Abstract
Image segmentation using a hierarchical sequence of piecewise constant approximations that minimally differ from the original image in terms of the total squared error is discussed. It is proposed to obtain these approximations by two combined clustering and segmentation methods based on clustering image pixels using Ward’s method. In the first method, the number of segments in clusters is reduced in the course of hierarchical clustering by reclassifying pixels from one cluster to another. In the second method, a limited number of superpixels representing connected segments of the image are formed by enlarging source pixels, and then the superpixels are clusterized by Ward’s method. To decompose the image into superpixels, the segmentation quality is improved while preserving the number of segments. As a result, a noticeable improvement in the quality of image approximations is achieved, and their invariant encoding gives a marking of the image for subsequent object detection.
Similar content being viewed by others
References
Gurevich, I.B. and Yashina, V.V. Descriptive approach to image analysis: Image formalization space, Pattern Recognit. Image Anal.: Adv. Math. Theory Appl., 2012, vol. 22, no. 4, pp. 495–518.
Vizil’ter, Yu.V and Zheltov, S.Yu., Computer vision in modern aviation systems, Mekh. Upr. Inform., 2011, no. 6, pp. 11–44.
Ablameiko, S.V., Nedz’ved’, A.M., Belotserkovskii, A.M., and Rutskaya, E.A., Segmentation of 3D computer tomography images based on global and local data, Vestn. Belarus. Gos. Univ., Ser. 1, Phys., Math, Inform., 2009, no. 1, pp. 58–64.
Mel’nikov, P.A. and Kopylov, A.V., An algorithm for finding a quasi-optimal markup for image processing using a rowwise combination of variables, Izv. Tul’skogo Gos. Univ., Natural Sci., 2012, no. 1, pp. 119–129.
Aivazyan, S.A., Bukhshtaber, V.M., Enyukov, I.S., and Meshalkin, L.D., Applied Statistics: Classification and Reduction of Dimensionality, Ìoscow: Finansy I Statistika, 1989.
Kharinov, M.V., A generalization of three approaches to the optimal segmentation of digital images, in Trudy SPIIRAN, 2013, no. 2(25), pp. 294–316.
Dvoenko, S.D., Meanless k-means as k-meanless clustering with the bi-partial approach, Pattern Recognit. Inf. Process. (PRIP’2014), Proc. of the 12th Int. Conf., Minsk, 2014, pp. 50–54.
Bugaev, A.S. and Khel’vas, A.V., Exploratory research and the development of methods and tools for the analysis and automatic recognition of stream data in global information systems, Report Latskan, (Ìoskow: Mosc. Fiz.-tekhn. Inst., 2001), vol. 1.
Mumford, D. and Shah, J., Boundary detection by minimizing functionals, in Proc. IEEE Comput. Vision Pattern Recognition Conf., San Francisco, 1985, pp. 22–26.
Crisp, D.J. and Tao, T.C., Fast Region merging algorithms for image segmentation, in The 5th Asian Conf. on Computer Vision (ACCV2002), Melbourne, Australia, 2002, pp. 1–6.
Koepfler, G., Lopez, C., and Morel, J., A multiscale algorithm for image segmentation by variational method, SIAM J. Numer. Anal., 1994, vol. 31, no. 1, pp. 282–299.
Kharinov, M.V., Memorizing and Adaptive Digital Image Processing, Yusupovm R.M., Ed., St. Peresburg: St. Pteresburg Gos. Univ., 2006.
Ward, J.H., ffixJr., Hierarchical grouping to optimize an objective function, J. Am. Stat. Assoc., 1963, vol. 58, no. 301, pp. 236–244.
Arifin, A.Z. and Asano, A., Image segmentation by histogram thresholding using hierarchical cluster analysis, Pattern Recognit. Lett., 2006, vol. 27, no. 13, pp. 1515–1521.
Otsu, N., A threshold selection method from gray-level histograms, IEEE Trans. Syst., Man, Cybernetics, 1979, vol. SMC-9, no. 1, pp. 62–66.
Tarjan, R.E., Efficiency of a good but not linear set union algorithm, J. ACM, 1975, vol. 22, no. 2, pp. 215–225.
Nock, R. and Nielsen, F., Statistical region merging, IEEE Trans. Pattern Anal. Mach. Intell., 2004, vol. 26, no. 11, pp. 1452–1458.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © M.V. Kharinov, 2015, published in Programmirovanie, 2015, Vol. 41, No. 5.
Rights and permissions
About this article
Cite this article
Kharinov, M.V. Pixel clustering for color image segmentation. Program Comput Soft 41, 258–266 (2015). https://doi.org/10.1134/S0361768815050047
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0361768815050047