Advertisement

Relating Statistical Image Differences and Degradation Features

  • Elisa Barney Smith
  • Xiaohui Qiu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2423)

Abstract

Document images are degraded through bilevel processes such as scanning, printing, and photocopying. The resulting image degradations can be categorized based either on observable degradation features or on degradation model parameters. The degradation features can be related mathematically to model parameters. In this paper we statistically compare pairs of populations of degraded character images created with different model parameters. The changes in the probability that the characters are from different populations when the model parameters vary correlate with the relationship between observable degradation features and the model parameters. The paper also shows which features have the largest impact on the image.

Keywords

Point Spread Function Document Image Degradation Model Image Degradation Degradation Feature 
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.
    H. S. Baird. Document image defect models. In H.S. Baird, H. Bunke, and K. Yamamoto(eds), editors, Structured Document Image Analysis. Springer-Verlag, June 1992.Google Scholar
  2. 2.
    H. S. Baird. Calibration of document image defect models. In Proc. of 2nd annual symposium on document analysis and information retrieval, Las Vegas, Nevada, pages 1–16, April 1993.Google Scholar
  3. 3.
    E. H. Barney Smith. Characterization of image degradation caused by scanning. Pattern Recognition Letters, 19(13):1191–1197, 1998.zbMATHCrossRefGoogle Scholar
  4. 4.
    E. H. Barney Smith. Optical Scanner Characterization Methods Using Bilevel Scans. PhD thesis, Rensselaer Polytechnic Institute, December 1998.Google Scholar
  5. 5.
    E. H. Barney Smith. Bilevel image degradations: Effects and estimation. In Proc. 2001 Symposium on Document Image Understanding Technology, pages 49–55, Columbia, MD, 2001.Google Scholar
  6. 6.
    E. H. Barney Smith. Estimating scanning characteristics from corners in bilevel images. In Proc. SPIE Document Recognition and Retrieval VIII, volume 4307, pages 176–183, San Jose, CA, 2001.Google Scholar
  7. 7.
    E. H. Barney Smith. Scanner parameter estimation using bilevel scans of star charts. In Proc. International Conference on Document Analysis and Recognition 2001, pages 1164–1168, Seattle, WA, 2001.Google Scholar
  8. 8.
    L. R. Blando, J. Kanai, and T. A. Nartker. Prediction of OCR accuracy using simple features. In Proc. of the Third International Conference on Document Analysis and Recognition, pages 319–322, Montreal, Quebec, Canada, 1995.Google Scholar
  9. 9.
    T. K. Ho and H. S. Baird. Large-scale simulation studies in image pattern recognition. IEEE PAMI, 19(10):1067–1079, 1997.Google Scholar
  10. 10.
    T. Kanungo, R. M. Haralick, H. S. Baird, and D. M. Werner Stuezle. A statistical, nonparametricmetho dology for document degradation model validation. IEEE PAMI, 22(11):1209–1223, 2000.Google Scholar
  11. 11.
    T. Pavlidis, M. Chen, and E. Joseph. Sampling and quantization of bilevel signals. Pattern Recognition Letters, 14:559–562, 1993.CrossRefGoogle Scholar
  12. 12.
    P. Sarkar, G. Nagy, J. Zhou, and D. Lopresti. Spatial sampling of printed patterns. IEEE PAMI, 20(3):344–351, 1998.Google Scholar
  13. 13.
    T. Sziriáinyi and Á. Böröczki. Overall picture degradation error for scanned images and the effciency of character recognition. Optical Engineering, 30(12):1878–1884, 1991.CrossRefGoogle Scholar
  14. 14.
    W. R. Throssell and P. R. Fryer. The measurement of print quality for optical character recognition systems. Pattern Recognition, 6:141–147, 1974.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Elisa Barney Smith
    • 1
  • Xiaohui Qiu
    • 2
  1. 1.Boise State UniversityBoiseUSA
  2. 2.Nanjing University of Post and TelecommunicationChina

Personalised recommendations