| Type: | Open University
Language: English |
|
|
|
|
|
|
|
|
Prerequisite: 92062
|
|
|
|
Recommended semester: 5th or 6th semester
|
|
Scope and form: Lectures and project work.
|
|
Examination: Evaluation of report and oral presentation (13-scale)
|
|
|
Remarks: This course may be cancelled, therefor always check with the contact person before signing on.
|
| Contact person: | Jørn Winge Bang, Building 451, Tel. +45 4525 5236, email jwb@iae.dtu.dk |
|
|
|
|
Department: Department of Applied Electronics
|
Aim: The objective of the course is to enable the student to:
- design and apply recognition algorithms for automatic recognition of objects in computer images e.g. recognition of errors in industrial products based on images of these,
- analyse, process and recognize patterns from time signals and digital computer images,
- understand classical and modern 1- and 2-dimensional pattern recognition methods, and
- apply pattern recognition methods using modern lab. tools.
|
Contents: Course subtitle: Image Processing and Pattern Recognition.
Introduction to pattern analysis, -processing, and -recognition.
Bayes decision methods. Parameter estimation and supervised learning.
Non-parametric methods. Linear discriminant functions. Unsupervised learning and clustering. Dynamic time warping.
Introduction to Markov-based pattern recognition. Pattern recognition methods.
Automatic speech recognition for machine control. Automatic recognition of objects in medical images.
Automatic recognition of objects in images from industrial production systems.
|
