SlideShare a Scribd company logo
Ch01_Introduction to Pattern Recognition
   (modified from 패턴인식개론/한학용)
Contents

 01_ Philosophical Debates on AI
 02_ Pattern Recognition (PR)
 03_ Features and Patterns
 04_ Components of PR and Design Cycle
 05_ Category of PR and Classifiers
 06_ Performance Evaluation of PR Algorithms
 07_ Approaches of PR and its Application Areas
 08_ Example of PR Applications




                                                   2
01_Philosophical Debates on AI

 Questions
  Is computer merely a calculating machine?
  Can computer think and understand languages like human?




                                                             3
01_Philosophical Debates on AI

 Positive opinions on the possibility of AI




A.M. Turing(1912~1954)                         Imitation Game


 Negative opinions on the possibility of AI




John Searle(1932~   )                   Chinese Room Arguments
                                                                 4
02_Definition of Pattern Recognition

 What is PR?
  An area of AI that deals with the problems to make computable machines
   (Turing Machines) to recognize certain objects




                                 Cognitive
                                 Science

                                      AI


                                      PR




                                                                            5
03_Features and Patterns

 What is feature?
  Discernible aspects, qualities, characteristics that a certain object has


 What is pattern?
  A set of traits or features of individual objects




                                                                               6
03_Features and Patterns

 Easy features and difficult features




 Categories of patterns




                                         7
04_Components of PR and Design Cycle

 Components of PR System and its Process




                                            8
04_ Components of PR and Design Cycle

 Design steps of PR system

  Step 1 : Data gathering
     Most time-consuming tedious process in PR tasks
     Necessary step to ensure stable PR performance
     For stable performance, we need to consider how many samples are needed before
      the gathering.


  Step 2 : Feature selection
     Essential part regarding PR system’s performance
     We need to decide what features to choose through sufficient prior analysis on the
      object patters.


  Step 3 : Model selection
     To decide what approach (model and algorithm) is to be constructed and applied
     Need prior knowledge on the features
     Need to set up parameters for the model according to the approach




                                                                                           9
04_ Components of PR and Design Cycle

 Step 4 : Learning
    Using the feature sets extracted from the collected data and chosen models, the
     learning algorithm generates or fills up the model (or hypothesis, classifier)
    According to the methods, there are supervised learning, unsupervised learning and
     reinforcement learning.


 Step 5 : Recognition
    Given a new feature set, the generated hypothesis decide a class or category that
     the feature set belongs to.




                                                                                          10
05_ Category of PR and Classifiers

 Categories of problems

  Classification
     In classification problem, the system needs to output one label in a set of finite
      number of labels.


  Regression
     Generalized version of classification
     Through regression, the PR system will return a real value score (usually between 0
      and 1)


  Clustering
     The problem of organizing a small number of multiple groups from a certain set
     The output of clustering system is a set of pairs (example and its class).
     The clustering can be processed in an hierarchical manner such as in phylogenetic
      tree.


  Description
     The problem of expressing an object using a set of a prototype or primitive terms

                                                                                            11
05_ Category of PR and Classifiers

 Classifier
     Most classification task in PR is done by classifiers
     Classification is to partitioning a feature space composed of feature vectors into
      decision regions of nominal classes.
     We call the boundaries of the regions as decision boundaries
     Classification of a feature vector x is to decide what decision region the feature
      vector belongs to, and to assign x to the class that represents the region




                                                                                           12
05_ Category of PR and Classifiers

 Classifier can be represented as a set of discriminant functions



 ∀j=i , if 𝑔 𝑖 𝑥 > 𝑔 𝑗 𝑥 , then we decide that the feature vector 𝑥 ∈ class 𝜔 𝑖




                                                                                   13
06_ Performance Evaluation of PR Algorithms

 Confusion Matrix

                          Actual Positive   Actual Negative
    Predicted Positive          TP                FP
   Predicted Negative           FN                TN


                  TP
 Recall rate =
                TP+FN
                TP
 Precision =
              TP+FP
                                TP
 True Positive Rate (TPR) =
                              TP+FN
                                FP
 False Positive Rate (FPR) =
                              FP+TN




                                                              14
06_ Performance Evaluation of PR Algorithms

 Receiver operating characteristic (ROC) Curve




                                                  15
06_ Performance Evaluation of PR Algorithms

 AUROC
   Ares under the region of ROC Curve
   Closer the curve to top-left corner, more accurate the recognition algorithm
   The performance can be evaluated by the amount of AUROC




                                                                                   16
07_ Approaches of PR and its Application Areas

 Approaches of PR
  Template matching
     Oldest and easiest
     First, prepare the template for the object to compare.
     Normalize the pattern to recognize for matching it with the template.
     And calculate similarity value such as cross-correlation or distance to perform the
      recognition
     Most important task is to prepare the most general template that explains all the samples
      in a certain category.
     Fast running time, but weak in variation of features


  Statistical approaches
       Decide the class of unknown pattern bases on decision boundaries of pattern sets.
       Each of the pattern sets represent a certain class.
       The statistical model of the patterns is a probability density function 𝑃 𝑥|𝑐 𝑖 .
       Learning is a process of creating a probability density function and calculating its
        parameters for each class


  Neural networks
     Model the relation of connection and integration of the biological neurons
     Calculate the response process of neural network for input stimulus
     Classify patterns based on the responses
                                                                                                  17
07_ Approaches of PR and its Application Areas

    Knowledge of the patterns is stored as weights that represent the connection
     strength of synapse.
    Learning is performed similar to biological ways, but the learning process is not a
     serial algorithm.
    The learned knowledge is considered as a black box.
    Minimal need for prior knowledge.
    With sufficient number of neurons, theoretically any complicated decision
     boundaries can be constructed, so this approach is very attractive.


 Structural approaches
    Instead of quantitative features, we consider the relationship among the basic
     prototypes what construct the pattern.
    Examples: Character, Fingerprint, Chromosome




                                                                                           18
07_ Approaches of PR and its Application Areas

 Approaches of PR




                                                   19
07_ Approaches of PR and its Application Areas

 Applications of PR
  Character recognition
     Convert a scanned text image into character codes which can be edited in a
      computer
     Mail classification, Handwriting recognition, Check and banknote recognition,
      License plate recognition




  Biological recognition and human behavioral pattern recognition
     Voice recognition, fingerprint recognition, face recognition, DNA mapping, walking
      pattern analysis and classification, utterance habit analysis and classification


  Diagnostic systems
     Car malfunction, medical diagnostics, EEG, ECG signal analysis and classification, X-
      Ray image pattern recognition


                                                                                              20
07_ Approaches of PR and its Application Areas




 Prediction system
    Weather forecasting based on satellite data, earthquake pattern analysis and
     earthquake prediction, stock price prediction, etc.


 Security and military area
    Intrusion detection based on network traffic pattern analysis, security screening
     system, search and attack of terrorist camp and targets using satellite terrain
     image analysis, radar signal classification, Identification Friend or Foe (IFF)




                                                                                         21
07_ Approaches of PR and its Application Areas


Related Areas                 Application Areas
•Adaptive signal processing   •Image
•Machine learning             processing/segmentation
•Artificial Neural networks   •Computer Vision
•Robotics and Vision          •Speech recognition
•Cognitive science            •Automatic target recognition
•Mathematical Statistics      •Optical character recognition
•Nonlinear optimization       •Seismic Analysis
•Exploratory Data analysis    •Man-machine interaction
•Fuzzy and Genetic System     •Bio recognition (fingerprint,
•Detection and Estimation     vein, iris)
Theory                        •Industrial inspection
•Formal language              •Financial forecast
•Structural modeling          •Medical analysis
•Biological cybernetics       •ECG signal analysis
•Computational neuroscience



                                                               22
08_ Example of PR Applications

 Simple English character recognition system
    feature V : # of vertical lines
    feature H : # of horizontal lines
    feature O : # of slopes
    feature C : # of curves


                      Feature
      Character
                      V   H O C
           L          1   1   0 0
           P          1   0   0 1
           O          0   0   0 1
           E          1   3   0 0
           Q          0   0   1 1




                                                23
08_ Example of PR Applications

 Automatic fish classification (Sea Bass or Salmon)
       A: Conveyor belt for fish
       B: Conveyor belt for classified fish
       C : Robot arm for grabbing fish
       D: Machine vision system with CCD camera
       E : Computer that analyze fish image and control the robot arm




                                                                         24
08_ Example of PR Applications

 Automatic fish classification
     Assume that fish is either salmon or sea bass
     Using machine vision system for acquiring new fish image
     Normalize the intensities of new fish image using image processing algorithm
     Segment fish from the background in the image processing process
     Using the prior knowledge that sea bass is bigger than salmon, extract features in
      the image to measure the length of the new fish
     From the training samples of the two fish categories, calculate the distribution of
      the length, and decide the threshold of decision boundary that minimize the
      classification error




                                                                  Accuracy : 60%




                                                                                            25
08_ Example of PR Applications

 Adding features for enhancing recognition rate
    The accuracy should be over 95% for stable pattern recognition system
    We find that average intensity level is a good feature.




                                                                             26
08_ Example of PR Applications

 Enhancing the recognition rate
    We generate 2 dimensional feature vector with length and average intensity.
    Using a simple linear discriminant function, we enhance the recognition rate.




                                                         Accuracy : 95.7%




                                                                                     27
08_ Example of PR Applications

 Cost vs. Classification Rate
    To minimize the cost, we adjust the decision boundary




                                                             28
08_Example of PR Applications

 Generalization problem
    Using neural network, the performance can be enhanced to 99.9975%
    Is this a good result?




                                                                         29
2013-1 Machine Learning Lecture 01 - Pattern Recognition

More Related Content

PPTX
Feature selection concepts and methods
PPTX
Probabilistic Reasoning
PDF
IRJET- Leaf Disease Detecting using CNN Technique
PDF
12 pattern recognition
PPTX
Convolutional Neural Network and Its Applications
PPTX
Logistic Regression | Logistic Regression In Python | Machine Learning Algori...
PPTX
Ai lecture 03 computer vision
PPTX
Pattern Recognition.pptx
Feature selection concepts and methods
Probabilistic Reasoning
IRJET- Leaf Disease Detecting using CNN Technique
12 pattern recognition
Convolutional Neural Network and Its Applications
Logistic Regression | Logistic Regression In Python | Machine Learning Algori...
Ai lecture 03 computer vision
Pattern Recognition.pptx

What's hot (20)

PPTX
PPT on Data Science Using Python
PPTX
Pattern recognition and Machine Learning.
PDF
Machine Learning Algorithms
PDF
Reinforcement learning
PPTX
Design cycles of pattern recognition
PDF
Bayesian Networks - A Brief Introduction
PDF
Image segmentation with deep learning
PDF
Module 4: Model Selection and Evaluation
PPTX
Artificial Neural Network
PPTX
Convolutional Neural Network (CNN) - image recognition
PPTX
Object detection with deep learning
PDF
Feature selection
PPTX
Image classification with Deep Neural Networks
PDF
Statistical Pattern recognition(1)
PPTX
PPTX
Image classification
PPT
Chapter 10. Cluster Analysis Basic Concepts and Methods.ppt
PDF
Object Detection Using R-CNN Deep Learning Framework
PDF
Machine learning workshop
PPT on Data Science Using Python
Pattern recognition and Machine Learning.
Machine Learning Algorithms
Reinforcement learning
Design cycles of pattern recognition
Bayesian Networks - A Brief Introduction
Image segmentation with deep learning
Module 4: Model Selection and Evaluation
Artificial Neural Network
Convolutional Neural Network (CNN) - image recognition
Object detection with deep learning
Feature selection
Image classification with Deep Neural Networks
Statistical Pattern recognition(1)
Image classification
Chapter 10. Cluster Analysis Basic Concepts and Methods.ppt
Object Detection Using R-CNN Deep Learning Framework
Machine learning workshop
Ad

Viewers also liked (20)

PPT
Pattern recognition
PPTX
Pattern recognition
PPTX
Pattern recognition
PPT
Pattern Recognition
PPT
Lektion 13 april
PPS
N korea65anniv-10-10-10
PPS
張家界之旅
PDF
ザックリとDocker
PPS
Presentation Document Manager
PPTX
零世界
PPT
U40@nagoya3
PPT
High definition performance fitness2
PPT
Secondary academic information session
 
PPTX
London!
PPTX
结婚速成手册
PPT
Power of one full version 2007 1
PDF
About miracles
PDF
G4SI_presentation_02_11_12
PPTX
4th weekly news economic times
PPS
北京故宮台北故宮鎮館之寶
Pattern recognition
Pattern recognition
Pattern recognition
Pattern Recognition
Lektion 13 april
N korea65anniv-10-10-10
張家界之旅
ザックリとDocker
Presentation Document Manager
零世界
U40@nagoya3
High definition performance fitness2
Secondary academic information session
 
London!
结婚速成手册
Power of one full version 2007 1
About miracles
G4SI_presentation_02_11_12
4th weekly news economic times
北京故宮台北故宮鎮館之寶
Ad

Similar to 2013-1 Machine Learning Lecture 01 - Pattern Recognition (20)

PPT
introduction to pattern recognition
PPTX
Artificial intelligence Pattern recognition system
PDF
IRJET- Pattern Recognition Process, Methods and Applications in Artificial In...
PDF
Lect#1_Pattern_Recognition_PGIT204D_By_Dr_TSSinha.pdf
PPT
pattern recognition.ppt
PDF
A Review on Pattern Recognition with Offline Signature Classification and Tec...
PDF
Pattern Matching AI.pdf
PDF
Design principle of pattern recognition system and STATISTICAL PATTERN RECOGN...
PPTX
06 cv mil_learning_and_inference
PDF
Pattern classification via unsupervised learners
PDF
An introduction to Machine Learning
PDF
An introduc on to Machine Learning
PPTX
Machine learning and pattern recognition
PDF
Machine Learning
PPTX
Avin jalal ai & pattern recognition1
PPT
Introduction to pattern recognization
PDF
Applications of Pattern Recognition Algorithms in Agriculture: A Review
PDF
PDF
Lecture4 - Machine Learning
PDF
BTech Pattern Recognition Notes
introduction to pattern recognition
Artificial intelligence Pattern recognition system
IRJET- Pattern Recognition Process, Methods and Applications in Artificial In...
Lect#1_Pattern_Recognition_PGIT204D_By_Dr_TSSinha.pdf
pattern recognition.ppt
A Review on Pattern Recognition with Offline Signature Classification and Tec...
Pattern Matching AI.pdf
Design principle of pattern recognition system and STATISTICAL PATTERN RECOGN...
06 cv mil_learning_and_inference
Pattern classification via unsupervised learners
An introduction to Machine Learning
An introduc on to Machine Learning
Machine learning and pattern recognition
Machine Learning
Avin jalal ai & pattern recognition1
Introduction to pattern recognization
Applications of Pattern Recognition Algorithms in Agriculture: A Review
Lecture4 - Machine Learning
BTech Pattern Recognition Notes

More from Dongseo University (20)

PDF
Linear Model (Summary of Chapter 3 of Machine Learning by Zhou Zhihua)
PDF
Model Evaluation and Model Selection (Summary of Chapter 2 of Machine Learnin...
PDF
Lecture_NaturalPolicyGradientsTRPOPPO.pdf
PDF
Evolutionary Computation Lecture notes03
PDF
Evolutionary Computation Lecture notes02
PDF
Evolutionary Computation Lecture notes01
PDF
Markov Chain Monte Carlo
PDF
Simplex Lecture Notes
PDF
Reinforcement Learning
PDF
Median of Medians
PDF
Average Linear Selection Algorithm
PDF
Lower Bound of Comparison Sort
PDF
Running Time of Building Binary Heap using Array
PDF
Running Time of MergeSort
PDF
Binary Trees
PDF
Proof By Math Induction Example
PDF
TRPO and PPO notes
PDF
Estimating probability distributions
PDF
2018-2 Machine Learning (Wasserstein GAN and BEGAN)
PDF
2018-2 Machine Learning (Linear regression, Logistic regression)
Linear Model (Summary of Chapter 3 of Machine Learning by Zhou Zhihua)
Model Evaluation and Model Selection (Summary of Chapter 2 of Machine Learnin...
Lecture_NaturalPolicyGradientsTRPOPPO.pdf
Evolutionary Computation Lecture notes03
Evolutionary Computation Lecture notes02
Evolutionary Computation Lecture notes01
Markov Chain Monte Carlo
Simplex Lecture Notes
Reinforcement Learning
Median of Medians
Average Linear Selection Algorithm
Lower Bound of Comparison Sort
Running Time of Building Binary Heap using Array
Running Time of MergeSort
Binary Trees
Proof By Math Induction Example
TRPO and PPO notes
Estimating probability distributions
2018-2 Machine Learning (Wasserstein GAN and BEGAN)
2018-2 Machine Learning (Linear regression, Logistic regression)

Recently uploaded (20)

PDF
Microbial disease of the cardiovascular and lymphatic systems
PDF
STATICS OF THE RIGID BODIES Hibbelers.pdf
DOC
Soft-furnishing-By-Architect-A.F.M.Mohiuddin-Akhand.doc
PPTX
GDM (1) (1).pptx small presentation for students
PDF
Chinmaya Tiranga quiz Grand Finale.pdf
PPTX
Tissue processing ( HISTOPATHOLOGICAL TECHNIQUE
PPTX
Introduction-to-Literarature-and-Literary-Studies-week-Prelim-coverage.pptx
PDF
OBE - B.A.(HON'S) IN INTERIOR ARCHITECTURE -Ar.MOHIUDDIN.pdf
PPTX
Cell Structure & Organelles in detailed.
PDF
Weekly quiz Compilation Jan -July 25.pdf
PPTX
Orientation - ARALprogram of Deped to the Parents.pptx
PPTX
PPT- ENG7_QUARTER1_LESSON1_WEEK1. IMAGERY -DESCRIPTIONS pptx.pptx
PPTX
master seminar digital applications in india
PDF
GENETICS IN BIOLOGY IN SECONDARY LEVEL FORM 3
PDF
RMMM.pdf make it easy to upload and study
PPTX
202450812 BayCHI UCSC-SV 20250812 v17.pptx
PDF
Supply Chain Operations Speaking Notes -ICLT Program
PDF
Chapter 2 Heredity, Prenatal Development, and Birth.pdf
PDF
Complications of Minimal Access Surgery at WLH
PDF
RTP_AR_KS1_Tutor's Guide_English [FOR REPRODUCTION].pdf
Microbial disease of the cardiovascular and lymphatic systems
STATICS OF THE RIGID BODIES Hibbelers.pdf
Soft-furnishing-By-Architect-A.F.M.Mohiuddin-Akhand.doc
GDM (1) (1).pptx small presentation for students
Chinmaya Tiranga quiz Grand Finale.pdf
Tissue processing ( HISTOPATHOLOGICAL TECHNIQUE
Introduction-to-Literarature-and-Literary-Studies-week-Prelim-coverage.pptx
OBE - B.A.(HON'S) IN INTERIOR ARCHITECTURE -Ar.MOHIUDDIN.pdf
Cell Structure & Organelles in detailed.
Weekly quiz Compilation Jan -July 25.pdf
Orientation - ARALprogram of Deped to the Parents.pptx
PPT- ENG7_QUARTER1_LESSON1_WEEK1. IMAGERY -DESCRIPTIONS pptx.pptx
master seminar digital applications in india
GENETICS IN BIOLOGY IN SECONDARY LEVEL FORM 3
RMMM.pdf make it easy to upload and study
202450812 BayCHI UCSC-SV 20250812 v17.pptx
Supply Chain Operations Speaking Notes -ICLT Program
Chapter 2 Heredity, Prenatal Development, and Birth.pdf
Complications of Minimal Access Surgery at WLH
RTP_AR_KS1_Tutor's Guide_English [FOR REPRODUCTION].pdf

2013-1 Machine Learning Lecture 01 - Pattern Recognition

  • 1. Ch01_Introduction to Pattern Recognition (modified from 패턴인식개론/한학용)
  • 2. Contents  01_ Philosophical Debates on AI  02_ Pattern Recognition (PR)  03_ Features and Patterns  04_ Components of PR and Design Cycle  05_ Category of PR and Classifiers  06_ Performance Evaluation of PR Algorithms  07_ Approaches of PR and its Application Areas  08_ Example of PR Applications 2
  • 3. 01_Philosophical Debates on AI  Questions  Is computer merely a calculating machine?  Can computer think and understand languages like human? 3
  • 4. 01_Philosophical Debates on AI  Positive opinions on the possibility of AI A.M. Turing(1912~1954) Imitation Game  Negative opinions on the possibility of AI John Searle(1932~ ) Chinese Room Arguments 4
  • 5. 02_Definition of Pattern Recognition  What is PR?  An area of AI that deals with the problems to make computable machines (Turing Machines) to recognize certain objects Cognitive Science AI PR 5
  • 6. 03_Features and Patterns  What is feature?  Discernible aspects, qualities, characteristics that a certain object has  What is pattern?  A set of traits or features of individual objects 6
  • 7. 03_Features and Patterns  Easy features and difficult features  Categories of patterns 7
  • 8. 04_Components of PR and Design Cycle  Components of PR System and its Process 8
  • 9. 04_ Components of PR and Design Cycle  Design steps of PR system  Step 1 : Data gathering  Most time-consuming tedious process in PR tasks  Necessary step to ensure stable PR performance  For stable performance, we need to consider how many samples are needed before the gathering.  Step 2 : Feature selection  Essential part regarding PR system’s performance  We need to decide what features to choose through sufficient prior analysis on the object patters.  Step 3 : Model selection  To decide what approach (model and algorithm) is to be constructed and applied  Need prior knowledge on the features  Need to set up parameters for the model according to the approach 9
  • 10. 04_ Components of PR and Design Cycle  Step 4 : Learning  Using the feature sets extracted from the collected data and chosen models, the learning algorithm generates or fills up the model (or hypothesis, classifier)  According to the methods, there are supervised learning, unsupervised learning and reinforcement learning.  Step 5 : Recognition  Given a new feature set, the generated hypothesis decide a class or category that the feature set belongs to. 10
  • 11. 05_ Category of PR and Classifiers  Categories of problems  Classification  In classification problem, the system needs to output one label in a set of finite number of labels.  Regression  Generalized version of classification  Through regression, the PR system will return a real value score (usually between 0 and 1)  Clustering  The problem of organizing a small number of multiple groups from a certain set  The output of clustering system is a set of pairs (example and its class).  The clustering can be processed in an hierarchical manner such as in phylogenetic tree.  Description  The problem of expressing an object using a set of a prototype or primitive terms 11
  • 12. 05_ Category of PR and Classifiers  Classifier  Most classification task in PR is done by classifiers  Classification is to partitioning a feature space composed of feature vectors into decision regions of nominal classes.  We call the boundaries of the regions as decision boundaries  Classification of a feature vector x is to decide what decision region the feature vector belongs to, and to assign x to the class that represents the region 12
  • 13. 05_ Category of PR and Classifiers  Classifier can be represented as a set of discriminant functions  ∀j=i , if 𝑔 𝑖 𝑥 > 𝑔 𝑗 𝑥 , then we decide that the feature vector 𝑥 ∈ class 𝜔 𝑖 13
  • 14. 06_ Performance Evaluation of PR Algorithms  Confusion Matrix Actual Positive Actual Negative Predicted Positive TP FP Predicted Negative FN TN TP  Recall rate = TP+FN TP  Precision = TP+FP TP  True Positive Rate (TPR) = TP+FN FP  False Positive Rate (FPR) = FP+TN 14
  • 15. 06_ Performance Evaluation of PR Algorithms  Receiver operating characteristic (ROC) Curve 15
  • 16. 06_ Performance Evaluation of PR Algorithms  AUROC  Ares under the region of ROC Curve  Closer the curve to top-left corner, more accurate the recognition algorithm  The performance can be evaluated by the amount of AUROC 16
  • 17. 07_ Approaches of PR and its Application Areas  Approaches of PR  Template matching  Oldest and easiest  First, prepare the template for the object to compare.  Normalize the pattern to recognize for matching it with the template.  And calculate similarity value such as cross-correlation or distance to perform the recognition  Most important task is to prepare the most general template that explains all the samples in a certain category.  Fast running time, but weak in variation of features  Statistical approaches  Decide the class of unknown pattern bases on decision boundaries of pattern sets.  Each of the pattern sets represent a certain class.  The statistical model of the patterns is a probability density function 𝑃 𝑥|𝑐 𝑖 .  Learning is a process of creating a probability density function and calculating its parameters for each class  Neural networks  Model the relation of connection and integration of the biological neurons  Calculate the response process of neural network for input stimulus  Classify patterns based on the responses 17
  • 18. 07_ Approaches of PR and its Application Areas  Knowledge of the patterns is stored as weights that represent the connection strength of synapse.  Learning is performed similar to biological ways, but the learning process is not a serial algorithm.  The learned knowledge is considered as a black box.  Minimal need for prior knowledge.  With sufficient number of neurons, theoretically any complicated decision boundaries can be constructed, so this approach is very attractive.  Structural approaches  Instead of quantitative features, we consider the relationship among the basic prototypes what construct the pattern.  Examples: Character, Fingerprint, Chromosome 18
  • 19. 07_ Approaches of PR and its Application Areas  Approaches of PR 19
  • 20. 07_ Approaches of PR and its Application Areas  Applications of PR  Character recognition  Convert a scanned text image into character codes which can be edited in a computer  Mail classification, Handwriting recognition, Check and banknote recognition, License plate recognition  Biological recognition and human behavioral pattern recognition  Voice recognition, fingerprint recognition, face recognition, DNA mapping, walking pattern analysis and classification, utterance habit analysis and classification  Diagnostic systems  Car malfunction, medical diagnostics, EEG, ECG signal analysis and classification, X- Ray image pattern recognition 20
  • 21. 07_ Approaches of PR and its Application Areas  Prediction system  Weather forecasting based on satellite data, earthquake pattern analysis and earthquake prediction, stock price prediction, etc.  Security and military area  Intrusion detection based on network traffic pattern analysis, security screening system, search and attack of terrorist camp and targets using satellite terrain image analysis, radar signal classification, Identification Friend or Foe (IFF) 21
  • 22. 07_ Approaches of PR and its Application Areas Related Areas Application Areas •Adaptive signal processing •Image •Machine learning processing/segmentation •Artificial Neural networks •Computer Vision •Robotics and Vision •Speech recognition •Cognitive science •Automatic target recognition •Mathematical Statistics •Optical character recognition •Nonlinear optimization •Seismic Analysis •Exploratory Data analysis •Man-machine interaction •Fuzzy and Genetic System •Bio recognition (fingerprint, •Detection and Estimation vein, iris) Theory •Industrial inspection •Formal language •Financial forecast •Structural modeling •Medical analysis •Biological cybernetics •ECG signal analysis •Computational neuroscience 22
  • 23. 08_ Example of PR Applications  Simple English character recognition system  feature V : # of vertical lines  feature H : # of horizontal lines  feature O : # of slopes  feature C : # of curves Feature Character V H O C L 1 1 0 0 P 1 0 0 1 O 0 0 0 1 E 1 3 0 0 Q 0 0 1 1 23
  • 24. 08_ Example of PR Applications  Automatic fish classification (Sea Bass or Salmon)  A: Conveyor belt for fish  B: Conveyor belt for classified fish  C : Robot arm for grabbing fish  D: Machine vision system with CCD camera  E : Computer that analyze fish image and control the robot arm 24
  • 25. 08_ Example of PR Applications  Automatic fish classification  Assume that fish is either salmon or sea bass  Using machine vision system for acquiring new fish image  Normalize the intensities of new fish image using image processing algorithm  Segment fish from the background in the image processing process  Using the prior knowledge that sea bass is bigger than salmon, extract features in the image to measure the length of the new fish  From the training samples of the two fish categories, calculate the distribution of the length, and decide the threshold of decision boundary that minimize the classification error Accuracy : 60% 25
  • 26. 08_ Example of PR Applications  Adding features for enhancing recognition rate  The accuracy should be over 95% for stable pattern recognition system  We find that average intensity level is a good feature. 26
  • 27. 08_ Example of PR Applications  Enhancing the recognition rate  We generate 2 dimensional feature vector with length and average intensity.  Using a simple linear discriminant function, we enhance the recognition rate. Accuracy : 95.7% 27
  • 28. 08_ Example of PR Applications  Cost vs. Classification Rate  To minimize the cost, we adjust the decision boundary 28
  • 29. 08_Example of PR Applications  Generalization problem  Using neural network, the performance can be enhanced to 99.9975%  Is this a good result? 29