ECE 64100 - Digital Imagine Processing II: Model-Based Image and Signal Processing

Course Details

Credits: 3

Areas of Specialization:

  • Communications, Networking, Signal & Image Processing

Counts as:

Normally Offered:

Fall - even years

Catalog Description:

An advanced treatment of the methods in model based signal and image processing including stochastic modeling of multidimensional signals, Bayesian estimation, inverse methods, doubly stochastic models, regularized inversion, the EM algorithm, Bayesian networks, Markov chains, optimization, convexity, majorization techniques, and stochastic simulation. The underlying theory is presented in the context of applications including image restoration, tomographic reconstruction, clustering, classification, and segmentation.

Supplementary Information:

The course presents the basic analytical and algorithmic tools used for processing information from a wide variety of physical sensors and applications ranging from medical CT scanners to speech signals. The basic theme of the course is the formulation of signal processing problems as inverse problems, and the solutions of inverse problems using the techniques of signal and system modeling along with parameter and signal estimation. The course also incorporates a number of computer-based laboratory exercises so that students can better understand how to implement the methods discussed in the class. The course is intended to be accessible to students with a variety of applications backgrounds, but they should have basic familiarity with probability, random variables, and random processes at the level of ECE 60000 or ECE 30200.

Required Text(s):


Recommended Text(s):


Lecture Outline:

Lectures Major Topics
4 Probability, estimation, and random processes
3 Causal Gaussian models
4 Non-causal Gaussian models
4 Image restoration using MAP estimate
4 Continuous non-Gaussian MRF models
5 MAP estimation with non-Gaussian Priors
5 The expectation-maximization (EM) algorithm
4 Markov chains and hidden Markov models
3 Discrete valued Markov random fields (MRF)
3 Stochastic simulation methods
3 Segmentation and MAP estimation with discrete priors

Assessment Method: