Nothing Special   »   [go: up one dir, main page]

skip to main content
article

Robust image classification

Published: 01 July 2006 Publication History

Abstract

The automatic classification of images is now widely used in a range of applications. These include the diagnosis of arthritis from joint images, the classification of environmental noise from spectrograms and automatic text analysis. However, satisfactory performance is difficult to achieve in uncontrolled environments, as images are often contaminated by high levels of noise, outliers and global contamination due to illumination changes and environmental effects. We address these issues using a semi-parametric modelling strategy and a novel robust Bayesian classifier. This model is driven by additive Gaussian noise with non-uniform variance to describe outliers and uses the parametric and non-parametric components to describe contamination of different types. We assess the performance of our approach in two experiments based on real and simulated data. These show that our approach can significantly outperform a number of competitors in uncontrolled environments.

References

[1]
{1} M. Soriano, C. Saloma, Improved classification robustness for noisy cell images represented as principal-component projections in a hybrid recognition system, Appl. Opt. 37 (1998) 3628-3638.
[2]
{2} A. Schwaighofer, V. Tresp, P. Mayer, A.K. Scheel, G. Muller, The RA scanner: prediction of rheumatoid joint inflammation based on laser imaging, IEEE Trans. Biomed. Imaging 50 (2003) 375-382.
[3]
{3} T. Hastie, D. Ikeda, R. Tibshirani, Statistical measures for the computer-aided diagnosis of mammographic masses, J. Comput. Graph. Statist. 8 (1999) 531-543.
[4]
{4} T. Hastie, A. Buja, R. Tibshirani, Penalized discriminant analysis, Ann. Statist. 23 (1995) 73-102.
[5]
{5} Y. Zheng, H. Li, D. Doermann, Machine printed text and handwriting identification in noisy document images, IEEE Trans. Pattern Anal. Mach. Intell. 26 (2004) 337.
[6]
{6} B.W. Gillespie, L.E. Atlas, Optimizing time-frequency kernels for classification, IEEE Trans. Signal Process. 49 (2001) 485-496.
[7]
{7} A. Kokaram, A statistical framework for picture recognition using 2D AR models, Image Vis. Comput. 22 (2004) 165-171.
[8]
{8} W.D. Philpot, The derivative ratio algorithm--avoiding atmospheric effects in remote sensing, IEEE Trans. Geosci. Remote Sensing 29 (1991) 350-357.
[9]
{9} J. Shiau, G. Wahba, D. Johnson, Partial spline models for the inclusion of tropopause and frontal boundary information, J. Atmospheric Ocean Tech. 3 (1986) 714-725.
[10]
{10} R.H. Glendinning, A.J. Goode, Semiparametric classification of noisy curves, Pattern Recognition 36 (2003) 35-44.
[11]
{11} D.J. Smith, T.C. Bailey, A.G. Munford, Robust classification of high-dimensional data using artificial neural networks, Statist. Comput. 3 (1993) 71-81.
[12]
{12} W. Guo, Y. Wang, M.B. Brown, A signal extraction approach to modelling hormone time series with pulses and a changing baseline, J. Amer. Statist. Assoc. 94 (1999) 746-756.
[13]
{13} Y.S. Kharin, Robustness in Statistical Pattern Recognition, Kluwer Academic Publishers, Dordrecht, 1996.
[14]
{14} C.C. Holmes, D.G.T. Denison, Classification with Bayesian MARS, Mach. Learn. 50 (2003) 159-173.
[15]
{15} C.K.I. Williams, Bayesian classification with Gaussian processes, IEEE Trans. Pattern Anal. Mach. Intell. 20 (1998) 1342-1351.
[16]
{16} P.J. Brown, T. Fearn, M.S. Haque, Discrimination with many variables, J. Amer. Statist. Assoc. 94 (1999) 1320-1329.
[17]
{17} S. Geisser, Bayesian discrimination, in: P.R. Krishnaiah, L.N. Kanal (Eds.), Handbook of Statistics, vol. 2, North-Holland, Amsterdam, 1982, pp. 101-120.
[18]
{18} G. Koop, D.J. Poirier, Bayesian variants of some classical semiparametric regression techniques, J. Econometrics 123 (2) (2004) 259-282.
[19]
{19} C.M. Wong, R. Kohn, A Bayesian approach to semiparametric regression, J. Econometrics 74 (1996) 209-235.
[20]
{20} M. Smith, R. Kohn, A Bayesian approach to non-parametric bivariate regression, J. Amer. Statist. Assoc. 92 (1997) 1522-1535.
[21]
{21} P.J. Lenk, Bayesian inference for semiparametric regression using a Fourier representation, J. Roy. Statist. Soc. 61 (1999) 863-880.
[22]
{22} Y. Wang, Smoothing spline models with correlated random errors, J. Amer. Statist. Assoc. 93 (1998) 341-348.
[23]
{23} M. Smith, P. Yau, T. Shively, R. Kohn, Estimating long-term trends in tropospheric ozone levels, Internat. Statist. Rev. 70 (2002) 99-124.
[24]
{24} C. Stone, M. Hansen, C. Kooperberg, Y. Truong, Polynomial splines and their tensor products in extended linear modelling, Ann. Statist. 25 (1996) 1371-1470.
[25]
{25} R. Kohn, M. Smith, D. Chan, Non-parametric regression using linear combinations of basis functions, Statist. Comput. 11 (2001) 313-322.
[26]
{26} T.A. Severini, On the relationship between Bayesian and non-Bayesian elimination of nuisance parameters, Statist. Sinica 9 (1999) 713-724.
[27]
{27} N.X. Ge, D.G. Simpson, Correlation and high-dimensional consistency in pattern recognition, J. Amer. Statist. Assoc. 93 (1998) 995-1006.
[28]
{28} C. Croux, C. Dehon, Robust linear discriminant analysis using S-estimates, Canad. J. Statist. 29 (2001) 473-493.
[29]
{29} C.K. Lee, J.K. Ord, Discriminant analysis using least absolute deviation, Decision Sci. 21 (1990) 86-96.
[30]
{30} R.H. Glendinning, Robust shape classification, Signal Processing 7 (1999) 121-138.
[31]
{31} R.H. Glendinning, Selecting sub-set auto-regressions from outlier contaminated data, Comput. Statist. Data Anal. 36 (1999) 179-207.
[32]
{32} R. Von Sachs, Estimating the spectrum of a stochastic process in the presence of a contaminating signal, IEEE Trans. Signal Process. 41 (1993) 323-353.
[33]
{33} R. Chellappa, S. Chatterjee, Classification of textures using Gaussian Markov random fields, IEEE Trans. Acoust. Speech Signal Process. ASSP-33 (1985) 959-963.
[34]
{34} N. Arica, F.T. Yarman-Vural, An overview of character recognition focused on off-line handwriting, IEEE Trans. Systems Man Cybernet. C-31 (2001) 216-233.
[35]
{35} S. Jaggi, W.C. Karl, S.G. Mallat, A.S. Willsky, Silhouette recognition using high-resolution pursuit, Pattern Recognition 32 (1999) 753-771.
[36]
{36} Y. Bengio, Y. Lecun, C. Nohl, C. Burges, LEREC--A NN/ HMM hybrid for online handwriting recognition, Neural Computation 7 (1995) 1289-1303.
[37]
{37} P. Refregier, Bayesian theory for target location in noise with unknown spectral density, J. Opt. Soc. Amer. 10 (1999) 276-283.
[38]
{38} P.H.C. Eilers, B.D. Marx, Multivariate calibration with temperature interaction using two-dimensional penalized signal regression, Chemometrics Intel. Lab. Systems 66 (2003) 159-174.
[39]
{39} G.M. James, T.J. Hastie, Functional linear discriminant analysis for irregularly sampled curves, J. Roy. Statist. Soc. 63 (2001) 533-550.
[40]
{40} M. Smith, R. Kohn, Non-parametric estimation of irregular functions with independent or autocorrelated errors, in: D. Dey, P. Muller, D. Sinha (Eds.), Practical Nonparametric and Semiparametric Bayesian Statistics, Springer, Berlin, 1998, pp. 157-180.

Cited By

View all

Recommendations

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image Signal Processing
Signal Processing  Volume 86, Issue 7
July 2006
397 pages

Publisher

Elsevier North-Holland, Inc.

United States

Publication History

Published: 01 July 2006

Author Tags

  1. Bayesian methods
  2. Markov chain Monte-Carlo
  3. image classification
  4. outliers
  5. semi-parametric model

Qualifiers

  • Article

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)0
  • Downloads (Last 6 weeks)0
Reflects downloads up to 30 Dec 2024

Other Metrics

Citations

Cited By

View all

View Options

View options

Media

Figures

Other

Tables

Share

Share

Share this Publication link

Share on social media