Extended visual appearance texture features

Simon-Frédéric Desage (1) , Gilles Pitard (1) , Maurice Pillet (1) , Hugues Favrelière (1) , Jean-Luc Maire (1) , Fabrice Frelin (1) , Serge Samper (1) , Gaëtan Le Goïc (2)

1 SYMME - Laboratoire SYstèmes et Matériaux pour la MEcatronique
2 Le2i - Laboratoire Electronique, Informatique et Image [UMR6306]

The research purpose is to improve surface characterization based on what is perceived by human eye and on the 2006 CIE report. This report denes four headings under which possible measures might be made: color, gloss, translucency and texture. It is therefore important to dene parameters able to discriminate surfaces, in accordance with the perception of human eye. Our starting point in assessing a surface is the measurement of its reectance (acquisition of ABRDF for visual rendering), i.e. evaluate a set of images from dierent angles of lighting rather than a single image. The research question is how calculate, from this enhanced information, some discriminating parameters. We propose to use an image processing approach of texture that reects spatial variations of pixel for translating changes in color, material and relief. From a set of images from dierent angles of light, we compute associated Haralick features for constructing new (extended) features, called Bidimensional Haralick Functions (BHF), and exploit them for discriminating surfaces. We propose another framework in three parts such as color, material and relief.

Mots clés: BHF; Haralick features; visual inspection; BTF; ABRDF; Texture; Soft metrology; surface characterization

Domaines: Mécanique [physics.med-ph]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Simon-Frédéric Desage, Gilles Pitard, Maurice Pillet, Hugues Favrelière, Jean-Luc Maire, et al.. Extended visual appearance texture features. Measuring, Modeling, and Reproducing Material Appearance, Feb 2015, San Francisco, CA, United States. ⟨10.1117/12.2081622⟩. ⟨hal-02136855⟩