MX4074 Pattern Recognition Techniques and Applications Syllabus:

MX4074 Pattern Recognition Techniques and Applications Syllabus – Anna University PG Syllabus Regulation 2021

COURSE OBJECTIVES:

 To understand the fundamentals of Pattern recognition
 To impart knowledge on various clustering techniques
 To study about feature extraction and selection
 To explore different classification models
 To understand Fuzzy Pattern Classifiers and applications

UNIT I PATTERN CLASSIFIER

Overview of Pattern recognition – Discriminant functions – Supervised learning – Parametric estimation – Maximum Likelihood Estimation – Bayesian parameter Estimation – Problems with Bayes approach– Non Parametric techniques, Perceptron Algorithm-LMSE Algorithm- Pattern classification by distance functions – Minimum distance pattern classifier.

UNIT II CLUSTERING

Clustering Concept – Hierarchical Clustering Procedures – Partitional Clustering, k- means algorithm – Clustering of Large Data Sets – EM Algorithm – Grid Based Clustering– Density Based Clustering.

UNIT III FEATURE EXTRACTION AND SELECTION

Entropy Minimization – KL Transforms – Regression-Linear, Non-linear and Logistic, Prediction, Feature Selection through Functions Approximation – Binary Feature Selection

UNIT IV HIDDEN MARKOV MODELS AND SUPPORT VECTOR MACHINE

State Machines – Hidden Markov Models: Maximum Likelihood for the HMM, Forward-Backward Algorithm, Sum and Product Algorithm for the HMM, Extensions of the Hidden Markov Model – Support Vector Machines: Maximum Margin Classifiers, Relevance Vector Machines

UNIT V RECENT ADVANCES AND APPLICATIONS

Elementary Neural Network for Pattern Recognition, Fuzzy pattern classifier, Application of PR in image segmentation, CAD system in breast cancer detection, ECG signal classification, Fingerprint recognition, cell cytology classification

45 PERIODS

PRACTICAL EXERCISES: 30 PERIODS

1. Implementation of Image classification using Perceptron model in Matlab/python.
2. Implementation of Fuzzy pattern classifier in Matlab/OpenCV/python.
3. Implementation of Feature extraction using KL transform in Matlab/OpenCV/python.
4. Implementation of partitional clustering in Matlab/OpenCV/python.
5. Implementation of density based clustering in Matlab/OpenCV/ python
6. Implementation of Classification using SVM in Matlab/OpenCV/python.
7. Implementation of Classification using HMM in Matlab/OpenCV/python.
8. Implementation of Bayes classifier in Matlab/OpenCV/python.
9. Implementation of Classification using Neural Networks in Matlab/OpenCV/python.
10. Implementation of image segmentation in Matlab/OpenCV/python

COURSE OUTCOMES:

On completion of this course the student will be able to:
CO1: Perform classification using Bayes approach
CO2: Implement clustering algorithms for classification
CO3: Perform Feature extraction, feature reduction
CO4: Apply HMM and SVM for real time applications
CO5: Apply pattern recognition techniques for biosignal and medical image applications

TOTAL:75 PERIODS

REFERENCES

1. Andrew Webb, “Statistical Pattern Recognition”, Arnold publishers, London,2002
2. C.M.Bishop, “Pattern Recognition and Machine Learning”, Springer, 2006
3. Earl Gose, Richard Johnsonbaugh Steve Jost, “Pattern Recognition and Image Analysis”, Prentice Hall of India Pvt Ltd., New Delhi, 1996
4. M. Narasimha Murthy and V. Susheela Devi, “Pattern Recognition”, Springer 2011
5. R.O.Duda, P.E.Hart and D.G.Stork, “Pattern Classification”, John Wiley, 2001
6. Robert J.Schalkoff, “Pattern Recognition Statistical, Structural and Neural
Approaches”, John Wiley & Sons Inc., New York, 1992
7. S.Theodoridis and K.Koutroumbas, “Pattern Recognition”, 4th Edition, Academic Press, 2008