The labs cover selected image processing methods. For each method, the students are provided with a Jupyter notebook that contains a brief explanation and a template for the implementation. The students are tasked to follow the template and complete the implementation.
Attendance at the labs is recommended but not mandatory.
See information about the semester work.
Week | Date | Topic |
---|---|---|
1 | 21. 9. | Introduction: B-splines |
2 | 28. 9. | Holiday |
3 | 5. 10. | Active contours |
4 | 12. 10. | Shape transforms |
5 | 19. 10. | Canceled |
6 | 26. 10. | Superpixel segmentation |
7 | 2. 11. | Graph cut |
8 | 9. 11. | Deep learning: U-Net |
9 | 16. 11. | Deep learning: U-net cont. |
10 | 23. 11. | Deep learning: classification |
11 | 30. 11. | Frangi vesselness filter |
12 | 7. 12. | Image registration |
13 | 14. 12. | Image registration with SimpleITK |
14 | 11. 1. | Semester work: presentation |
Although the students can use the lab computers, they are encouraged to use their own devices, which will allow them to work on the assignments at home. The easiest way to install all required software is through the Conda system. Use the Miniconda installer to obtain a Conda installation.
After installing Conda, create a new environment and install the required dependencies:
conda create -n zmo python=3 jupyter matplotlib numpy
To verify the installation, activate the fresh environment
conda activate zmo
and check that the following commands exit successfuly:
python -c "import numpy; import matplotlib" jupyter notebook --version
conda activate zmo
before working on the assignments.