Engineering 252

Pattern Recoginition and Computer Vision

Text: Pattern Classification (2nd ed.) by R.O. Duda, P.E. Hart, D.G. Stork

Plus some application material on Computer Vision for illustrating the
theory and methodology.
Grade Basis:
  1. Midterm Examination
  2. Final Examination
  3. Two MATLAB Projects

Homework Problems: 7 assignments

Topics

  1. Pattern representation by features and stochastic processes.
  2. Bayesian classification theory.
  3. Supervised learning: parametric estimation and recognition.
  4. Supervised learning: nonparametric estimation and learning 
  5. (e.g. nearest neighbor methods).
  6. Unsupervised learning: parametric estimation and recognition.
  7. Unsupervised learning, nonparametric: clustering and recognition.
  8. Artificial neural networks.
  9. Learning theory, complexity, and high dimensional feature spaces.

Computer Vision Applications To:

  1. Texture recognition.
  2. Stereo reconstruction of 3D surfaces.
  3. 2D and 3D shape modelling and recognition using algrebraic curves and surfaces.
Prerequisite:
A senior level undergraduate course in statistics or probability. Some linear algebra.


Instructor: David B. Cooper
Office: Barus-Holley, Room 224
Phone: 863-2601
E-mail: cooper@lems.brown.edu