Call for Paper - November 2023 Edition
IJCA solicits original research papers for the November 2023 Edition. Last date of manuscript submission is October 20, 2023. Read More

Regional Decomposition of Images using Three Parameter Logistic Type Mixture Model with K-Means

International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Year of Publication: 2018
K. V. Satyanarayana, K. Srinivasa Rao, P. Srinivasa Rao

K V Satyanarayana, Srinivasa K Rao and Srinivasa P Rao. Regional Decomposition of Images using Three Parameter Logistic Type Mixture Model with K-Means. International Journal of Computer Applications 181(32):13-20, December 2018. BibTeX

	author = {K. V. Satyanarayana and K. Srinivasa Rao and P. Srinivasa Rao},
	title = {Regional Decomposition of Images using Three Parameter Logistic Type Mixture Model with K-Means},
	journal = {International Journal of Computer Applications},
	issue_date = {December 2018},
	volume = {181},
	number = {32},
	month = {Dec},
	year = {2018},
	issn = {0975-8887},
	pages = {13-20},
	numpages = {8},
	url = {},
	doi = {10.5120/ijca2018918213},
	publisher = {Foundation of Computer Science (FCS), NY, USA},
	address = {New York, USA}


For image analysis image decomposition or segmenting the images is a basic requirement. For decomposing the images probability models play a vital role. This paper addresses image decomposition using three parameter logistic type mixture distribution. Here it is assumed that the pixel intensities of image region follow a three parameter logistic type probability distribution. The estimation of parameters is carried utilizing Expectation and Maximization algorithm. The initialization of the parameters is done with K-means algorithm and moment method of estimation the number of image regions is obtained counting the peaks of the histogram drawn for the pixel intensities of the whole image. The decomposition algorithm (segmentation) is developed under maximum component likelihood function with Bayesian considerations. The efficiency of the proposed algorithm is studied by computing the metrics for segmentation such as GCE, VOI, PRI. The experimentation is conducted with five randomly chosen images taken from Berkeley image database revealed that the proposed algorithm is superior to the other model based segmentation algorithms for some images, which are having laptykurtic image regions.A comparative study with that of segmentation algorithm based on GMM is also presented.


  1. Srinivas.Y and SrinivasRao.K (2007), “Unsupervised image segmentation using finite doubly truncated Gaussian mixture model and Hierarchical clustering ”, Journal of Current Science,Vol.93,No.4, pp.507-514.
  2. T. Yamazaki (1998) “Introduction of algorithm into color image segmentation,” Proceedings of ICIRS’98, pp. 368–371
  3. T.Lie et al.(1993), “ Performance evaluation of Finite normal mixture model based image segmentation, IEEE Transactions on Image processing, Vol.12(10), pp 1153-1169.
  4. Z.H.Zhang et al (2003).“ EM Algorithms for Gaussian Mixtures with Split-and-mergeOperation”, Pattern Recognition, Vol. 36(9),pp1973-1983.
  5. T. Jyothirmayi, et al(2016)., “Image Segmentation Based on Doubly Truncated Generalized Laplace Mixture Model and K Means Clustering International Journal of Electrical and Computer Engineering (IJECE), Vol. 6, No. 5, October 2016, pp. 2188~2196.
  6. T. Jyothirmayi, et al (2017).,“Performance Evaluation of Image Segmentation Method based on Doubly Truncated Generalized Laplace Mixture Model and Hierarchical Clustering” .J. Image, Graphics and Signal Processing, 2017, 1, 41-49
  7. The Berkeley segmentation dataset CS/vision/bsds/BSDS300/html/dataset/images.html.
  8. SrinivasRao. K,C.V.S.R.Vijay Kumar, J.LakshmiNarayana, (1997) “ On a New Symmetrical Distribution ”, Journal of Indian Society for Agricultural Statistics, Vol.50(1), pp 95-102
  9. M.Seshashayee, K.Srinivasarao, Ch.Satyanarayana and P.Srinivasarao- (2011) – Studies on Image Segmentation method Based on a New Symmetric Mixture Model with K – Means, Global journal of Computer Science and Technology, Vol.11, No.18, pp.51-58. ISSN: 0975-4172,0975-4350
  10. GVS.Rajkumar, K.Srinivasarao, and P.Srinivasarao-(2011) – Studies on color Image segmentation technique based on finite left truncated Bivariate Gaussian mixture model with k - means, Global Journal of computer Science and Technology, Vol.11, No.18, pp 21- 30. ISSN: 0975-4172,0975-4350
  11. GVS.Rajkumar K.Srinivasa Rao P.Srinivasarao (2011)studies on color image segmentation technique based on finite left truncated bivariate Gaussian mixture model with k-means, global journal of computer science and technology, vol.11, no.18. Issn: 0975-4172, 0975-4350
  12. Mclanchlan G. And Krishnan T (1997)., “ The EM Algorithm and Extensions”, John Wiley and Sons, New York -1997.
  13. Mclanchlan G. and Peel.D (2000), “ The EM Algorithm For Parameter Estimations”, John Wiley and Sons, New York -2000.
  14. Jeff A.Bilmes (1997), “ A Gentle Tutorial of the EM Algorithm and its application to Parameter Estimation for Gaussian Mixture and Hidden Markov Models”, Technical Report, University of Berkeley, ICSI-TR-97-021.
  15. R.Unnikrishnan, C,Pantofaru, and M.Hernbert ( 2007), “ Toward objective evaluation of image segmentation algorithms,” IEEE Trans. Pattern Analysis and Machine Intelligence . Vol.29, no.6, pp.929-944.


Image decomposition,Three parameter logistic type mixture distribution, Expectation and Maximization algorithm, Metrics of segmentation,K-means algorithm.