| Recommended semester: 7th - 9th semester |
| Scope and form: Lectures, mandatory Matlab computer exercises and a miniproject. |
Evaluation: Oral exam
Approval of compulsory coursework is a prerequisite for taking part in the exam. |
| Examination: 13-scale |
| Previous course: 04364 |
| Prerequisites: 02451 / 04361 / 04362 / C4232 |
| Aim: To provide the student with a basis for developing algorithms and planning digital systems with emphasis on non-linear signal processing. The signals can be of many types, but the course will mainly deal with speech signals and biomedical signals. |
| Contents: The lectures will be chosen among the following topics: Neural Networks. An introduction to the theory of neural networks is given. The theory will be illustrated by applications in the areas of multimedia, bio-medicine, and datamining. Methods for recognition of speech, especially methods based on Hidden Markov Models and / or neural networks. Signal detection and pattern recognition. Bayesian decision theory and mathematical modelling of pattern recognition systems. |
| Remarks: This course forms together with course 02455 advance courses in the area of digital signal processing. The corresponding introductory courses are 02451, 02453. |
| Contact: Lars Kai Hansen, building 321, (+45) 4525 3889, lkh@imm.dtu.dk |
| Department: 002 Informatics and Mathematical Modelling |
| Course URL: http://www.imm.dtu.dk/courses/02457 |
| Keywords: Neural networks, hidden Markiv Models, pattern recognition, multimedia signal processing |
| Updated: 16-08-2001 |
