1. komponenta

Lecture type	Total
Lectures	45

* Load is given in academic hour (1 academic hour = 45 minutes)

COURSE AIMS AND OBJECTIVES:
This is an introductory course in inverse problems using techniques of statistical inversion theory. Prototype problems will be in image processing with tasks like image denoising, image reconstruction and such.

COURSE DESCRIPTION AND SYLLABUS:

1. Inverse problems and a model of measurement
2. Singular value decomposition
3. Randomised algorithms and low rank approximations
4. Convex optimisation
5. Sparse data representation
6. Regularization and LASSO
7. Bayesian inversion
8. Statistical inversion
9. Methods of deep learning
10. Fourier and Radon transform
11. Applications in image processing (Roentgen CT)

1. Statistical and Computational Inverse Problems, Jari Kaipio, Erkki Somersalo, Springer, 2005.
2. Inverse Problem Theory and Methods for Model Parameter Estimation (http://www.ipgp.fr/~tarantola/Files/Professional/Books/InverseProblemTheory.pdf), Albert Tarantola.
3. Introduction to Bayesian Scientific Computing, D. Calvetti, E. Somersalo, Springer, 2007.
4. Compressive Imaging: Structure, Sampling, Learning, Ben Adcock, Anders Hansen.

Ostali izborni predmeti - Regular study - Computer Science and Mathematics
Vezani kolegiji C - Regular study - Computer Science and Mathematics
Vezani kolegiji D - Regular study - Computer Science and Mathematics

Ostali izborni predmeti - Regular study - Computer Science and Mathematics
Vezani kolegiji C - Regular study - Computer Science and Mathematics
Vezani kolegiji D - Regular study - Computer Science and Mathematics