česky

Appearance of real scenes is highly dependent on actual conditions as illumination and viewpoint, which significantly complicates automatic analysis of images of such scenes. In this thesis, we introduce novel textural features, which are suitable for robust recognition of natural and artificial materials (textures) present in real scenes. These features are based on efficient modelling of spatial relations by a type of Markov Random Field (MRF) model and we proved that they are invariant to illumination colour, cast shadows, and texture rotation. Moreover, the features are robust to illumination direction and degradation by Gaussian noise, they are also related to human perception of textures.

The features were favourably tested on current textural databases (Outex, Bonn BTF, CUReT, ALOT, and KTH-TIPS2), where they outperformed state of the art methods (including opponent Gabor features, LBP, LBP-HF, and MR8-LINC) in almost all experiments, e.g. the results on ALOT dataset was improved by 20%. We applied our features in construction of a content-based tile retrieval system, optimisation of texture compression parameters in accordance with human perception, invariant segmentation of multimodal textures, and recognition of glaucomatous tissue in retina images. The presented methods can improve in existing CBIR systems or they can be utilised in domain specific CBIR systems focused on structural/textural similarity. Other applications include computer vision, since the analysis of real scenes often requires a recognition of textures in variable conditions.

The thesis won Antonin Svoboda prize 2012 for the best PhD thesis in computer sciences and cybernetics awarded by Czech Society for Cybernetics and Informatics and got into finalists of European award from promising yound researchers in computer science and applied mathematics - ERCIM Cor Baayen Award 2013.