IF4094 Pattern Recognition Syllabus:
IF4094 Pattern Recognition Syllabus – Anna University PG Syllabus Regulation 2021
COURSE OBJECTIVES:
Understand the in-depth concept of Pattern Recognition
Implement Bayes Decision Theory
Understand the in-depth concept of Perception and related Concepts
Understand the concept of ML Pattern Classification
Understand the concept of DL Pattern Recognition
UNIT I PATTERN RECOGNITION
Induction Algorithms. Rule Induction. Decision Trees. Bayesian Methods. Overview. NaiveBayes. The Basic Na¨ıve Bayes Classifier. Naive Bayes Induction for Numeric Attributes. Correction to the Probability Estimation. Laplace Correction. No Match. Other Bayesian Methods. Other Induction Methods. Neural Networks. Genetic Algorithms. Instance-based Learning. Support Vector Machines.
UNIT II STATISTICAL PATTERN RECOGNITION
About Statistical Pattern Recognition. Classification and regression. Features, Feature Vectors, and Classifiers. Pre-processing and feature extraction. The curse of dimensionality. Polynomial curve fitting. Model complexity. Multivariate non-linear functions. Bayes’ theorem. Decision boundaries. Parametric methods. Sequential parameter estimation. Linear discriminant functions. Fisher’s linear discriminant. Feed-forward network mappings.
UNIT III BAYES DECISION THEORY CLASSIFIERS
Bayes Decision Theory. Discriminant Functions and Decision Surfaces. The Gaussian Probability Density Function. The Bayesian Classifier for Normally Distributed Classes. Exact interpolation. Radial basis function networks. Network training. Regularization theory. Noisy interpolation theory. Relation to kernel regression. Radial basis function networks for classification. Comparison with the multi-layer perceptron. Basis function optimization.
UNIT IV LINEAR DISCRIMINANT FUNCTIONS
Linear Discriminant Functions and Decision Surfaces. The Two-Category Case. The Multicategory Case. The Perceptron Criterion Function. Batch Perceptron. Perceptron Algorithm Convergence. The Pocket Algorithm. Mean Square Error Estimation. Stochastic Approximation and the LMS Algorithm. Convergence Proof for Single-Sample Correction. Fixed increment descent. Some Direct Generalizations. Fixed increment descent. Batch variable increment Perceptron. Balanced Winnow algorithm. Relaxation Procedures. The Descent Algorithm.
UNIT V NONLINEAR CLASSIFIERS
The Two Layer Perception. The Three Layer Perception. Algorithms Based On Exact Classification Of The Training Set. Feedforward operation and classification. General feedforward operation. Expressive power of multilayer networks. Backpropagation algorithm. Network learning. Training protocols. Stochastic Backpropagation. Batch Backpropagation. Radial basis function networks (RBF). Special bases. Time delay neural networks (TDNN). Recurrent networks. Counter propagation. Cascade-Correlation. Cascade-correlation. Neocognitron
SUGGESTED ACTIVITIES:
1: Car Sales Pattern Classification using Support Vector Classifier
2: Avocado Sales Pattern Recognition using Linear regression
3: Tracking Movements by implementing Pattern Recognition
4: Detecting Lanes by implementing Pattern Recognition
5: Pattern Detection in SAR Images
COURSE OUTCOMES:
CO1: Discover imaging, and interpretation of temporal patterns
CO2: Identify Structural Data Patterns
CO3: Implement Pattern Classification using Machine Learning Classifiers
CO4: Implement Pattern Recognition using Deep Learning Models
CO5: Implement Image Pattern Recognition
TOTAL:45 PERIODS
REFERENCES
1. Pattern Classification, 2nd Edition, Richard O. Duda, Peter E. Hart, and David G. Stork. Wiley, 2000
2. Pattern Recognition, Jürgen Beyerer, Matthias Richter, and Matthias Nagel. 2018
3. Pattern Recognition and Machine Learning, Christopher M. Bishop. Springer, 2010
4. Pattern Recognition and Classification, Dougherty, and Geoff. Springer, 2013
5. Practical Machine Learning and Image Processing, Himanshu Singh. Apress, 2019