course: Statistical Pattern Recognition

number:
148228
teaching methods:
lecture with integrated tutorials
responsible person:
Prof. Dr.-Ing. Aydin Sezgin
lecturer:
Prof. Dr. Mireille Boutin (extern)
language:
english
HWS:
2
CP:
3
offered in:

goals

The students have learned the fundamental principles of statistical pattern recognition along with the relationship between statistical pattern recognition methods and other methods of pattern recognition.

content

  • Introduction to Statistical Pattern Recognition
  • Discriminant Functions and Decision Hypersurfaces
  • Bayes' Decision Rule
  • Discriminant Functions for Normally Distributed Feature Vectors
  • Density Estimation and Pattern Recognition using Maximum Likelihood Estimation
  • Density Estimation and Pattern Recognition using Parzen Windows
  • Density Estimation and Pattern Recognition using the K-Nearest Neighbors
  • Density Estimation and Pattern Recognition using the Nearest Neighbor

requirements

none

recommended knowledge

  • Familiarity with MATLAB or some other programming language such as C
  • Familiarity with probability theory for discrete and continuous random variables