Advertisement

A New Graph-Based Image Segmentation Algorithm

  • Qian Zhang
  • Fujian Feng
  • Lin Xin
  • Lin Wang
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 208)

Abstract

Based on graph theory, we choose two-dimensional Gaussian as a dynamic adaptive index for weighting function, difference function Dex for inter-area and Din for one area were defined by structural similarity index (SSIM), the function determines different area to be merged or segmented is achieved. The algorithm was implemented on Mat lab successfully. Experimental results show that the algorithm the segmentation is better than others in effect and calculate time.

Keywords

Image segmentation Graph Two-dimensional gaussian distribution SSIM 

References

  1. 1.
    Liu J, Wu J (2010) An improved image segmentation based on gradient-threshold histogram method. Comput Digital Eng 38(4):131–133MATHGoogle Scholar
  2. 2.
    Duarte A, Sanchez A, Fernandez F, Antonio S (2006) Monte mayor Improving image segmentation quality through effective region merging using a hierarchical social met heuristic. Pattern Recogn Lett 55(26):1239–1251CrossRefGoogle Scholar
  3. 3.
    Liu X, Wang D (2006) Image and texture segmentation using local spectral histograms. IEEE Trans Image Process 15(10):3066–3077CrossRefGoogle Scholar
  4. 4.
    Leahy R (1993) An optimal graph theoretic approach to data clustering: theory and its application to image segmentation. IEEE Trans Pattern Anal Mach Intell 15(11):1101–1113CrossRefGoogle Scholar
  5. 5.
    Shi J, Malk J (2000) Normalized cuts and image segmentation. IEEE Trans Pattern Anal Mach Intell 22(8):888–905CrossRefGoogle Scholar
  6. 6.
    Felzenszwalb PF, Huttenlocher DP (2004) Efficient graph-based image segmentation. Int J Comput Vision 59(2):167–181CrossRefGoogle Scholar
  7. 7.
    Zhang T (2011) An improved image segmentation method based on graph. J West China Univ Nat Sci 30(1):61–64Google Scholar
  8. 8.
    Liu DB, Hou MS, Qu ZX, Wu H (2011) An efficient dynamically updated algorithm for shortest path tree. Comput Sci 38(7):96–99Google Scholar

Copyright information

© Springer-Verlag London 2013

Authors and Affiliations

  1. 1.Pattern Recognition and Intelligent Systems Laboratory of GuizhouGuiyangChina
  2. 2.Academic Affairs Office of Guizhou University for NationalitiesGuiyangChina

Personalised recommendations