Colour and rotation invariant textural features based on Markov random fields

A visual appearance of natural materials significantly depends on acquisition circumstances, particularly illumination conditions and viewpoint position, whose variations cause difficulties in the analysis of real scenes. We address this issue with novel texture features, based on fast estimates of Markovian statistics, that are simultaneously rotation and illumination invariant. The proposed features are invariant to in-plane material rotation and illumination spectrum (colour invariance), they are robust to local intensity changes (cast shadows) and illumination direction. No knowledge of illumination conditions is required and recognition is possible from a single training image per material. The material recognition is tested on the currently most realistic visual representation – Bidirectional Texture Function (BTF), using CUReT and ALOT texture datasets with more than 250 natural materials. Our proposed features significantly outperform leading alternatives including Local Binary Patterns (LBP, LBP-HF) and texton MR8 methods.

Research highlights
► Novel colour textural features, which are simultaneously rotation and illumination invariant.
► Generalized for several illumination sources with variable spectra but fixed position.
► Generalized for local intensity changes.
► Outperform leading alternative LBP, LBP-HF, MR8 features.
► Verified on the BTF CUReT, KTH-TIPS2, and ALOT texture datasets and on the Outex database.