Infopage for IP1: Image Processing (Mo. 10-12h & Th. 12-14h)

This lecture forms the foundation to deepen knowlede in the resarch area of ''Image Processing''. All master students, who are interested in this exciting research area of image prcessing and scene analyiss are invited to join this lecture series. This series also provides a perfect basement for the upcoming lectures of image processing 2 (IP2) in the next semester. Since this lecture is a part of the International Master Course Informatics of the University of Hamburg the language used in this course is English. However, oral examinations can be performed in German or English.
The procedure of this course consists of lectures with integrated exercises.

Topics

• Image processing for multimedia applications (~ 4 weeks):
  • Introduction
  • Digital images and their properties
  • Data structures for image processing
  • Pre-processing
  • Image compression
• Image analysis (~ 6 weeks):
  • Image Segmentation
  • Edge-based Segmentation
  • Shape decription
  • Image morphology
  • Texture analysis
  • Motion analysis
• See and act (~ 3 weeks):
  • 3D image analysis
  • Object recognition
  • Scene analyis
  • Konwledge-based scene interpretation

Lecture slides

The English lecture slides will be made accessible trough this site on each Monday and Thursday before the lectures.

1. Mo 12.10.2015: Introduction
2. Th 15.10.2015: Python Introduction
3. Mo 19.10.2015: Image Understanding and Image Formation
4. Th 22.10.2015: Discussion of exercise 1, Sampling and the frequency domain cont.
5. Mo 26.10.2015: Shape-preserving Sampling and Thresholding
6. Th 29.10.2015: Discussion of exercise 2, Perspective transformations
7. Mo 02.11.2015: Image properties and filters
8. Th 05.11.2015: Image properties and filters cont.
9. Mo 09.11.2015: Spectral image processing and convolution
10. Th 12.11.2015: Discussion of exercise 3, Spectral image processing and convolution cont.
11. Mo 16.11.2015: Image Compression (1)
12. Th 19.11.2015: Image Compression (2)
13. Mo 23.11.2015: Image Segmentation (1)
14. Th 26.11.2015: Image Segmentation (2)
15. Mo 30.11.2015: Image Segmentation (2) cont.
16. Th 03.12.2015: Discussion of exercise 4, Grouping and Searching
17. Mo 07.12.2015: Grouping and Shape Features
18. Th 10.12.2015: Discussion of exercise 5, Grouping and Shape Features cont.
19. Mo 14.12.2015: Project Introduction, Skeletonization and Matching
20. Th 17.12.2015: Pattern recognition
21. Mo 04.01.2016: Decision Theory
22. Th 07.01.2016: Motion Analysis (1), Derivation of Horn and Schuncks Solution (German)
23. Mo 11.01.2016: Motion Analysis (2) and Camera Geometry and 3D Image Analysis
24. Th 14.01.2016: Shape from Shading
25. Mo 18.01.2016: Object Recognition (1)
26. Th 21.01.2016: Object Recognition (2)
27. Mo 25.01.2016: High Level Vision
28. Th 28.01.2016: Project Presentations, A walk through the lecture series

Exercises

The exercises are colocated to the course on a two-weekly cycle, solutions will be discussed in each following week. The exercises consist of five exercise sheets and a small project at the end.
Please send your solutions until Thursday 12 am to Benjamin Seppke.

1. 15.10. - 22.10.2015: Getting started
2. 22.10. - 29.10.2015: Image formation
3. 29.10. - 12.11.2015: Perspective projections
4. 12.11. - 26.11.2015: Histograms, Filters, Convolution and Fourier-Transform
5. 26.11. - 10.12.2015: Image Compression and Segmentation
6. 14.12.2015 - 28.01.2016: Project: Writer identification in historic manuscripts (Project definition, Project dataset)

References

• Richard Szeliski (2010): Computer Vision: Algorithms and Applications (Homepage and eBook download)
• Sonka, Hlaváč und Boyle (3. Auflage 2008): Image Processing, Analysis and Machine Vision, Thomson
• D.A. Forsyth, J. Ponce: Computer Vision, A Modern Approach, Prentice-Hall 2003
• R.C. Gonzalez, R.E. Woods: Digital Image Processing, Prentice-Hall 2001
• B. Jähne (1997): Digitale Bildverarbeitung (4. Auflage), Springer-Verlag

Additional references will be announced in the course.