Invitation to WSPS 6
The Section of Mathematical and Information Sciences of the Association of Hungarian PhD and DLA Students anticipates your attendance at the 6th Winter School of PhD Students in Informatics and Mathematics, which will be held between:
22nd-24th February, 2019 at University of Szeged, Szeged, Hungary.
The co-organisers and the sponsors of the event are John von Neumann Computer Society and the Institute of Informatics of the University of Szeged.
The aim of our winter school is dual:
- improve the multidisciplinary scientific network of PhD students by inviting them to present their work in poster sessions, and
- improve professional skills in an intensive workshop.
Each year, the workshop has a topic relevant for a wide audience. Internationally renowned scientists will give a number of in-depth lectures; these will be accompanied by seminar sessions where attendees will be able to explore topics in an interactive, hands-on way.
This year’s topic is digital image processing.
Poster section and publication
All participants are invited to present their work in the form of posters, be it relevant to the topic of the workshop or to some other area of informatics or mathematics. The printing charge of the submitted posters is included in the registration fee! Each presented poster participates in the competition for the best poster award, which will be accompanied by a remuneration of total 100 000 HUF offered by NJSZT.
All poster abstracts will be published in the conference proceedings, which will have an ISBN number.
Some additional information
Posters and abstracts should be uploaded here after registration.
Accepted formats are:
- for posters: PDF, in A0 size
- for abstracts: LaTeX (accepted template can be downloaded from here)
The registration fee is: 27 000 HUF, which covers all expenses, the printing of the submitted posters and the accommodation for 2 nights.
The meals during the conference are supported by the project “Integrated program for training new generation of scientists in the fields of computer science” №EFOP-3.6.3-VEKOP-16-2017-00002.
Should you have any special requests please contact the organizers directly.
A special price of 17 000 HUF is available for the full members of the Section of Mathematical and Information Sciences of the Association of Hungarian PhD and DLA Students and members of John von Neumann Computer Society (NJSZT).
The deadline for the registration is the
13th January, 2019 27th January, 2019.
Poster and abstract submission deadline are the
20th January, 2019 3rd February, 2019.
Registration is considered complete when the registration fee is transferred to the bank account of the Association of Hungarian PhD and DLA Students.
Address: 1055 Budapest Falk Miksa utca 1.
Tax number: 18397358-1-42
Bank account: 11600006-00000000-62169239
International bank account: HU50 1160 0006 0000 0000 6216 9239
Name of the bank: Erste Bank ZRt.
SWIFT code: GIBAHUHB
The transfer remark should contain the keyword “MITO” and your name.
Please note that places are limited and will be allocated in order of registration.
Day 1. – February 22. Friday:
14:00 – 15:00 Registration
15:00 – 15:30 Opening ceremony
15:30 – 17:00 Medical image processing using deep learning techniques (Dr. László Ruskó)
17:00 – 17:20 Coffee break
17:20 – 19:00 Poster section
19:30 – 21:30 Gala Dinner
Day 2. – February 23. Saturday:
7:00 – 9:00 Breakfast
9:00 – 10:30 Task-oriented Computer Vision in 2D and 3D / 1. (Dr. Csaba Beleznai)
10:30 – 10:50 Coffee break
10:50 – 12:20 Task-oriented Computer Vision in 2D and 3D / 2. (Dr. Csaba Beleznai)
12:20 – 13:30 Lunch
13:30 – 15:00 Cultural program – Visiting the Dóm
15:00 – 16:30 Industrial Image Processing / 1. (Dr. Gábor Németh)
16:30 – 16:50 Coffee break
16:50 – 18:20 Industrial Image Processing / 2. (Dr. Gábor Németh)
18:20 – 19:40 Dinner
Day 3. – February 24. Sunday:
7:00 – 9:00 Breakfast
9:00 – 10:30 Cultural program – Visiting the Computer Exhibition
10:30 – 13:00 Continuous and Discrete Image Reconstruction (Dr. Péter Balázs)
13:00 – 14:00 Lunch
The programme committee reserves the right to modify the schedule.
Dr. Péter Balázs received MSc in Mathematics and Computer Science from the University of Szeged, Hungary, in 1999 and 2000, respectively, and the PhD in Computer Science in 2007. Since 2000 he has been with the Institute of Informatics, University of Szeged. Currently he is Associate Professor at the Department of Image Processing and Computer Graphics. His research interests include Discrete Tomography, Digital Image Analysis, Machine Learning and Pattern Recognition.
Dr. Csaba Beleznai is a Senior Scientist at the Center for Vision, Automation & Control at the AIT Austrian Institute of Technology GmbH. He received his M.S. degree from the Technical University of Ilmenau (Germany) and received his Ph.D. degree in Physics from the Claude Bernard University, Lyon (France) in 1999. His research involves scientific project coordination aiming at complex computer vision systems for applications in demanding context.
From 2007 to 2008 he served as Vice-President of the Austrian Association for Pattern Recognition. Since 2013 he is a member of the governing board of the Hungarian Association for Image Processing and Pattern Recognition.
Dr. Gábor Németh graduated as Software Engineer at University of Szeged in 2007. After graduation he studied digital image processing in Doctoral School of Computer Science, University of Szeged, and earned his PhD degree in 2012. His research interest is image processing, especially feature extraction.
Since 2013 he is an assistant professor at the Department of Image Processing and Computer Graphics, Institute of Informatics, University of Szeged.
Dr. László Ruskó made his MSc in computer science in 2001 at the University of Szeged. After a few years of academic work on discrete tomography and medical image processing he joined the R&D office of GE in Szeged. He was involved in various University collaborations and R&D projects, among which the algorithm development for a liver specialized application (Advantage Workstation) Hepatic VCAR was the most important. After earning his PhD in medical image processing, he joined the GE Healthcare Digital team. Now he is working on image segmentation topics (such as anatomy identification in CT/MR images, organ segmentation) using machine/deep learning techniques. He has experience in CT, MR, PET, US, and XRAY image processing that is demonstrated by several publications and patents.
Continuous and Discrete Image Reconstruction (Dr. Péter Balázs)
Computerized Tomography (CT) was originally a method of diagnostic radiology to obtain the density distribution within the human body based on X-ray projection samples. From a mathematical point of view it seeks to determine an unknown function defined over the 3D Euclidean space from weighted integrals over subspaces, called projections. Since the values of the function can vary over a wide range, a huge number of projections are needed to ensure an accurate reconstruction. In the first part of the lecture we introduce two widely used methods to solve the above mentioned image reconstruction problem.
There are applications where the aim is to reconstruct objects from just a small number of their projections, e.g., in industrial non-destructive quality testing or in Electron Tomography. Here, CT reconstruction methods are no longer successfully applicable. However, there is still a chance to get an accurate reconstruction from just a small number of projections, by exploiting prior knowledge that the range of the image function is discrete and consists of only a small number of known values. This leads us to the field of Discrete Tomography that will be discussed in the second part of the lecture.
During the practical session, we study continuous and discrete methods to reconstruct objects from their projections, under different circumstances.
Task-oriented Computer Vision in 2D and 3D (Dr. Csaba Beleznai)
Applications of Computer Vision slowly make the step towards practical use under demanding real-world conditions. This talk presents considerations from scientific and practitioner’s point of view describing the process how a solution for a given task can be accomplished. These aspects are illustrated by a number of application examples targeting some selected challenging vision problems such as text detection and recognition in cluttered environments, video analytics and 3D vision-based characterization of crowd movement. The talk attempts to provide a look under the hood of these systems by detailing considerations for the algorithmic choice, describing employed algorithmic concepts, presenting relevant implementation details, putting special emphasis on the interplay between Matlab and C++ for a rapid development process and demonstrating numerous results of completed real-time 2D and 3D vision systems.
During the practical part of this presentation we will jointly investigate and solve some 2D/3D analysis tasks in Matlab, for detecting humans in depth data computed from stereo vision image pairs.
Industrial Image Processing (Dr. Gábor Németh)
Industrial Image Processing is special field of digital image processing including the following topics:
- Special Lenses used in industries for image processing.
- Camera Calibration
- Feature point detection, description and matching.
- Epipolar geometry
- 3D reconstruction
- Optical flow
- Visual measurement
In this course we are going to learn the features and usage of special lenses. Then we are going to deal with visual measurement in a lab course. The course consists of 1 hour of lecture and 3 hours experimental lab course.
Medical image processing using deep learning techniques (Dr. László Ruskó)
In the daily routine of medical imaging the number of cases is rapidly increasing which challenges the radiology departments in healthcare. Since the human resourced are limited, there is an increasing demand for automated (pre)processing of medical images. Deep leaning techniques were proved to be efficient in solving various image processing problems such as character recognition, face detection, or photo classification. As result of the technology improvement (in GPU computation power) of the recent years we can train complex deep-learning models for automated processing of large (high-resolution or 3-dimensional) images. Thus, medical image classification (based on pathology) and segmentation (of organs or lesions) became a highlighted research field in computer science. This lecture will present example for a classification and a segmentation problem which were solved using deep learning technique.
|Supported by the project “Integrated program for training new generation of scientists in the fields of computer science”, № EFOP-3.6.3-VEKOP-16-2017-00002. The project has been supported by the European Union and co-funded by the European Social Fund.|
mito [at] dosz [dot] hu