Universität Wien FIND

Due to the COVID-19 pandemic, changes to courses and exams may be necessary at short notice. Inform yourself about the current status on u:find and check your e-mails regularly. Registration is mandatory for courses and exams. Wearing a FFP2 face mask and a valid evidence of being tested, vaccinated or have recovered from an infection are mandatory on site.

Please read the information on studieren.univie.ac.at/en/info.

052600 VU Signal and Image Processing (2021W)

Continuous assessment of course work
MIXED

Registration/Deregistration

Note: The time of your registration within the registration period has no effect on the allocation of places (no first come, first served).

Details

max. 50 participants
Language: English

Lecturers

Classes (iCal) - next class is marked with N

We will adopt a mixed / hybrid format that complements pre-recorded video lectures with in-person review sessions in the lecture hall. New video lectures and tutorials will be made available on Moodle on an ongoing basis, typically on Fridays. These videos form the basis for the review sessions, which will be held in-person in the lecture hall and simultaneously streamed on Moodle via Collaborate sessions. In the review sessions, we will review the most important concepts introduced in the videos and answer any questions you may have. Attendance of the review sessions is not mandatory but *strongly recommended* to understand the material at a depth that is required for passing the course.

Attendance of each review session in person in the lecture hall is limited to 30 students on a first-come-first-serve basis. The 3G rule applies, i.e., you must have a valid negative COVID test, a valid vaccination, or a certificate of recovery to be admitted into the lecture hall.

Tuesday 05.10. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 06.10. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 12.10. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 13.10. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 20.10. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 27.10. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 03.11. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 09.11. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 10.11. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 16.11. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 17.11. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 23.11. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 24.11. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 30.11. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 01.12. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 07.12. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 14.12. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 15.12. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 11.01. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 12.01. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 18.01. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 19.01. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG
Tuesday 25.01. 11:30 - 13:00 Hörsaal 2, Währinger Straße 29 2.OG
Wednesday 26.01. 15:00 - 16:30 Hörsaal 2, Währinger Straße 29 2.OG

Information

Aims, contents and method of the course

Algorithms for data analysis are often based on the assumption of independent and identically distributed (i.i.d) data. The world, however, often violates the first "i", i.e., it generates data with a rich spatial and temporal structure such as time-series and images. Representing, understanding, and processing this structure is the domain of signal processing. As such, a firm grasp of signal processing is essential to understand structure in data and design systems that exploit this structure.

In the first part of this course, we will approach signal processing from the perspective of linear time-invariant (LTI) systems, i.e., we will consider signals as outputs of LTI-systems [1]. This approach will lead us to study the discrete(-time) Fourier transform (D(T)FT) and its applications, including sampling and filter design. In the second part of the course, we will study several variants and extensions of the Fourier transform, including the Hilbert-, Discrete Cosine- and Wavelet transforms. In the third part of the course, we will take an alternative approach to signal processing and consider signals as realizations of stationary stochastic processes [2]. This will lead us to the field of stochastic spectral analysis. We will conclude the course with an introduction to information theory and compression algorithms, e.g., the Lempel-Ziv-Welch (LZW) algorithm that is used in data formats such as ZIP and TIFF.

The lectures are complemented by tutorials, pen & paper exercises and coding assignments on simulated and experimental data to foster a deeper understanding of the topics covered in the lectures.

Assessment and permitted materials

There will be four assignments (two pen & paper assignments and two Pythong coding exercises), one mid-term exam, and one final exam. The various assignments and exams count towards the final grade as follows:

* Assignments: 51%
* Two feedback sheets: 4%
* Midterm: 20%
* Final: 25%

In addition, you can earn up to 10% of bonus points by answering questions on Moodle about the pre-recorded videos prior to each review session.

Minimum requirements and assessment criteria

Prerequisites: StEOP, PR2, MG2, THI, MOD, ADS
Recommended prerequisites: NUM

Grading will be done according to the following scheme:

1. At least 87.5%
2. At least 75.0%
3. At least 62.5%
4. At least 50.0%

In addition, you need at least 10% of the points on each assignment and on each exam to pass the course.

Examination topics

The major goals of this course include:
* Understanding the theory of signals and linear time-invariant systems.
* Becoming familiar with spectral transformations and data compression algorithms.
* Being able to implement common transformations in Python and applying them to time-series and images.

Reading list

1. Alan V. Oppenheim, Ronald W. Schafer, Discrete-Time Signal Processing, 3rd Edition, Pearson, 2010
2. Donald B. Percival, Andrew T. Walden, Spectral Analysis for Physical Applications, Cambridge University Press, 1993
3. Rafael C. Gonzales, Richard E. Woods Digital Image Processing 4th edition, Addison-Wesley, 2018.
4. Boaz Porat, Digital Processing of Random Signals, Dover Publications, 2008.

Association in the course directory

Last modified: Tu 05.10.2021 17:28