Advertisement

Measuring the Quality Evaluation for Image Segmentation

  • Rodrigo Janasievicz Gomes Pinheiro
  • Jacques Facon
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3773)

Abstract

This paper proposes a measure of quality for evaluating the performance of region-based segmentation methods. Degradation mechanisms are used to compare segmentation evaluation methods onto deteriorated ground-truth segmentation images. Experiments showed the significance of using degradation mechanisms to compare segmentation evaluation methods. Encouraging results were obtained for a selection of degradation effects.

Keywords

Image Segmentation Quality Evaluation Segmentation Process Pixel Number Pepper Noise 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Zhang, Y.: A survey on evaluation methods for image segmentation. Pattern Recognition 29, 1335–1346 (1996)CrossRefGoogle Scholar
  2. 2.
    Yasnoff, W., Mui, J.K., Bacus, J.W.: Error measures for scene segmentation. Pattern Recognition 9, 217–231 (1977)CrossRefGoogle Scholar
  3. 3.
    Menoti, D., Facon, J., Borges, D.L., Souza, A.: Segmentation of postal envelopes for address block location: an approach based on feature selection in wavelet space. In: ICDAR 2003 - 7th InternationalConference on Document Analysis and Recognition, vol. 2, pp. 699–703 (2003)Google Scholar
  4. 4.
    Eiterer, L.F., Facon, J., Menoti, D.: Fractal-based approach for segmentation of address block in postal envelopes. 9TH Iberoamerican Congress on Pattern Recognition. LNCS Lecture Notes in Computer Science 1, 454–461 (2004)Google Scholar
  5. 5.
    Roldán, R.R., Lopera, J.F.G., Allah, C.A., Aroza, J.M., Escamilla, P.L.L.: A measure of quality for evaluating methods of segmentation and edge detection. Pattern Recognition 34, 969–980 (2001)zbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Rodrigo Janasievicz Gomes Pinheiro
    • 1
  • Jacques Facon
    • 1
  1. 1.PUCPR-Pontifícia Universidade Católica do ParanáCuritibaBrazil

Personalised recommendations