The Pattern Recognition department of the Institute of Information Theory and Automation of the Czech Academy of Sciences is a basic research unit oriented mainly to statistical pattern recognition and computer vision areas. The emphasis is being put on finite mixtures, modelling of Markov random fields for scene interpretation, physically correct visualization, and visual data restoration. The model-based pattern recognition makes use of new theoretical results from probabilistic neural networks, statistical feature selection, unsupervised segmentation, and the illumination invariants theory.
Department theoretical results can be categorized into the following thematic areas: