESEARCH in software engineering
RESEARCH in software engineering
 
[2015/2016] RESEARCH in software engineering
[2015/2016] RESEARCH in software engineering[2015/2016] RESEARCH in software engineering
[2015/2016] RESEARCH in software engineering
 
Arrangement of module in teaching methodology
Arrangement of module in teaching methodologyArrangement of module in teaching methodology
Arrangement of module in teaching methodology
 
Replicable Evaluation of Recommender Systems
Replicable Evaluation of Recommender SystemsReplicable Evaluation of Recommender Systems
Replicable Evaluation of Recommender Systems
 
Intro_to_ML
Intro_to_MLIntro_to_ML
Intro_to_ML
 
The Art of Intelligence – Introduction Machine Learning for Oracle profession...
The Art of Intelligence – Introduction Machine Learning for Oracle profession...The Art of Intelligence – Introduction Machine Learning for Oracle profession...
The Art of Intelligence – Introduction Machine Learning for Oracle profession...
 
A Self-Adaptive Evolutionary Negative Selection Approach for Anom
A Self-Adaptive Evolutionary Negative Selection Approach for AnomA Self-Adaptive Evolutionary Negative Selection Approach for Anom
A Self-Adaptive Evolutionary Negative Selection Approach for Anom
 
Empirical Software Engineering
Empirical Software EngineeringEmpirical Software Engineering
Empirical Software Engineering
 

Último

Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu SubbuApidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbuapidays
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDropbox
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProduct Anonymous
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWERMadyBayot
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native ApplicationsWSO2
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...apidays
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MIND CTI
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024The Digital Insurer
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobeapidays
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...apidays
 
A Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusA Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusZilliz
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdflior mazor
 

Último (20)

Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu SubbuApidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
 
A Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusA Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source Milvus
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdf
 

Avin jalal ai & pattern recognition1

  • 1. By : Avin Jalal . Supervisor: Ms.Shaheen Abdlkarim. AI & PATTERN RECOGNITION
  • 2. SUBJECTS:  WHAT IS PATTERN?  TYPE OF PATTERNS.  IDEA OF PATTERN RECOGNITION.  WHAT IS PATTERN RECOGNITION?  PROCESS PHASE IN PR.  DESIGN PATTERN RECOGNITION.  PATTERN RECOGNITION PROCESS.  WHAT IS PATTERN CLASS?  APPROACHES FOR PATTERN RECOGNITION.  PROBLEMS & APPLICATIONS.
  • 3. WHAT IS PATTERN? PATTERN : PATTERN IS A SET OF PHYSICAL OBJECTS WHERE THE ELEMENTS OF THE SET ARE SIMILAR TO ONE ANOTHER IN CERTAIN WAYS.[1]
  • 4. TYPE OF PATTERNS: [1] SPATIAL PATTERNS: TEMPORAL PATTERN: ABSTRACT PATTERNS:
  • 5. PATTERN RECOGNITION: THE IDEA OF PATTERN RECOGNITION: [6] HUMAN PERCEPTION: [2]
  • 6. PATTERN RECOGNITION: PATTERN RECOGNITION: Is the science for how machines can: [6] – Observing (sensing) the environment, – Learning to distinguish patterns , – and making sound decisions about the patterns or pattern classes. RECOGNITION PATTERN WAYS: [9] 1. Statistical way. 2. Artificial Intelligence.
  • 7. PROCESSING PHASES IN PR: [10] 1.Training/Learning 2.Detecting/Classifying Input processing Feature Measurement Classification Pre-processing Feature extraction selection Learning Classification Train Feedback
  • 8. THE DESIGN OF A PATTERN RECOGNITION SYSTEM : [6] 1. DATA ACQUISITION AND PREPROCESSING: 2. DATA REPRESENTATION: 3. TRAINING: 4. DECISION-MAKING:
  • 12. PATTERN RECOGNITION: WHAT IS A PATTERN CLASS (OR CATEGORY)?[10]  SUPERVISED LEARNING: [4] Examples: [5] • The Perceptron learning algorithm,  UNSUPERVISED LEARNING: [4] Examples : [5] • k-Means clustering algorithm,
  • 13. CLASSIFICATION IN (PR): • SUPERVISED TRAINING/LEARNING: [10]
  • 14. CLASSIFICATION IN (PR): • UNSUPERVISED TRAINING/LEARNING: [10]
  • 15. APPROACHES FOR PATTERN RECOGNITION : [8] TEMPLATE MATCHING: STATISTICAL CLASSIFICATION: SYNTACTIC OR STRUCTURAL MATCHING: ARTIFICIAL NEURAL NETWORKS:
  • 16. TEMPLATE MATCHING : [7] Class Z Class Y Input pattern Class X Z Y Temp X
  • 19. STATISTICAL CLASSIFICATION : [8] Feature Extraction Algorithm X3 X2 X1 Y3 Y1 Y2 Feature Extraction Algorithm Set Of Feature X Set Of Feature Y ...
  • 21.
  • 22. REFERENCES: 1. Dr. C. N. Ravi Kumar, Pattern Classification,(http://uni-mysore.ac) 2. Pradeep Mishra, Pattern Recognition, (http://http://www.slideshare.net) 3. Shyh-Kang Jeng, ―Introduction‖, Pattern recognition Course Website, 2009. [online] Available: http://cc.ee.ntu.edu.tw/~skjeng/PatternRecognition2007.htm. [Accessed Sep. 30, 2009]. 4. Talal A. Alsubaie, SFDA, Pattern Recognition,(http://http://www.slideshare.net) 5. M. Tim Jones, Artificial Intelligence A Systems Approach,page-257 6. Anil K. Jain, Robert P.W. Duin, Introduction to Pattern Recognition1,(citeseerx.ist.psu.edu) 7. , , - ,page 182-191 8. Pattern Recognition and Image Processing, (http://lessons.bdr130.net/1586.html) 9. , Pattern recognition ” 2,1 ,(http://www.teedoz.com) 10. Luís Gustavo Martins, Introduction To Pattern Recognition, (http://http://www.slideshare.net)

Notas del editor

  1. PATTERN : Pattern is a set of physical objects or phenomena or concepts=المفاهيمِ where the elements of the set are similar to one another in certain ways. The Pattern are described by certain quantities, qualities, traits, notable features and so on.[SIT_PC.ppt]
  2. 1.SPATIAL PATTERNS- These patterns are located in space.Eg:- characters in character recognition * images of ground covers in remote sensing * images of medical diagnosis.2.TEMPORAL PATTERN-These are distributed in time.Eg:- Radar signal, speech recognition, sonar signal etc.3.ABSTRACT PATTERNS-Here the patterns are distributed neither in space nor time.Eg:- classification of people based on psychological tests. * Medical diagnosis based on medical history and other medical tests. * Classification of people based on language they speak.[SIT_PC.ppt]
  3. The idea of pattern recognition -Since our early childhood, we have been observing patterns in the objects around us (e.g., toys, flowers, pets, and faces).By the time children are five years old, most can recognize digits and letters. Small and large characters, handwritten and machine printed characters, characters of different colors and orientations and partially occluded letters - all are easily recognized by the young. We take this ability for granted until we face the task of teaching a machine how to recognize the characters. [10.1.1.108.1064.pdf]///////////////////////////////////////////Humans have developed highly sophisticated skills forsensing their environment and taking actions according towhat they observe, e.g.,– Recognizing a face.– Understanding spoken words.– Reading handwriting.– Distinguishing fresh food from its smell.– ...[11848aipresentationonpatternrecognition-111204123817-phpapp02.ppt]
  4. What is pattern recognition -Pattern recognition is the science for how machines can: observing (sensing) the environment, learning to distinguish patterns of interest (e.g., animals) from their background (e.g., sky, trees, ground), and making sound decisions about the patterns (e.g., Fido) or pattern classes (e.g., a dog, a mammal, an animal).[10.1.1.108.1064.pdf]
  5. • Two phase Process1.Training/Learning• Learning is hard and time consuming• System must be exposed to several examples of each class• Creates a “model” for each class• Once learned, it becomes natural2.Detecting/Classifying[introductiontopatternrecognition-110316123301-phpapp02.pdf]//////1-     التعليم : تبدأ بجمع صور لوجوه الموظفين وأخذ مميزات “feature extraction  ”  دقيقة تعتمدها لكل وجه مثل المسافة بين العينين أو أقصى مسافة بين جانبي الوجه أو غيرها من المميزات ومن الممكن أن تختار أكثر من مميز وتدخلها إلى النظام لتصبح كقاعدة بيانات لديه وتخزن بأن س من الناس له هذه المميزات بتلك المقاييس “انظر للصورة في الأسفل “.2-    التمييز : عند وقوف احدهم امام الكميرا للدخول للشركة تلتلقط الكاميرا لوجهه صورة وتعالج بشكل أو بآخر ” من مشكلات الإضاء أو استدارة الرأس  أو القرب والبعد وغيرها – فيما يعرف بـــpreprocessing  “  ” – ومن ثم تقارن هذه المميزات بالمميزات المسجلة في قواعد البيانات فإذا تشابهت إلى حد كبير مع احد المميزات الموجودة فقد تعرف النظام عليها وتكون لأحد الموظفين وإلا فلا  . .. فكر الآن في عدة أمور . . . ما مقدار التشابه بين المميزات هل هناك مقدار بسيط يُسمح به ؟هل عدد الوجوه للشخص الواحد بظروف الإضاءة المختلفة مثلا وتدريب النظام عليها تزيد الدقة ؟هل من الممكن خداع النظام بوجه مشابه ؟كل هذه الأسئلة يجب أن تسأل لاختيار النظام المناسب وتجهيزه بالشكل الذي يراعي الدقة وسولة الاستخدام؟
  6. The design of a pattern recognition system essentially involves the following four aspects:(i) data acquisition and preprocessing, e.g., taking a picture of an object and removing the irrelevant background,(ii) data representation, e.g. deriving relevant object properties (like its size, shape and color) which efficiently offer pertinent information needed for pattern recognition,(iii) training, e.g., imparting pattern class definition into the system, often, by showing a few typical examples of the pattern, and (iv) decision-making that involves finding the pattern class or pattern description of new, unseen objects based on a training set of examples. [10.1.1.108.1064.pdf]
  7. Data acquisition and sensing:Measurements of physical variables.Important issues: bandwidth, resolution , etc.Pre-processing:Removal of noise in data.Isolation of patterns of interest from the background.Feature extraction:Finding a new representation in terms of features.ClassificationUsing features and learned models to assign a pattern to a category.Post-processingEvaluation of confidence in decisions.:[talalalsubaie-1220737011220266-9.ppt]
  8. 1- التدريب ” training” : والهدف منها بناء نموذج يمثل نتائج البيانات التي تم تدريب النظام عليها فكما هو ملاحظ في الصورة ” من اليمين “:ا- جمع البيانات المراد تدريب النظام عليها : نجمع صور الموظفين !ب- القيام بعمليات فلترة أوعمليات اخرى ” مثلا في حال تمييز الوجه ” نعالج مشكلة الإضاءة المختلفة والاستدارة والحجم و غيرها : نأخذ الصور للموظفين بوضعية متشابهة وبنفس البعد عن الكاميرا وبنفس الإضاءة – قدر المستطاع – ونقوم ببعض عمليات الفلترة وتحديد شكل الرأس ونعالج بعض المشاكل إن وجدتج- اختيار الخصائص التي تميز كل عينة كما في الصورة مثلا  ,وقد تستطيع أخذ مميزات تراها أنت دقيقة وهذا ما يجعل تطبيقك مختلفا عن الآخرين فقد تعتمد مميزات للوجه تزيد نسبة الدقة أكثر كما في الــ 3D face recognition التي اعتمدت بالاضافة للميزات السابقة مميزات أخرى تعتمد على المسافة بين ” مثلا ” البعد بين العينين والانف إذا أخذت بالبعد كمحور عمودي على الوجه !فيكون لدينا نموذج يمثل بيانات التدريب للنظام وسيرجع إليه النظام في الخطوة التالية لعمل الـــ “matching”.2-  التمييز “recognition or test” : يتم الآن اختبار النظام وفي هذه الخطوة يتم التعرف على الأشياء وفق هذه الخطوات:ا- ادخال العينة المراد التعرف عليها : مثل صورة الوجه أو الصوت المراد التعرف عليه أو صورة لقزحية العين أو غير ذلك حسب النظامب- القيام بعمليات فلترة أوعمليات اخرى ” مثلا في حال تمييز الوجه ” نعالج نشكلة الإضاءة المختلفة والاستدارة والحجم و غيرها.ج- اختيار الخصائص وانتزاعها من الصورة وفق نفس المعادلة التي طبقت في مرحلة التدريب.د- في هذه المرحلة بالضبط يحدث التعرف والتمييز وذلك بتصنيف العينة التي تم إدخالها إلى الأنواع التي تم تدريب النظام عليها فتحصل مقارنة بين العينة المدخلة وجميع عينات التددريب بطرق بحث وخوارزميات تراعي الدقة والسرعة.
  9. What is a Pattern Class (or category)?– is a set of patterns sharing common attributes– a collection of “similar”, not necessarily identical, objects– During recognition, given objects are assigned to a prescribed class[introductiontopatternrecognition-110316123301-phpapp02.pdf]
  10. Supervised Training/Learning [introductiontopatternrecognition-110316123301-phpapp02.pdf]– a “teacher” provides labeled training sets, used to train a classifier
  11. Unsupervised Training/Learning [introductiontopatternrecognition-110316123301-phpapp02.pdf]– No labeled training sets are provided– System applies a specified clustering/grouping criteria to unlabeled dataset– Clusters/groups together “most similar” objects (according to given criteria)
  12. Template matching:Match with stored template considering translation, rotation and scale changes; measure similarity (correlation) based on training set.Statistical classification:Each pattern is represented in terms of d features(measurements) andViewed as a point in a d-dimensional space. using training sets establish decision boundaries in the feature space - following decision theoretic or discriminate analysis approaches.Syntactic or structural matching: complex pattern is composed of sub-patterns and the relations; they themselves are built from simpler / elementary sub-patterns are called primitives. the patterns are viewed as sentences belonging to a language, primitives are viewed as the alphabet of the language. the sentences are generated according to a grammar. a large collection of complex patterns can be described by a small number of primitives and grammatical rules. the grammar for each pattern class are inferred from the training samples.neural networks:are viewed as weighted directed graphs in which the nodes are artificial neurons and directed edges (with weights) are connections between neurons input-output. neural networks have the ability to learn complex nonlinear input-output relationships from the sequential training procedures, and adapt themselves to input data.
  13. (Template-Matching and Correlation Method) طريقة المطابقة القالبية(templates) مرحلة التعليم في هذه الطريقة تقوم على تخزين مجموعة من القوالبقالب من كل صنف في الحاسوب ويوضح الشكل رقم ( 3) القوالب (prototypes) أو النماذج. [ المدخلة المراد تمييزها مع القوالب المصدر المخزونة داخل الحاسبة [في مرحلة التصنيف تقارن الصورة الداخلةY أكبر من نتيجة مقارنتها مع الصنف X فإن كانت نتيجة مقارنتها مع الصنف (template)حيث تحسب درجة الاختلاف بين الصورتين وتقارن بقيمة X فإنها تصنف ضمن الصنفوإذا تجاوزت درجة الاختلاف قيمة .(threshold value) محددة سلفا تسمى قيمة العتبةالعتبة، فهذا يعني إن الصورتين غير متشابهتين وبعكسه يعني إن الصورتين متشابهتان أومتطابقتان. وقد تم استخدام هذه الطريقة في تمييز الحروف المطبوعة وقراءة صكوك البنوكوهكذا [ 9]. يوضح الشكل رقم ( 4) علمية التصنيف بالحاسبة.تتم عملية المقارنة بإدخال صورة الحرف على شكل مصفوفة ذات بعدين وتقارن معفإذا كانت نسبة الخطأ ضمن منطقة pixel by pixel القوالب المخزونة في ملف بالحاسبة1-,1 ) فان الحرف يطابق هذا الحرف أما إذا كانت اكبر ) (threshold value) مدى العتبةفان الحرف لا يتطابق. أما المطابقة بالنسبة للصور فتؤخذ الصورة المقطع وتقارن معالصورة الكبيرة وكما في الإشكال ( 4,3 ) حيث تجري عملية تمرير الصورة المقطع علىالصورة الكبيرة ومن البداية إلى النهاية باستخدام اللافوف الرياضي إلى إن يتم إيجاد منطقةمشابهه للمقطع أو لا. وتتم عملية المقارنة ببساطة بخزن الصورة الداخلة على شكل مصفوفةثم تقارن مع القوالب الموجودة في الحاسوب وتعطي قيمة للمقارنة [ 2,3,7 ]. تعد هذه الطريقةطريقة سهلة جدًا، والصعوبة الوحيدة في هذه الطريقة هي الاختيار الجيد للقوالب من كلصنف فض ً لا عن تحديد معايير المقارنة وخصوصًا لو كانت الصورة الداخلة تحمل تشوهات!فمث ُ لا لو استخدمنا هذه الطريقة للتعرف على المجرمين، لابد أن نأخذ لكل مجرم عدة لقطاتكي تخزن على جهاز الحاسوب لقطتان جانبيتان واحدة من كل جهة، لقطة أمامية، ولقطتانبزاوية نظر 45 درجة عن الكاميرا. ولكم أن تتخيلوا المساحات التخزينية اللازمة لكل هذهالقوالب !.(correlation) الترابطلمعرفة أو تحديد ما إذا كانت الصورة تحتوي على منطقة مشابهة لمنطقة ما، فانM*N ذات مقياس F(x,y) الترابط هو إحدى الطرق المستخدمة. فمث ً لا لدينا صورة رقميةW(x,y) وإننا نريد أن نحدد ما إذا كانت هذه الصورة تحتوي على منطقة مشابهة لمنطقة ماF(x,y), W(x,y) سنستخدم الترابط بين هاتين المنطقتين K<N و J<M حيث J*K ذات قياسكما في المعادلة الآتية:R(m,n)= ΣΣ − −x yf (x, y).w(x m, y n) .............(1حيثn=0,1,2….N-1, m=0,1,2……M-1نطبق المعادلة رقم ( 1) للحصول على F(x,y) داخل (m,n) من اجل أية قيمة لفي أرجاء منطقة الصورة باستخدام W(x,y) تتحرك n,mومع تغير R قيمة واحدة لR(m,n) بعدئذ تدل القيمة العظمى ل R(m,n) اللافوف الرياضي ونحصل عندئذ على الدالةوان عملية المسح تكون كما F(x,y) مع W(x,y) على الموضع الذي تطابقت فيه الصورتانفي الشكلين التاليين رقم ( 5,6 ) حيث يوضحان عملية المسح للصورة باستخدام اللافوفالرياضي [ 1,6,8 ]. على الرغم من إن دالة الترابط في معادلة ( 1) سهلة في طبيعتها إلا أنهناك دالة ( 2) وهي أكفأ في استعمالها وهي على النحو الآتي:ΣΣ[ ( , ) − ( , )]2.ΣΣ[ ( − , − ) − ]2 ]1/ 2x y x yf x y f x y w x m y n wR(m,n)حيثn=0,1,2….N-1,m=0,1,2……M-1هو متوسط الشدة للقالب (تحسب هذه القيمة مرة واحدة) Ww(x,y) في المنطقة المتطابقة مع f(x,y) هي متوسط القيمة ل f (x,y)تكون ضمن المدى (- 1، 1 ) حيث القيم الموجبة العالية تدل على R(m,n) إن قيمةأن الترابط موجب وعال بين الصورتين والقيم السالبة العالية تدل على أن الترابط سالبوعال بين الصورتين في حين إن القيم القريبة من الصفر تدل على انعدام الترابط بينإذ تكون هذه Source لنافذة معينة من صورة ال R(m,n) الصورتين. يتم حساب قيمةعلى صورة template وبعدها يتم تزحيف صورة ال template النافذة بحجم صورة ال. ( كما في الشكل ( 7 R(m,n) وتحسب قيمة Source ال(convolution) اللافوف الرياضيحيث يتم تحديد إطار للصورة من الصف الأول والأخير والعمود الأول والأخيرووضع النافذة على الصورة ابتداء من العنصر الأول كما في الشكلين التاليين ( 6,5 ) أي إنالطريقة تقوم بمطابقة ثنائية بين عناصر الصورة الكبيرة وعناصر الصورة الصغيرة أو النافذة.ثم نتقدم بالنافذة إلى اليمين وبمقدار عنصر واحد فقط وتعاد عملية مقارنة عناصر النافذة مععناصر الصورة الكبيرة وهكذا حتى نهاية الصف ، بعدها يتم الانتقال إلى الصف التالي لتعاد.[ العملية ، وهكذا إلى نهاية الصورة الكبيرة [ 4
  14. في هذه الطريقة، يوصّف كل pattern بواسطة مجموعة من الخصائص set of features والتي من الممكن أن نعبر عنها بقيم حقيقية. في مرحلة التعلم: يقدّم كل نمط pattern كمتجه من الخصائص feature vector كما توضح الصورة:
  15. أما في مرحلة التعرف او التمييز أو التصنيف، فهذه عادة تتم عن طريق تقسيم مساحة الصورة إلى مناطق مجزأة، كل منطقة تقارن مع صنف كما توضح الصورة:فمثلاً لو كنّا نريد التعرف على صورة تفاحة، ماهي خصائص التفاحة التي نخزنها في مرحلة التعلم؟! هي على سبيل المثال: اللون، الشكل، الدوران، المنطقة السفى، المنطقة العليا.... ألخ. وكذلك يتم التعرف على التفاحة، تقسم الصورة إلى أجزاء وكل جزأ نقارن الخصائص الموجودة فيه مع خصائص الصنف المخزنة وهكذا.الصعوبة هنا هي في اختيار مجموعة الخصائص لكل فئة وقواعد القرار في التعرف على النمط.