ML Anthology

Are Multifractal Multipermuted Multinomial Measures Good Enough for Unsupervised Image Segmentation?

Lui Kam, Jacques Blanc-Talon
CVPR 2000 pp. 1058-1063
doi:10.1109/CVPR.2000.855799 /cvpr/2000/kam2000cvpr-multifractal/

Abstract

By extending multinomial measures, a new class of self-similar multifractal measures is developed for texture representation. Two multifractal features have been shown to be suitable for texture discrimination and classification. Their use within a supervised segmentation framework provides us with satisfactory results. In this paper we complete the survey on these features by showing their rotation invariant property and their scaling behaviour. Both properties are particularly important for analyzing aerial images because the geographical elements can appear in different orientations and scales. Then, an automatic clustering algorithm based on a watershed technique is used for the segmentation of real world images. The experimental results are encouraging.

CVPR Semantic Scholar

Cite

Text

Kam and Blanc-Talon. "Are Multifractal Multipermuted Multinomial Measures Good Enough for Unsupervised Image Segmentation?." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2000. doi:10.1109/CVPR.2000.855799

Markdown

[Kam and Blanc-Talon. "Are Multifractal Multipermuted Multinomial Measures Good Enough for Unsupervised Image Segmentation?." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2000.](https://mlanthology.org/cvpr/2000/kam2000cvpr-multifractal/) doi:10.1109/CVPR.2000.855799

BibTeX

@inproceedings{kam2000cvpr-multifractal,
  title     = {{Are Multifractal Multipermuted Multinomial Measures Good Enough for Unsupervised Image Segmentation?}},
  author    = {Kam, Lui and Blanc-Talon, Jacques},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2000},
  pages     = {1058-1063},
  doi       = {10.1109/CVPR.2000.855799},
  url       = {https://mlanthology.org/cvpr/2000/kam2000cvpr-multifractal/}
}