Digital signal and video processing

Course description

Image formation (image acquisition, sampling and reconstruction, and perception of visual information); color and color images (physical interpretation of colors, color models, spaces and metrics, comparison and conversion among color spaces); image histograms and information theory; image interpolation and decimation; image visualization; image transformations (intensity and geometrical transformations); image compression; spatial image filtering (with kernel, statistical and morphological filtering); image segmentation (with derivatives – edge detection, thresholding, classification and models, Hough transform, principal component analysis); image analysis in the frequency domain; image filtering in the frequency domain; video processing and analysis (motion detection, analysis and evaluation, motion vector, noise filtering, blotch detection and removal, flicker correction); standards and quality of imaging and video communication services.

Course is carried out on study programme

Elektrotehnika 2. stopnja

Objectives and competences

The course objective is to obtain knowledge in the field of digital image and video processing, including the understanding of digital processing of multidimensional signals, obtaining qualifications for choosing relevant methods for image acquisition, processing and storage, interpretation of colors in digital imaging systems, understanding of the representation of multidimensional signals in the frequency domain and understanding image and video compression algorithms. Practical laboratory work consists of implementing computerized and information-assisted techniques for digital image and video processing.

Learning and teaching methods

During lectures, theoretical aspects of techniques, existing standards and established methods are given, which are additionally supported by descriptions of practical examples from different fields of application. During laboratory practice, techniques for computerized image processing and analysis are developed and implemented.

Intended learning outcomes

With successful course completion, the students should be able to:

  • define the image and video as a multidimensional signal;
  • compare different image and video visualization techniques and their representation in color spaces;
  • differentiate among different aspects of image and video sampling (pixel, spatial, spectral, temporal and radiometric resolution);
  • differentiate among different standards for image and video compression in terms information theory;
  • choose an adequate method for image filtering in spatial or frequency domain for the purpose of the given application;
  • use different techniques for image transformation (intensity transformations, geometrical transformations);
  • develop simplified computer algorithms for problem solving in the field of image and video processing;
  • evaluate existing and new approaches for image and video processing.

Reference nosilca

  1. Bulat Ibragimov, Robert Korez, Boštjan Likar, Franjo Pernuš, Lei Xing in Tomaž Vrtovec. Segmentation of pathological structures by landmark-assisted deformable models. IEEE Transactions on Medical Imaging, 36(7):1457-1469, 2017. [doi:10.1109/TMI.2017.2667578] [FV: 3.942 (2016); 9/105 computer science, interdisciplinary applications; četrtina]
  2. Ching-Wei Wang, Cheng-Ta Huang, Jia-Hong Lee, Chung-Hsing Li, Sheng-Wei Chang, Ming-Jhih Siao, Tat-Ming Lai, Bulat Ibragimov, Tomaž Vrtovec, Olaf Ronneberger, Philipp Fischer, Tim F. Cootes in Claudia Lindner. A benchmark for comparison of dental radiography analysis algorithms. Medical Image Analysis, 31:63-76, 2016. [doi:10.1016/] [FV: 4.188 (2016); 8/105 computer science, interdisciplinary applications; 1. četrtina]
  3. Robert Korez, Bulat Ibragimov, Boštjan Likar, Franjo Pernuš in Tomaž Vrtovec. A framework for automated spine and vertebrae interpolation-based detection and model-based segmentation. IEEE Transactions on Medical Imaging, 34(8):1649-1662, 2015. [doi:10.1109/TMI.2015.2389334] [FV: 390 (2014); 18/249 engineering, electrical & electronic; 1. četrtina]
  4. Bulat Ibragimov, Boštjan Likar, Franjo Pernuš in Tomaž Vrtovec. Shape representation for efficient landmark-based segmentation in 3D. IEEE Transactions on Medical Imaging, 33(4):861-874, 201 [doi:10.1109/TMI.2013.2296976] [FV: 3.390 (2014); 18/249 engineering, electrical & electronic; 1. četrtina]
  5. Bulat Ibragimov, Boštjan Likar, Franjo Pernuš in Tomaž Vrtovec. A game-theoretic framework for landmark-based image segmentation. IEEE Transactions on Medical Imaging, 31(9):1761-1776, 2012. [doi:10.1109/TMI.2012.2202915] [FV: 4.027 (2012); 8/242 engineering, electrical & electronic; 1. četrtina]

Study materials

  1. R.C. Gonzalez, R.E. Woods: Digital Image Processing, Pearson Prentice Hall, 3. izdaja, 2008.
  2. B. Likar: Biomedicinska slikovna informatika in diagnostika, Založba FE in FRI, 2008.
  3. W. Burger, M.J. Burge: Principles of Digital Image Processing: Fundamental Techniques, Springer, 2009.
  4. W. Burger, M.J. Burge: Principles of Digital Image Processing: Core Algorithms, Springer, 2009.
  5. W. Burger, M.J. Burge: Principles of Digital Image Processing: Advanced Methods, Springer, 2013.

Bodi na tekočem

Univerza v Ljubljani, Fakulteta za elektrotehniko, Tržaška cesta 25, 1000 Ljubljana

E: T:  01 4768 411