Demonstration of image retrieval based on illumination invariant textural MRF features

Content-based image retrieval (CBIR) systems target database images using feature similarities with respect to the query. Our CBIR demonstration utilises novel illumination invariant features, which are extracted from Markov random field (MRF) based texture representations. These features allow retrieving images with similar scenes comprising colour-textured objects viewed with different illumination brightness or spectrum. The illumination invariant retrieval is verified on textures from the Outex database. Categories and Subject Descriptors H.3.3 [Information Storage and Retrieval]: Information Search and Retrieval; I.4.7 [Image Processing and Computer Vision]: Feature Measurement—Texture,Invariants General Terms Algorithms, Verification Keywords Content-based Image retrieval (CBIR), illumination invariants, Markov random fields (MRF), demonstration

• Type: Conference
• Year: 2007
• Where: CIVR
• Authors: Pavel Vacha, Michal Haindl
• Keywords: CIVR 2007, Content-based Image Retrieval, Illumination Invariant Retrieval, Illumination Invariants