Unsupervised texture segmentation using local feature distribution
Two-stage unsupervised local feature distribution clustering algorithm for texture segmentation proposed. Gabor filter coefficients are used to extract texture features. The method is being applied to unsupervised natural texture segmentation and restoration of bi-level images corrupted with correlated noise.
Download file
Counter downloads: 136
Keywords
Authors
| Name | Organization | |
| Gubanov Peter V. | Tomsk State University | peter@tsu.ru |
References
Jaln A.,Farrokfmia F. Unsupervised texture segmentation using Cabor filters//Pattern Recognition. 1991. Vol. 24. № 12.P. 1167-1186.
Hofinann Т., Puzicha J., Buhman J. Unsupervised texture segmentation in a deterministic annealing framework // IEEE Trans, on PAMI. 1998. Vol. 20. № 8. P. 803-818.
Daugman J. Uncertanity relation /or resolution in space, spatial frequency, and orientation optimized by two-dimiensional visual cortical filters // Journal of the optical society Am. A. 1985. Vol. 2. № 7. P. 1160-1169.
Айвазян C.A., Мхитарян B.C. Прикладная статистика и основы эконометрики. М.: ЮНИТИ, 1998. 1022 с.
Ojala Т., Pietikdinen М. Unsupervised texture segmentation using feature distributions // Pattern Recognition. 1999. Vol. 32. № 3.
Brodatz P. Textures: A Photographic album for artists and designers. New York: Dover Publications, 1966.
