Local Multiple Orientation Estimation: Isotropic and Recursive Oriented Network
Résumé
In this paper we propose a new operator for texture orientation estimation. We focus on directional textures which can have more than a single orientation at the same point. Our operator consists in a steerable network of paral-lel lines along which a homogeneity feature is computed in the spatial domain. The analysis of the network re-sponse along each direction allows us to set up the pres-ence of single or multiple orientations. In order to reduce the computing cost, we propose a recursive implementation of our operator, thanks to the rotations of the image instead of the rotation of the net-work. Our operator works on a small support, and thus pro-vides a local estimation of the orientations. Results ob-tained both with synthetic textures and natural images are accurate and show the selectivity and the isotropic behav-iour of our operator.