1. Rekonstrukce obrazu pomocí grafických modelů
- Author
-
Ficová, Klára, Kazda, Alexandr, and Bulín, Jakub
- Subjects
zpracování obrazu ,Grafický model ,Graphical model ,optimization ,optimalizace ,image processing ,Computer Science::Computer Vision and Pattern Recognition ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION - Abstract
Graphical models represent probability relations among random variables using a graph. They offer an effective way of modelling real life situations and are frequently used in machine learning and statistical thinking. The main goal of this thesis is to describe and implement techniques of denoising an image using graphical models. The graphical model we choose is factorgraph, representing pixels in image and interactions between them as nodes. Using algorithms on the graph, we determine the most probable true image. We also apply those theoretical methods on noisy images and compare the results. 1
- Published
- 2020