Abstract
We present the results of numerical simulation for image segmentation based on the chain distance clustering algorithm. The key issue is the use of the p-adic metric, where p > 1 is a prime number, at the scale of levels of brightness (pixel wise). In previous studies the p-adic metric was used mainly in combination with spectral methods. In this paper this metric is explored directly, without preparatory transformations of images. The main distinguishing feature of the p-adic metric is that it reflects the hierarchic structure of information presented in an image. Different classes of images match with in general different prime p (although the choice p = 2 works on average). Therefore the presented image segmentation procedure has to be combined with a kind of learning to select the prime p corresponding to the class of images under consideration.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
This is the nearest neighbor chain algorithm, that, with the reciprocal or mutual nearest neighbor; this is the standard efficient implementation for all standard agglomerative hierarchical clustering algorithms.
- 2.
This (thresholding chains or relation representations, from, e.g., single link clustering, or the minimal spanning tree) has been used for image segmentation in some previous work. A review that includes some of this work is available, e.g., in [28].
- 3.
We noted that future work would consider other color encoding beyond RGB (YUV, etc.); also dynamic range, here 256, to be extended.
- 4.
However, the following comment can be useful to clarify the situation with computational speed up. Clustering that uses a p-adic distance can be considered by analogy with (1) creating a hierarchical clustering, and then (2) cutting the hierarchy to produce a partition. This analogy is more the case if the hierarchy is a contiguity-constrained one. The computational requirements that are referred to quite negatively as regards this work may not be all that bad, especially if one makes comparison relative to straightforward, Euclidean distance or related, segmentation. A further comparison in future work is likely to be beneficial too: the use of wavelet transforms—hierarchical and hence ideally represented p-adically for image and signal segmentation.
References
Ilea, D.E., Whelan, P.F.: Image segmentation based on the integration of colour–texture descriptors—a review. Pattern Recogn. 44(N10–11), 2479–2501 (2011)
Hill, J., Corona, E., Ao, J., Mitra, S., Nutter, B.: Information theoretic clustering for medical image segmentation. In: Advanced Computational Approaches to Biomedical Engineering, pp. 47–70. Springer, Berlin-Heidelberg (2014)
Arbelaez, P., Maire, M., Fowlkes, C., Malik, J.: Contour detection and hierarchical image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 33(5), 898–916 (2011)
Vieux, R., Benois-Pineau, J., Domenger, J.-P., Braquelair, A.: Segmentation-based multi-class semantic object detection. Multimedia Tools and Applications, pp. 1–22. Springer Verlag (2011)
Rossi, E., Benini, S., Leonardi, R., Mansencal, B., Benois-Pineau, J., Clustering of scene repeats for essential rushes preview. International Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS), pp. 234–237, London, United Kingdom (2009)
IEEE Explore Digital Library. http://ieeexplore.ieee.org.
Benois-Pineau, J., Khrennikov, A.Y., Kotovich, N.V.: Segmentation of images in p-adic and Euclidean metrics. Dokl. Math. 64, 450–455 (2001)
Benois-Pineau, J., Khrennikov, A. (2002). Image segmentation in compressed domain by clustering methods with Euclidean and p-adic metrics. In: Proceedings in Interational Conference in Management of Uncertainty in Knowledge-based Systems (IPMU-2), pp. 969–976, Annecy, France
Khrennikov, A.Y., Kotovich, N.V.: Representation and compression of images with the aid of the m-adic coordinate system. Dokl. Akad. Nauk. 387, 159–163 (2002)
Khrennikov, A., Kotovich, N., Borzistaya, E.: Compression of images with the aid of representation by p-adic maps and approximation by Mahler’s polynomials. Dokl. Math. 69(N3), 373–377 (2004)
Khrennikov, A.Y.: Non-Archimedean analysis: quantum paradoxes, dynamical systems and biological models. Kluwer, Dordrecht (1997)
Albeverio, S., Khrennikov, A.Y., Kloeden, P.: Memory retrieval as a p-adic dynamical system. Biosystems. 49, 105–115 (1997)
Dubischar, D., Gundlach, V.M., Steinkamp, O., Khrennikov, A.Y.: A p-adic model for the process of thinking disturbed by physiological and information noise. J. Theor. Biol. 197, 451–467 (1999)
Khrennikov, A.Y.: Toward an adequate mathematical model of mental space: conscious/unconscious dynamics on m-adic trees. Biosystems. 90, 656–675 (2007)
Murtagh, F.: Ultrametric model of mind, I: review. P-adic Numbers Ultrametr. Anal. Appl. 4, 193–206 (2012)
Murtagh, F.: Ultrametric model of mind, II: application to text content analysis. P-adic Numbers Ultrametr. Anal. Appl. 4, 207–221 (2012)
Lauro-Grotto, R.: The unconscious as an ultrametric set. Am. Imago. 64, 535–543 (2007)
Khrennikov, A., Kotovich, N.: Towards ultrametric modeling of unconscious creativity. Int. J. Cogn. Inform. Nat. Intell. 8(4), 12 (2014)
Murtagh, F.: On ultrametricity, data coding, and computation. J. Classif. 21, 167–184 (2004)
Murtagh, F., Contreras, P.: Fast, linear time, m-adic hierarchical clustering for search and retrieval using the baire metric, with linkages to generalized ultrametrics, hashing, formal concept analysis, and precision of data measurement. P-Adic Numbers Ultrametr. Anal. Appl. 4, 45–56 (2012)
Bradley, P.: Degenerating families of dendrograms. J. Classif. 25, 27–42 (2008)
Bradley, P.: From image processing to topological modelling with p-adic numbers. P-Adic Numbers Ultrametr. Anal. Appl. 2, 293–304 (2010)
Bradley, P.: Mumford dendrograms. Comput. J. 53, 393–404 (2010)
Bradley, P.: Comparing G-maps with other topological data structures. GeoInformatica. 18, 595–620 (2014)
Benois-Pineau, J., Khrennikov, A.: Significance delta reasoning with p-adic neural networks: application to shot change detection in video. Comput. J. 53, 417–431 (2010)
Kaufman, L., Rousseeuw, P.J.: Finding groups in data: an introduction to cluster analysis, 1st edn. John Wiley, New York (1990)
Hastie, T., Tibshirani, R., Friedman, J.: Hierarchical clustering. The elements of statistical learning, 2nd edn, pp. 520–528. Springer, New York (2009)
Murtagh, F.: Algorithms for contiguity-constrained clustering. Comput. J. 28, 82–88 (1985)
Khrennikov, A., Kozyrev, S.V., Oleschko, K., Jaramillo, A.G., de Jesus Correa Lopez, M.: Application of p-adic analysis to time series. Infinite Dimens. Anal. Quantum Probab. Relat. Top. 16, 1350030 (2013)
Khrennikov, A., Oleschko, K., de Jesús Correa Lopez, M.: Modeling fluid’s dynamics with master equations in ultrametric spaces representing the treelike structure of capillary networks. Entropy. 18(7), 249 (2016)
Khrennikov, A., Oleschko, K., de Jesús Correa Lopez, M.: Application of p-adic wavelets to model reaction-diffusion dynamics in random porous media. J. Fourier Anal. Appl. 22, 809–822 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Khrennikov, A., Kotovich, N. (2017). Image Segmentation with the Aid of the p-Adic Metrics. In: Toni, B. (eds) New Trends and Advanced Methods in Interdisciplinary Mathematical Sciences. STEAM-H: Science, Technology, Engineering, Agriculture, Mathematics & Health. Springer, Cham. https://doi.org/10.1007/978-3-319-55612-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-319-55612-3_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-55611-6
Online ISBN: 978-3-319-55612-3
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)