Finnish Inverse Problems Summer School 2019
(Summer School on Very Finnish Inverse Problems)
University of Helsinki, Finland, June 3-7, 2019
The Finnish Centre of Excellence in Inverse Modelling and Imaging and Finnish Inverse Problems Society (FIPS) are proud to organize a Summer School on inverse problems on June 3-7, 2019.
The event will take place at the Kumpula Campus of University of Helsinki, Finland.
Monday and Tuesday sessions were in hall CK112 in the bottom floor of the Exactum building.
Wednesday, Thursday and Friday sessions will be in hall "Brainstorm" in the bottom floor of the Dynamicum building. The address is Erik Palménin Aukio 1. All registered participants will get a nametag in the lobby of Dynamicum.
The content is primarily intended for PhD students and postdoctoral researchers in mathematics or physics working in the field of inverse problems.
Slides of Samuli Siltanen (PDF 12 MB)
Slides of Matti Lassas (PDF 43 MB)
Tatiana Bubba (University of Helsinki): X-ray tomography (slides I, PDF 2.1 MB) (slides II, PDF 700 KB) (slides III, PDF 12 MB)
Janne Hakkarainen (Finnish Meteorological Institute): Kalman filter (slides, codes and data, ZIP 379 MB)
Ville Kolehmainen (University of Eastern Finland): Bayesian inversion and MCMC (slides, PDF 3.8 MB) (exercise, PDF 100 KB)
Marko Laine (Finnish Meteorological Institute): Markov Chain Monte Carlo (MCMC) methods (slides, PDF 3 MB) (exercises, PDF 6 KB)
Mikko Salo (University of Jyväskylä): Applications of microlocal analysis to inverse problems (slides, PDF 364 KB)
Hanne Kekkonen (University of Cambridge): Bayesian inverse problems and uncertainty quantification (slides, PDF 2.3 MB)
Matti Lassas (University of Helsinki): Geometric inversion (slides, PDF 4 MB)
Aku Seppänen (University of Eastern Finland): Electrical Impedance Tomography, inverse problems, material characterization & structural health monitoring (slides, PDF 2.3 MB)
Samuli Siltanen (University of Helsinki): X-ray tomography (slides, PDF 24 MB)
Antti Solonen (Leanheat Oy, LUT University): Hierarchical Bayesian models for marine vessels
Johanna Tamminen (Finnish Meteorological Institute): Satellite remote sensing (slides, PPT 50 MB)
Tanja Tarvainen (University of Eastern Finland): Tomographic imaging with light and sound (slides, PDF 2.4 MB)
To register, please send email to summer.school "at" fips.fi. Registration fee is 60 euros. The registration fee can be paid in cash at the beginning of the course, or to the following bank account:
Account owner: Suomen inversioseura
IBAN number: FI72 1012 3000 2079 18
SWIFT/BIC Code: NDEAFIHH
Reference number: 600 4
Deadline for registration: April 30, 2019.
Schedule for the week June 3-7, 2019
Information for students
You can earn 3 credit points of the summer school. In order to obtain the credits, you need to attend the lecturers and exercises and write a learning diary (2-3 pages). The learning diaries can be written in English or Finnish. Send the learning diaries in pdf-form to Tanja Tarvainen (tanja.tarvainen "at" uef.fi). Deadline for returning the diary is June 16, 2019. Remember to write clearly your name, email address and the name of the university where you are conducting your studies, so that we are able to send a sertificate of your work with correct information.
Data collection and reconstruction workshop, June 10-12
In the week after the summer school you have a chance to measure your own X-ray tomography data and try the methods you learned in practice.
Schedule of the data collection workshop
10-12 Introduction to measuring real X-ray data, hall Exactum B222
Samuli Siltanen introduces the mathematical steps of data calibration (slides, PDF 5.4 MB)
Alexander Meaney adds practical details
12-13 Lunch (at own cost)
13-15 Introduction to FIPS open datasets
Alexander Meaney describes the Lotus and Walnut datasets, including the construction of the system matrix with the ASTRA toolbox
Salla Latva-Äijö describes time-dependent datasets
15-16 First visit to the lab
Alexander Meaney hosts two groups at the lab, half hour for each group. Recording of a single X-ray image.
The other group may discuss with Samu and Salla meanwhile.
16-17 Discussion of individual needs and wishes of the participants. Based on the discussion, division into two groups.
Download the two projection images Key.tif (10 MB) and toy.tif (10 MB) by clicking these thumbnail images:
The related Matlab codes are here: bin_projection.m, yoda_reconstructions_20190610.m, yoda_sinogram_generation_20190610.m.
Tomographic measurement sessions. First group at 9-12 and second group at 13-16.
Hall Exactum B222 is available as working space 9-16.
Computational "homework" for both groups for the time outside the lab:
Run the Matlab codes in the blog post Simple simulation of X-ray tomography.
Run the Matlab codes in the blog post Total variation regularization for X-ray tomography.
Run the Matlab codes (Tikhonov regularization and approximate total variation regularization) on the page Matrix-free X-ray tomography.
The codes on the third page above are applicable to the measured data by replacing "radon.m" by multiplication with matrix A and replacing "iradon.m" by multiplication by the transpose of matrix A. However, remember that radon.m and iradon.m take images as arguments, whereas multiplication by matrices requires dropping the images into vertical vectors.
Reconstruction workshop 9-17.
Let's work together on reconstructions from the measured data. There are two working spaces available:
Computer classroom Physicum D210. Ask for access to University of Helsinki computing resources if needed.
Organising committee (Summer School)
Matti Lassas, University of Helsinki
Samuli Siltanen, University of Helsinki
Johanna Tamminen, Finnish Meteorological Institute
Tanja Tarvainen, University of Eastern Finland
Organising committee (Data collection and reconstruction workshop)
Tatiana Bubba, University of Helsinki
Samuli Siltanen, University of Helsinki
Alexander Meaney, University of Helsinki
Salla Latva-Äijö, University of Helsinki
Last updated on Mar 5, 2019.