Abstract
Content-based image retrieval (CBIR) is the problem of searching for digital images in large databases. In order to search and index the images, different models of Self-organizing Maps (SOM), such as the Tree-Structured self labeled self-organizing maps (TS-SL-SOM), were used because combines features of data quantization and topology preservation with good results. TS-SL-SOM has a dynamic and data-driven tree topology with maps as nodes. From top to down, maps are segmented and each labeled region generates a new child map in the next level. For image retrieval, the feature vectors of images were mapped into the TS-SL-SOM neural network. Although it has presented good results when applied for indexing images, in comparison with other structures. In this paper a new SOM segmentation is proposed for improving the results in CBIR. The proposed algorithm is based in the well known idea that in the edges of segments the activation of neurons is less than in the center of segments. Better retrieval results were accomplished when using the proposed method in comparison with the traditional method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Chow, W.S.T., Rahman, M.K.M., Wu, S.: Content-based image retrieval by using tree-structured features and multilayer self-organizing map. Pattern Analysis Application 9, 1–20 (2006)
Laaksonen, J., Koskela, M., Laakso, S., Oja, E.: Picsom - content-based image retrieval with self-organizing maps. Pattern Recognition Letters 21(13-14), 1199–1207 (2000)
Wu, S., Rahman, M.K.M., Chow, T.: Content-based image retrieval using growing hierarchical self-organizing quadtree map. Pattern Recognition 38(5), 707–722 (2005)
Patiño-Escarcina, R.E., Costa, J.A.F.: Content-based image retrieval using descriptos hierarchy. In: Int. Conf. on Hybrid Intelligent Systems (2007)
Costa, J.A.F.: Classificação Automática e Análise de Dados por Redes Neurais Auto-organizaveis. PhD thesis, Facultade de Engenharia Elétrica e de Computação, Universidade Estadual de Campinas (1999)
Costa, J.A.F., Netto, M.A.: A new tree-structured self-organizing map for data analysis. In: Press, I. (ed.) Int. Joint Conf. on Neural Networks, Washington DC, pp. 1931–1936 (2001)
Costa, J.A.F., Netto, M.: Estimating the number of clusters in multivariate data by self-organizing maps. Int. Journal of Neural Systems 9(3), 195–202 (1999)
Costa, J.A.F., Netto, M.L.A.: Clustering of complex shaped data sets via kohonen maps and mathematical morphology. In: Dasarathy, B. (ed.) Proc. of the SPIE, Data Mining and Knowledge Discovery, vol. 4384, pp. 16–27 (2001)
Wu, B.Y.: On the intercluster distance of a tree metric. Theoretical Computer Science 369(1-3), 136–141 (2006)
Nene, A.S., Nayar, S.K., Murase, H.: Columbia object image library (coil-100). Technical report, CUCS-006-96 (February 1996)
Salembier, P., Smith, J.R.: Mpeg-7 multimedia, description schemes. IEEE Trans. on Circ. and Sys. For video Tech. 11(6), 748–759 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Patiño-Escarcina, R.E., Costa, J.A.F. (2009). A New Segmentation Approach in Structured Self-Organizing Maps for Image Retrieval. In: Corchado, E., Yin, H. (eds) Intelligent Data Engineering and Automated Learning - IDEAL 2009. IDEAL 2009. Lecture Notes in Computer Science, vol 5788. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-04394-9_54
Download citation
DOI: https://doi.org/10.1007/978-3-642-04394-9_54
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-04393-2
Online ISBN: 978-3-642-04394-9
eBook Packages: Computer ScienceComputer Science (R0)