Abstract
Comparing probability or possibility distributions is important in many fields of information processing under uncertainty. In this paper we address the question of defining and computing Hausdorff distances between distributions in a general sense. We propose several dilations of distributions, and exhibit some links between Lévy-Prokhorov distances and dilation-based distances. In particular, mathematical morphology provides an elegant way to deal with periodic distributions. The case of possibility distributions is addressed using fuzzy mathematical morphology. As an illustration, the proposed approaches are applied to the comparison of spatial relations between objects in an image or a video sequence, when these relations are represented as distributions.
References
[1] A. Abou-Elailah, V. Gouet-Brunet, I. Bloch: Detection of ruptures in spatial relationships in video sequences. In: International Conference on Pattern Recognition Applications and Methods - ICPRAM. pp. 110–120. Lisbon, Portugal (2015) 10.5220/0005213501100120Search in Google Scholar
[2] R. Aliev, W. Pedrycz, B. Fazlollahi, O.H. Huseynov, A.V. Alizadeh, B. G. Guirimov: Fuzzy logic-based generalized decision theory with imperfect information. Information Sciences 189, 18–42 (2012) 10.1016/j.ins.2011.11.027Search in Google Scholar
[3] I. Bloch: On Fuzzy Distances and their Use in Image Processing under Imprecision. Pattern Recognition 32(11), 1873–1895 (1999) 10.1016/S0031-3203(99)00011-4Search in Google Scholar
[4] I. Bloch, J. Atif: Comparaison de relations spatiales floues - Approches par transport optimal et morphologiemathématique. In: Rencontres Francophones sur la Logique Floue et ses Applications - LFA. pp. 133–140. Cargèse, France (2014) Search in Google Scholar
[5] I. Bloch, J. Atif: Deux approches pour la comparaison de relations spatiales floues : transport optimal et morphologiemathématique. Revue d’Intelligence Artificielle 29(5), 595–619 (2015) 10.3166/ria.29.595-619Search in Google Scholar
[6] I. Bloch, J. Atif: Hausdorff distances between distributions using optimal transport and mathematical morphology. In: 12th International Symposium on Mathematical Morphology (ISMM), LNCS 9082. pp. 522–534. Reykjavik, Iceland (2015) 10.1007/978-3-319-18720-4_44Search in Google Scholar
[7] L. Boxer: On Hausdorff-like Metrics for Fuzzy Sets. Pattern Recognition Letters 18, 115–118 (1997) 10.1016/S0167-8655(97)00006-8Search in Google Scholar
[8] B.B. Chauduri, A. Rosenfeld: On a Metric Distance between Fuzzy Sets. Pattern Recognition Letters 17, 1157–1160 (1996) 10.1016/0167-8655(96)00077-3Search in Google Scholar
[9] E.R. Dougherty: Application of the Hausdorff metric in gray-scale mathematical morphology via truncated umbrae. Journal of Visual Communication and Image Representation 2(2), 177–187 (1991) 10.1016/1047-3203(91)90007-3Search in Google Scholar
[10] D. Dubois, H. Prade: On Distance between Fuzzy Points and their Use for Plausible Reasoning. In: International Conference on Systems, Man, and Cybernetics. pp. 300–303 (1983) Search in Google Scholar
[11] R. M. Dudley: Distances of probability measures and random variables. The Annals of Mathematical Statistics 39(5), 1563– 1572 (1968) 10.1214/aoms/1177698137Search in Google Scholar
[12] M. Gromov: Metric Structures for Riemannian and Non-Riemannian Spaces. Progress in Mathematics 152 (1999) Search in Google Scholar
[13] J. Lindblad, N. Sladoje: Linear time distances between fuzzy sets with applications to pattern matching and classification. IEEE Transactions on Image Processing 23(1), 126–136 (2014) 10.1109/TIP.2013.2286904Search in Google Scholar PubMed
[14] K. Miyajima, A. Ralescu: Spatial Organization in 2D Images. In: Third IEEE International Conference on Fuzzy Systems, FUZZIEEE’ 94. pp. 100–105. Orlando, FL (Jun 1994) Search in Google Scholar
[15] O. Pele, M. Werman: A linear time histogram metric for improved SIFT matching. In: European Conference on Computer Vision, ECCV. pp. 495–508 (2008) 10.1007/978-3-540-88690-7_37Search in Google Scholar
[16] A.T. Popov: Hausdorff distance and fractal dimension estimation bymathematical morphology revisited. In: Nonlinar Signal and Image Processing, NSIP. pp. 90–94 (1999) Search in Google Scholar
[17] Y. Prokhorov: Convergence of random processes and limit theorems in probability theory. Theory of Probability & Its Applications 1(2), 157–214 (1956) 10.1137/1101016Search in Google Scholar
[18] M.L. Puri, D.A. Ralescu: Différentielle d’une fonction floue. C. R. Académie de Sciences de Paris, Série I 293, 237–239 (1981) Search in Google Scholar
[19] J. Rabin, J. Delon, Y. Gousseau: Transportation distances on the circle. Journal of Mathematical Imaging and Vision 41(1-2), 147–167 (2011) 10.1007/s10851-011-0284-0Search in Google Scholar
[20] S.T. Rachev: Minimal metrics in the real random variables space. In: Stability Problems for Stochastic Models. pp. 172–190 (1983) 10.1007/BFb0082069Search in Google Scholar
[21] A. Rosenfeld: Distances between Fuzzy Sets. Pattern Recognition Letters, 3, 229–233 (1985) 10.1016/0167-8655(85)90002-9Search in Google Scholar
[22] B. Sendov: Hausdorff approximations, vol. 50. Springer (1990) 10.1007/978-94-009-0673-0Search in Google Scholar
[23] J. Serra: Image Analysis and Mathematical Morphology. Academic Press, New-York (1982) Search in Google Scholar
[24] C. Villani: Optimal transport: old and new. Springer, Berlin (2003) Search in Google Scholar
[25] M. Werman, S. Peleg, A. Rosenfeld: A distance metric for multidimensional histograms. Computer Vision, Graphics, and Image Processing 32(3), 328–336 (1985) 10.1016/0734-189X(85)90055-6Search in Google Scholar
[26] R. Zwick, E. Carlstein, D.V. Budescu: Measures of Similarity Among Fuzzy Concepts: A Comparative Analysis. International Journal of Approximate Reasoning 1, 221–242 (1987) 10.1016/0888-613X(87)90015-6Search in Google Scholar
© 2016 Isabelle Bloch and Jamal Atif
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
