Texture analysis by multi-resolution fractal descriptors
This is an incremental improvement for texture analysis in computer vision.
The authors tackled texture classification by proposing a multi-resolution fractal descriptor based on Bouligand-Minkowski measures, achieving better results than classical and state-of-the-art methods like Gabor-wavelets and co-occurrence matrix on Brodatz and Vistex datasets.
This work proposes a texture descriptor based on fractal theory. The method is based on the Bouligand-Minkowski descriptors. We decompose the original image recursively into 4 equal parts. In each recursion step, we estimate the average and the deviation of the Bouligand-Minkowski descriptors computed over each part. Thus, we extract entropy features from both average and deviation. The proposed descriptors are provided by the concatenation of such measures. The method is tested in a classification experiment under well known datasets, that is, Brodatz and Vistex. The results demonstrate that the proposed technique achieves better results than classical and state-of-the-art texture descriptors, such as Gabor-wavelets and co-occurrence matrix.