Advanced search

Unsupervised texture segmentation and labeling using biologically inspired features

Gaëtan Martens (UGent) , Chris Poppe (UGent) , Peter Lambert (UGent) and Rik Van de Walle (UGent)
Author
Organization
Abstract
Due to the semantic gap, describing high-level semantic concepts with low-level visual features is a very challenging task. The classification of textures in scene images is intricate because of the high variation of the data. Therefore, the application of appropriate features is of utter importance. This paper presents biologically inspired features for texture segmentation and an unsupervised method to link those texture features with semantic concepts. The calculation of the features is inspired by the human visual system and corresponds to cell outputs in the first stage of the visual cortex. Analogously to the processing principles of the cortex, self-organizing maps are employed for classification. The performance of the texture segmentation and labeling is evaluated on textures from the Brodatz album and on a real-life scenery image dataset. For both methods, a high percentage of pixels is correctly classified.

Citation

Please use this url to cite or link to this publication:

Chicago
Martens, Gaëtan, Chris Poppe, Peter Lambert, and Rik Van de Walle. 2008. “Unsupervised Texture Segmentation and Labeling Using Biologically Inspired Features.” In 2008 IEEE 10TH WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING, VOLS 1 AND 2, ed. D Feng, D Sikora, WC Siu, J Zhang, L Guan, J Dugelay, Q Wu, and W Li, 159–164. New York, NY, USA: IEEE.
APA
Martens, Gaëtan, Poppe, C., Lambert, P., & Van de Walle, R. (2008). Unsupervised texture segmentation and labeling using biologically inspired features. In D. Feng, D. Sikora, W. Siu, J. Zhang, L. Guan, J. Dugelay, Q. Wu, et al. (Eds.), 2008 IEEE 10TH WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING, VOLS 1 AND 2 (pp. 159–164). Presented at the 10th IEEE Workshop on Multimedia Signal Processing, New York, NY, USA: IEEE.
Vancouver
1.
Martens G, Poppe C, Lambert P, Van de Walle R. Unsupervised texture segmentation and labeling using biologically inspired features. In: Feng D, Sikora D, Siu W, Zhang J, Guan L, Dugelay J, et al., editors. 2008 IEEE 10TH WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING, VOLS 1 AND 2. New York, NY, USA: IEEE; 2008. p. 159–64.
MLA
Martens, Gaëtan, Chris Poppe, Peter Lambert, et al. “Unsupervised Texture Segmentation and Labeling Using Biologically Inspired Features.” 2008 IEEE 10TH WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING, VOLS 1 AND 2. Ed. D Feng et al. New York, NY, USA: IEEE, 2008. 159–164. Print.
@inproceedings{678745,
  abstract     = {Due to the semantic gap, describing high-level semantic concepts with low-level visual features is a very challenging task. The classification of textures in scene images is intricate because of the high variation of the data. Therefore, the application of appropriate features is of utter importance. This paper presents biologically inspired features for texture segmentation and an unsupervised method to link those texture features with semantic concepts. The calculation of the features is inspired by the human visual system and corresponds to cell outputs in the first stage of the visual cortex. Analogously to the processing principles of the cortex, self-organizing maps are employed for classification. The performance of the texture segmentation and labeling is evaluated on textures from the Brodatz album and on a real-life scenery image dataset. For both methods, a high percentage of pixels is correctly classified.},
  author       = {Martens, Ga{\"e}tan and Poppe, Chris and Lambert, Peter and Van de Walle, Rik},
  booktitle    = {2008 IEEE 10TH WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING, VOLS 1 AND 2},
  editor       = {Feng, D and Sikora, D and Siu, WC and Zhang, J and Guan, L and Dugelay, J and Wu, Q and Li, W},
  isbn         = {9781424422944},
  language     = {eng},
  location     = {Cairns, Australia},
  pages        = {159--164},
  publisher    = {IEEE},
  title        = {Unsupervised texture segmentation and labeling using biologically inspired features},
  year         = {2008},
}

Web of Science
Times cited: