Bézier modelling of cracks
In this paper we show how arbitrary patterns of cracks can be fitted by Bézier curves of unknown order. The Reversible Jump MCMC technique is used to estimate both the number of curves in the image, and the positions of the knots and control points for each curve. The technique described in this paper is suited to a variety of line fitting applications.
Unable to display preview. Download preview PDF.
- 1.Morris, R.: Image sequence restoration using Gibbs distributions. PhD. thesis, Cambridge University, (1995)Google Scholar
- 2.Rue, H., Syversveen, A. R.: Bayesian object recognition using Baddeley's delta loss. Department of Mathematical Sciences, NTH, Norway, (1995)Google Scholar
- 3.Clifford, P., Nicholls, G.: A Metropolis sampler for polygonal image reconstruction. Department of Statistics, Oxford University, (1994)Google Scholar
- 4.Nicholls, G.: Bayesian image analysis with Markov chain Monte Carlo and colored continuum triangulation models. Department of Mathematics, Aukland University, (1996)Google Scholar
- 5.Bucklow, S.: Formal connoisseurship and the characterisation of craquelure. PhD. thesis, Cambridge University, (1997)Google Scholar
- 6.Ripley, B., Sutherland, A.: Finding spiral structures in images of galaxies. Philosophical Transactions of the Royal Society of London 332 (1990) 477–485Google Scholar
- 7.Green, P.: Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika 82 (1996) 711–732Google Scholar
- 8.Foley, J., van Dam, A., Feiner, S., Hughes, J.: Computer graphics: principles and practice. Addison-Wesley (1991)Google Scholar
- 9.Denison, D., Mallick, B., Smith, A.: Automatic Bayesian curve fitting. Department of Mathematics, Imperial College, (1996)Google Scholar