Talk will present an innovative alternative to automatic image parametrization on multiple resolutions, based on texture description with specialized association rules. Image evaluation with machine learning methods will be presented. The algorithm ArTex for parameterizing textures with association rules belonging to structural parametrization algorithms was developed. In order to improve the classification accuracy a multiresolution approach is used. The algorithm ARes for finding more informative resolutions based on the SIFT algorithm will be described. The presented algorithms are evaluated on several public domains and the results are compared to other well-known parametrization algorithms belonging to statistical and spectral parametrization algorithms. Significant improvement of classification results was observed when combining parametrization attributes at several image resolutions for most parametrization algorithms. Our results show that multiresolution image parametrization should be considered when improvement of classification accuracy in textural domains is required. These resolutions have to be selected carefully and may depend on the domain itself.

Reference:
ŠAJN, Luka, KONONENKO, Igor. Multiresolution image parametrization for improving texture classification. EURASIP J. Adv. Signal Process. (Print). [Print ed.], 2008.
Luka Šajn, PhD
Assistant Professor
Luka Šajn was born on 24.10.1977 in Ljubljana, Slovenia. He finished his PhD studies from the University of Ljubljana in 2007 with the thesis "Multi-resolution parametrization for texture classification and its application in analysis of scintigraphy images". His research interests are: multi-resolution pattern parametrization, ischaemic hearth disease diagnosing from scintigraphic images and whole-body bone scintigraphy segmentation. Lately also research in ICF (International Classification of Functioning, Health and Disability / i.e. text mining) and some other medical domains is conducted.
Currently he is a researcher and an assistant professor at the same faculty.