ML Anthology

Improving Recognition Effectiveness of Noisy Texture Concepts

Peter Pachowicz, Jerzy W. Bala
ICML 1991 pp. 625-629
doi:10.1016/B978-1-55860-200-7.50127-6 /icml/1991/pachowicz1991icml-improving/

Abstract

Learning from noisy engineering data requires to develop either noise-tolerant learning tools or concept optimization methods. We present and compare several concept optimization approaches. This comparison is provided for texture recognition and image segmentation task. Training and test data represent 12 texture classes extracted from low quality images incorporating Laws' energy masks.

ICML Semantic Scholar

Cite

Text

Pachowicz and Bala. "Improving Recognition Effectiveness of Noisy Texture Concepts." International Conference on Machine Learning, 1991. doi:10.1016/B978-1-55860-200-7.50127-6

Markdown

[Pachowicz and Bala. "Improving Recognition Effectiveness of Noisy Texture Concepts." International Conference on Machine Learning, 1991.](https://mlanthology.org/icml/1991/pachowicz1991icml-improving/) doi:10.1016/B978-1-55860-200-7.50127-6

BibTeX

@inproceedings{pachowicz1991icml-improving,
  title     = {{Improving Recognition Effectiveness of Noisy Texture Concepts}},
  author    = {Pachowicz, Peter and Bala, Jerzy W.},
  booktitle = {International Conference on Machine Learning},
  year      = {1991},
  pages     = {625-629},
  doi       = {10.1016/B978-1-55860-200-7.50127-6},
  url       = {https://mlanthology.org/icml/1991/pachowicz1991icml-improving/}
}