Unsupervised Image Segmentation via Graph-Based Community Detection

  • Abdelmalik MoujahidEmail author
  • Fadi Dornaika
  • Blanca Cases
Conference paper


Community detection arises in a variety of fields ranging from mathematics, physics, biology, computer science, and the social sciences, among many others. It differs from the classical problem of graph partitioning in that the number and size of the groups into which the network is divided are not specified by the user. This makes community identification algorithms ideally suited to deal with image segmentation from a non-supervised perspective. In this chapter we present an unsupervised framework that automatically identifies semantic objects in images by formulating the general problem of semantic segmentation as community detection problem in graphs. The framework broadly follows a four-step procedure. First, we perform an over-segmentation of the original image using the well-known statistical region merging (SRM) algorithm which presents the advantage of not requiring any quantization or color space transformations. Second, we compute the feature descriptors of the resulting segmented regions. For encoding color and other textural information, each region is described by a hybrid descriptor based on color histograms and covariance matrix descriptor. Third, from the set of descriptors we construct different weighted graphs using various graph construction algorithms. Finally, the resulting graphs are then divided into groups or communities using a community detection algorithm based on spectral modularity maximization. This algorithm makes use of the eigenspectrum of matrices such as the graph Laplacian matrix and the modularity matrix which are more likely to reveal the community structure of the graph. Experiments conducted on large orthophotos depicting several zones in the region of Belfort city situated on the northeastern of France provide promising results. The proposed framework can be used by semiautomatic approaches to handle the challenging problems of scene parsing.


  1. 1.
    T. Ahonen, A. Hadid, and M. Pietikäinen. Face description with local binary patterns: application to face recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(12):2037–2041, 2006.CrossRefGoogle Scholar
  2. 2.
    S. Bandyopadhyay, G. Chowdhary, and D. Sengupta. Focs: Fast overlapped community search. IEEE Transactions on Knowledge and Data Engineering, 27(11):2974–2985, Nov 2015.CrossRefGoogle Scholar
  3. 3.
    M. Belkin and P. Niyogi. Laplacian eigenmaps for dimensionality reduction and data representation. Neural Comput., 15(6):1373–1396, June 2003.CrossRefGoogle Scholar
  4. 4.
    M. Bereta, P. Karczmarek, W. Pedrycz, and M. Reformat. Local descriptors in application to the aging problem in face recognition. Pattern Recognition, 46:2634–2646, 2013.CrossRefGoogle Scholar
  5. 5.
    C. X. C, D. Guanzhong, and Y. Libing. Survey on spectral clustering algorithm. Computer Science, 35:14–18, 2008.Google Scholar
  6. 6.
    F. Dornaika, A. Moujahid, Y. E. Merabet, and Y. Ruichek. Building detection from orthophotos using a machine learning approach: An empirical study on image segmentation and descriptors. Expert Systems with Applications, 58:130 – 142, 2016.CrossRefGoogle Scholar
  7. 7.
    C. Farabet, C. Couprie, L. Najman, and Y. LeCun. Learning hierarchical features for scene labeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8):1915–1929, Aug 2013.CrossRefGoogle Scholar
  8. 8.
    S. Fortunato. Community detection in graphs. Physics Reports, 486(3–5):75 – 174, 2010.MathSciNetCrossRefGoogle Scholar
  9. 9.
    Garima, H. Gulati, and P. K. Singh. Clustering techniques in data mining: A comparison. In 2015 2nd International Conference on Computing for Sustainable Global Development (INDIACom), pages 410–415, March 2015.Google Scholar
  10. 10.
    Y. C. Gong and C. C. L. spectral clustering. Advances in Artificial Intelligence, chapter Locality spectral clustering, pages 348–354. Springer Berlin Heidelberg, 2008.CrossRefGoogle Scholar
  11. 11.
    H. Hu. Graph Based Models for Unsupervised High Dimensional Data Clustering and Network Analysis. PhD thesis, University of California, 2015.Google Scholar
  12. 12.
    D. Huang, C. Shan, M. Ardabilian, and Y. Wang. Adaptive particle sampling and adaptive appearance for multiple video object tracking. IEEE Trans. on Systems, Man, and Cybernetics-Part C: Applications and reviews, 41(6):765–781, November 2011.Google Scholar
  13. 13.
    B. W. Kernighan and S. Lin. An efficient heuristic procedure for partitioning graphs. Bell System Technical Journal, 49(2):291–307, 1970.CrossRefGoogle Scholar
  14. 14.
    M. E. J. Newman. Finding community structure in networks using the eigenvectors of matrices. Phys. Rev. E, 74:036104, Sep 2006.Google Scholar
  15. 15.
    M. E. J. Newman. Modularity and community structure in networks. Proceedings of the National Academy of Sciences, 103(23):8577–8582, 2006.CrossRefGoogle Scholar
  16. 16.
    R. Nock and F. Nielsen. Statistical region merging. IEEE Trans. Pattern Anal. Mach, vol. 26, no 11:1452–1458, 2004.CrossRefGoogle Scholar
  17. 17.
    T. Ojala, M. Pietikäinen, and T. Maenpaa. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. Transactions on Pattern Analysis and Machine Intelligence, 24:971–987, 2002.CrossRefGoogle Scholar
  18. 18.
    S. T. Roweis and L. K. Saul. Nonlinear dimensionality reduction by locally linear embedding. Science, 290(5500):2323–2326, 2000.CrossRefGoogle Scholar
  19. 19.
    T. Sajana, C. M. S. Rani, and K. V. Narayana. A survey on clustering techniques for big data mining. Indian journal of Science and Technology, 9(3):1–12, 2016.CrossRefGoogle Scholar
  20. 20.
    S. E. Schaeffer. Graph clustering. Computer Science Review, 1(1):2007, 2007.Google Scholar
  21. 21.
    P. Sharma and J. Suji. A review on image segmentation with its clustering techniques. International Journal of Signal Processing, Image Processing and Pattern Recognition, 9(5):209–218, 2016.CrossRefGoogle Scholar
  22. 22.
    R. Socher, C. C. Lin, A. Y. Ng, and C. D. Manning. Parsing natural scenes and natural language with recursive neural networks. In International Conference on Machine Learning, 2011.Google Scholar
  23. 23.
    C. Sousa, S. Rezende, and G. Batista. Influence of graph construction on semi-supervised learning. In European Conference on Machine Learning, pages 160–175, 2013.Google Scholar
  24. 24.
    V. Takala, T. Ahonen, and M. Pietikäinen. Block-based methods for image retrieval using local binary patterns. In Image Analysis, SCIA, volume LNCS, 3540, 2005.CrossRefGoogle Scholar
  25. 25.
    O. Tuzel, F. Porikli, and P. Meer. A fast descriptor for detection and classification. In European Conf. on Computer Vision, pages 589–600, 2006.Google Scholar
  26. 26.
    G. Wacquet, E. P. Caillault, D. Hamad, and P. A. Hebert. Constrained spectral embedding for k-way data clustering. Pattern Recognition Letters, 34(9):1009–1017, 2013.CrossRefGoogle Scholar
  27. 27.
    X. Wang, B. Qian, and I. Davidson. On constrained spectral clustering and its applications. Data Mining and Knowledge Discovery, 28(1):1–30, 2014.MathSciNetCrossRefGoogle Scholar
  28. 28.
    L. Wolf, T. Hassner, and Y. Taigman. Descriptor based methods in the wild. In Faces in Real-Life Images Workshop in ECCV, 2008.Google Scholar
  29. 29.
    D. Xu and Y. Tian. A comprehensive survey of clustering algorithms. Annals of Data Science, 2(2):165–193, 2015.MathSciNetCrossRefGoogle Scholar
  30. 30.
    P. Yang, Q. Zhu, and B. Huang. Spectral clustering with density sensitive similarity function. Knowledge-Based Systems, 24:621–628, 2011.CrossRefGoogle Scholar
  31. 31.
    L. Zhang, S. Chen, and L. Qiao. Graph optimization for dimensionality reduction with sparsity constraints. Pattern Recognition, 45:1205–1210, 2012.CrossRefGoogle Scholar
  32. 32.
    L. Zhang, L. Qiao, and S. Chen. Graph-optimized locality preserving projections. Pattern Recognition, 43:1993–2002, 2010.CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Abdelmalik Moujahid
    • 1
    Email author
  • Fadi Dornaika
    • 1
    • 2
  • Blanca Cases
    • 1
  1. 1.University of the Basque Country UPV/EHUSan SebastianSpain
  2. 2.Ikerbasque FoundationSan SebastianSpain

Personalised recommendations