International Journal of
Physical Sciences

  • Abbreviation: Int. J. Phys. Sci.
  • Language: English
  • ISSN: 1992-1950
  • DOI: 10.5897/IJPS
  • Start Year: 2006
  • Published Articles: 2557

Full Length Research Paper

An alternative approach for unsupervised cluster based image retrieval

P. Pattanasethanon1* and B. Attachoo2
  1Faculty of Information Technology, King Mongkut’s Institute of Technology Ladkrabang, Bangkok, 10520, Thailand. 2Faculty of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok, 10520, Thailand.
Email: [email protected]

  •  Accepted: 02 August 2012
  •  Published: 16 October 2012



In this research, we proposed an alternative method to retrieve images from its clusters which is determined by its color histogram. In other approach, it is called cluster based image retrieval model. One common problem with conventional content based image retrieval (CBIR) system is that the output data contains too many non-relevant feedbacks due to the indistinct symmetry of semantic gap. In this paper, two clustering techniques are performed to solve the problem mentioned above; including the misconception of semantic gap, called k-means and hierarchical clustering technique. The proposed method shows a promising retrieval procedure in terms of runtime errors reduction and accuracy. The experimental results determine that this alternative image retrieval model does not only retrieve images to the homogeneity of its features but it is also practically sufficient.


Key wordsContent based image retrieval, color histogram, K-means, clustering image retrieval.