Aerial Scene Classification and Information Retrieval via Fast Kernel Based Fuzzy C-Means Clustering

  • Zhengmao YeEmail author
  • Hang Yin
  • Yongmao Ye
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 898)


Fast kernel based fuzzy C-Means clustering is proposed in this article to accomplish both accurate and robust segmentation, via integration with watershed transform and fast level set schemes. Aerial scenes are inherently linked to (noise and artifact) sensitivities, intensity inhomogeneity, blurry boundary, and information complexity. It is thus necessary to combine the edge or contour based level set method with region based fuzzy C-Means clustering. To achieve fast segmentation, watershed transform is used to secure the initial contour of the fast level set method, so that initial cluster centers of fuzzy C-Means clustering are selected on those closed contour to avoid misclassification and to enhance separability. It reduces time for lengthy computation iteration. Using multiple densely distributed aerial images, robust and fast clustering is observed after comparing between classical and fast kernel based fuzzy C-Means clustering. To further analyze the role of hybrid fast kernel based scheme on scene classification and information retrieval, frequency domain histogram analyses for several clustering cases are conducted on aerial digital images.


Fuzzy C-Means clustering Gaussian kernel Fast level set Watershed transform Histogram analysis 


  1. 1.
    Duda, R., Hart, P., Stork, D.: Pattern Classification, 2nd edn. Wiley, Hoboken (2001)zbMATHGoogle Scholar
  2. 2.
    Gonzalez, R., Woods, R.: Digital Image Processing, 3rd edn. Prentice Hall, Upper Saddle River (2007)Google Scholar
  3. 3.
    Schilling, R., Harris, S.: Fundamental of Digital Signal Processing using MATLAB. Cengage Learning, Stamford (2005)Google Scholar
  4. 4.
    Ye, Z., Mohamadian, H., Yin, H., Ye, Y.: Integration of fuzzy C-Means clustering and fast level set for aerial RGB image segmentation. In: Proceedings of 2015 International Brazilian Meeting on Cognitive Science, Sao Paulo, Brazil, 7–11 December 2015 (2015)Google Scholar
  5. 5.
    Li, B., Chui, C., Chang, S.: Integrating spatial fuzzy clustering with level set methods for automated medical image segmentation. Comput. Biol. Med. 41, 1–10 (2011)CrossRefGoogle Scholar
  6. 6.
    Ye, Z., Mohamadian, H., Ye, Y.: Gray level image processing using contrast enhancement and watershed segmentation with quantitative evaluation. In: Proceedings of 2008 IEEE International Conference on Content-Based Multimedia Indexing, London, UK, 18–20 June 2008, pp. 470–475 (2008)Google Scholar
  7. 7.
    Saikumar, T., Yugander, P., Murthy, P., Smitha, B.: Image segmentation algorithm using watershed transform and fuzzy C-Means clustering on level set method. Int. J. Comput. Theory Eng. 5, 209–213 (2013)CrossRefGoogle Scholar
  8. 8.
    Ye, Z., Ye, Y., Mohamadian, H.: Fuzzy filtering and fuzzy K-Means clustering on biomedical sample characterization. In: Proceedings of the 2005 IEEE International Conference on Control Applications, Toronto, Canada, pp. 90–95 (2005)Google Scholar
  9. 9.
    Ye, Z., Mohamadian, H.: The role of quantitative metrics in enhancing spatial information retrieval via fuzzy C-Means clustering. In: Proceedings of 2011 International Symposium on Remote Sensing of Environment, Australia, 10–15 April 2011 (2011)Google Scholar
  10. 10.
    Fazendeiro, P., Oliveira, J.: Observer-biased fuzzy clustering. IEEE Trans. Fuzzy Syst. 23(1), 85–97 (2015)CrossRefGoogle Scholar
  11. 11.
    Ye, Z., Mohamadian, H.: Enhancing decision support for pattern classification via fuzzy entropy based fuzzy C-Means clustering. In: Proceedings of the 2013 52nd IEEE Conference on Decision and Control, Florence, Italy, 10–13 December 2013, pp. 7432–7436 (2013)Google Scholar
  12. 12.
    Connor, B., Roy, K., Shelton, J., Dozier, G.: Iris recognition using fuzzy level set and GEFE. Int. J. Mach. Learn. Comput. 4(3), 225–231 (2014)CrossRefGoogle Scholar
  13. 13.
    Alipour, S., Shanbehzadeh, J.: Fast automatic medical image segmentation based on spatial kernel fuzzy C-Means on level set method. Mach. Vis. Appl. 25, 1469–1488 (2014)CrossRefGoogle Scholar
  14. 14.
    Arabe, S., Gao, X., Wang, B.: A fast and robust level set method for image segmentation using fuzzy clustering and lattice boltzmann method. IEEE Trans. Cybern. 43, 910–920 (2013)CrossRefGoogle Scholar
  15. 15.
    Gong, M., Liang, Y., Shi, J., Ma, W., Ma, J.: Fuzzy C-Means clustering with local information and kernel metric for image segmentation. IEEE Trans. Image Process. 22(2), 573–584 (2013)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.College of EngineeringSouthern UniversityBaton RougeUSA
  2. 2.Liaoning Radio and Television StationShenyangChina

Personalised recommendations