Biodiversty and math on TV

TV news enterview shows SCG's research

Multiscale Fractal Descriptors Applied to Texture Classification

João Batista Florindo, Odemir Martinez Bruno

arXiv:1304.1568v1, 2013

This work proposes the combination of multiscale transform with fractal descriptors employed in the classification of gray-level texture images. We apply the space-scale transform (derivative + Gaussian filter) over the Bouligand-Minkowski fractal descriptors, followed by a threshold over the filter response, aiming at attenuating noise effects caused by the final part of this response. The method is tested in the classification of a well-known data set (Brodatz) and compared with other classical texture descriptor techniques. The results demonstrate the advantage of the proposed approach, achieving a higher success rate with a reduced amount of descriptors.