Abstract
Area openings and closings are morphological filters which efficiently suppress impulse noise from an image, by removing small connected components of level sets. The problem of an objective choice of threshold for the area remains open. Here, a mathematical model for random images will be considered. Under this model, a Poisson approximation for the probability of appearance of any local pattern can be computed. In particular, the probability of observing a component with size larger than k in pure impulse noise has an explicit form. This permits the definition of a statistical test on the significance of connected components, thus providing an explicit formula for the area threshold of the denoising filter, as a function of the impulse noise probability parameter. Finally, using threshold decomposition, a denoising algorithm for grey level images is proposed.
Similar content being viewed by others
References
B. Bollobás,Random Graphs, Academic Press: London, 1985.
A.R. Conway and A.J. Guttmann, “On two-dimensional percolation”J. Phys. A: Math. Gen., Vol. 28, pp. 891–904, 1995.
D. Coupier, A. Desolneux, and B. Ycart, A zero-one law for first-order logic on random images, inProceedings Math-Info’04, Drmota et al. (Ed.), to appear 2004.
A. Desolneux, L. Moisan, and J.-M. Morel, “Meaningful alignments”Int. J. Computer Vision, Vol. 40, No. 1, pp. 7–23, 2000.
A. Desolneux, L. Moisan, and J.-M. Morel, “Edge detection by Helmholtz principle”Journal of Mathematical Imaging and Vision, Vol. 14, No. 3, pp. 271–284, 2001.
D.L. Donoho and I.M. Johnstone, “Adapting to unknown smoothness via wavelet shrinkage”Journal of Amer. Stat. Assoc., Vol. 90, pp. 1200–1224, 1995.
H.J.A.M. Heijmans, “Self-dual morphological operators and filters”Journal of Mathematical Imaging and Vision, Vol. 6, No. 1, pp. 15–36, 1996.
H.J.A.M. Heijmans, “Connected morphological operators for binary images”Computer Vision and Image Understanding, Vol. 73, No. 1, pp. 99–120, 1999.
I. Jensen and A.J. Guttmann, “Statistics of lattice animals (polyominoes) and polygons”J. Phys. A: Math. Gen., Vol. 33, pp. L257–L263, 2000.
D.A. Klarner, “Cell growth problems”Canad. J. Math., Vol. 19, pp. 851–863, 1967.
D.A. Klarner and R.L. Rivest, “A procedure for improving the upper bound for the number ofn-ominoes”Canad. J. Math., Vol. 25, pp. 585–602, 1973.
S. Masnou and J.-M. Morel, “Image restoration involving connectedness” inProc. DIP’97, Vienna, Austria, Vol. 3346. SPIE, 1998.
P. Monasse, “Contrast invariant representation of digital images and application to registration” PhD thesis, University Paris 9, 2000.
P. Monasse and F. Guichard, “Fast computation of a contrast invariant image representation”IEEE Transactions on Image Processing, Vol. 9, No. 5, pp. 860–872, 2000.
B. Sapoval and M. Rosso, “Gradient percolation and fractal frontiers in image processing”Fractals, Vol. 3, pp. 23–31, 1995.
J. Serra,Image Analysis and Mathematical Morphology, Part. I, Academic Press: New York, 1982.
J. Serra,Image Analysis and Mathematical Morphology, Part. II: Theoretical Advances, Academic Press: London, 1988.
J. Serra and L. Vincent, “An overview of morphological filtering”Circuits, Systems and Signal Processing, Vol. 11, No. 1, pp. 47–108, 1992.
J. Spencer, “Nine lectures on random graphs” inEcole d’été de probabilité de Saint-Flour XXI, P. Bernard (Ed.), Vol. 1541 ofL.N. in Mathematics, Springer-Verlag, New York, 1991, pp. 293–343.
L. Vincent, “Morphological area opening and closing for grayscale images” inProc. NATO Shape in Picture Workshop, Driebergen, Springer-Verlag: The Netherlands, 1992, pp. 197–208.
L. Vincent, “Grayscale area openings and closings: Their applications and efficient implementation” inProc. EURASIP Workshop on Mathematical Morphology and its Applications to Signal Processing, Barcelona, Spain, 1993, pp. 22–27.
Author information
Authors and Affiliations
Corresponding author
Additional information
David Coupier is 25 years old, he is PhD student at the MAP5. He has studied mathematics at the University of Orsay—Paris XI. He is working on zero-one laws and Poisson approximations for random images. Web page: http://www.math-info.univ-paris5.fr/~coupier/
Agnés Desolneux is 30 years old, she is CNRS researcher at the MAP5. She defended her PhD thesis in applied mathematics in 2000 under the direction of Jean-Michel Morel at the ENS Cachan. She is working on statistical methods in image analysis. Web page: http://www.math-info.univ-paris5.fr/~desolneux/
Bernard Ycart is 45 years old, he is Professor of mathematics at the University Paris 5 and director of the MAP5 (FRE CNRS 2428). He is specialist of applied probabilities (Markov processes, stochastic algorithms). Web page: http://www.math-info.univ-paris5.fr/~ycart/
Rights and permissions
About this article
Cite this article
Coupier, D., Desolneux, A. & Ycart, B. Image Denoising by Statistical Area Thresholding. J Math Imaging Vis 22, 183–197 (2005). https://doi.org/10.1007/s10851-005-4889-z
Issue Date:
DOI: https://doi.org/10.1007/s10851-005-4889-z