Variational methods in image processing, Fall term 2009

During the Fall term 2009, I will lecture a course on variational methods in image processing based on the book Mathematical Problems in Image Processing by Aubert and Kornprobst (2nd ed.). This is an advanced course worth 10 ECTS (=syventävä kurssi, 10 op).

The course consists of two parts. In the first part, I will cover mathematical foundations, including some functional analysis, function spaces, variational methods and PDE. The exact content will be determined by the background of the participants.

In the second part, the aforementioned methods will be put to use in some problems of image processing. I intend to cover at least some models for image restoration as well as some for image segmentation. Additional topics can be covered according to participants interests if there is time.

The course will operate on a "transition to independence" basis, providing participants with the support they need to explore independently a mathematical image processing research article of their choice. The course evaluation consists of a poster produced based on this article. Students in the applied mathematics route can alternatively study the implementation of some method and report that in their poster.

There are lots of options for writing Master's theses (=Pro gradu) on the topic of the course.

Classes: Tuesdays 2:15–4:00 p.m. in room M101, from September 15
Prerequisites: Analysis III
Target audience: 4th-5th year students, graduate students
Evaluation: based on poster which should be handed in by the end of January, 2010. Evaluation criteria will be available at the beginning of the course.
Course material: above mentioned book, articles

Note that the course consists of only one weekly meeting - participants are expected to spend a correspondingly larger share of the time on self-study!

Back to Peter's home-page