Color Image Segmentation Based on Dynamic Model and Power Spectrum

Abstract:

In this paper, the locally coupled neural oscillator model and power spectrum are introduced to solve the color image segmentation problem. Firstly the single neuron oscillator model is developed, the model parameter is set so as to create limit circle simulating the activity of neuron. Then the locally coupled network model is developed, taking R, G, Bas external input separately, the network output three dynamic orbits representing R, G, B for each image point. Lastly, the power spectrum of the orbits is calculated for each image point, the color image is segmented using the calculated power spectrum. The performance is comparable to the results from other segmentation methods.

• Periodical: Advanced Materials Research (Volume 621)
• Pages: 364-369
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.621.364
• Online since: December 2012
• Authors: Yu Pu Chen, Yuan Hua Qiao, Jun Miao
• Keywords: Estimation, Image Segmentation, Neuronal Network, Power Spectrum, Vision

References

[1] C. von der Malsburg, The correlation theory of brain function, Max-Planck-Institute for Biophysical Chemistry, Internal Rep. 81-2 (1981).

[2] R.M. Haralick and L.G. Shapiro, in: Image segmentation techniques, Comput. Graph. Image Process., vol. 29 (1985), pp.100-132.

[3] N.R. Pal and S.K. Pal, A review on image segmentation techniques, Paattern Recognit, vol. 26(1993), pp.1277-1294.

[4] R. Kohler, A segmentation system based on thresholding, Comput. Graph. Image Process., vol. 15(1981), pp.319-338.

[5] A. Mitiche and J.K. Aggarwal, Detection of edges using range information, IEEE Trans. Pattern Anal. Machine Intell., vol. 5(1983), pp.174-178.

DOI: 10.1109/tpami.1983.4767369

[6] B. Bhanu,S. Lee C.C. Ho, and T. Henderson, Range data processing: Representation of surfaces by edges, in Proc. IEEE Int. Pattern Recognition Conf. (1986), pp.236-238.

[10] A. Hoover and G. Jean-Baptiste, et al., An experimental comparison of range image segmentation algorithms, IEEE Trans. Pattern Anal. Machine Intell., vol. 18(1996), pp.673-689.

DOI: 10.1109/34.506791

[11] S.M. Bhan darkar, J. Koh, and M. Suk, Multiscale Image segmentation using a hierarchical self-organizing map, Neurocomput. , vol. 14(1997), pp.241-272.

DOI: 10.1016/s0925-2312(96)00048-3

[12] A.L. Hodgkin and A.F. Huxey. A quantitative description of ion currents and its application in nerve membranes, J. Physiol, vol. 117(1952), pp.500-544.

[13] Liang Zhao, Chaotic synchronization for scene segmentation. International Journal of Modern Physics B Vol. 17, Nos. 22, 23 & 24 (2002), p.4387–4394.

[14] Liang Zhao, Chaotic synchronization in general network topology for scene segmentation. Neurocomputing vol. 71(2008), pp.3360-3366.

DOI: 10.1016/j.neucom.2008.02.024

[15] D. L. Wang and D. Terman, Locally excitatory globally inhibitory oscillator networks, IEEE Trans. Neural Networks, vol. 6(1995), p.283–286.

DOI: 10.1109/72.363423

[16] C. von der Malsburg, The correlation theory of brain function, Max-Planck-Inst. Biophys. Chemistry, Int(1981). Rep. 81-2.

[17] C. M. Gray, P. K¨onig, A. K. Engel, and W. Singer, Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties, Nature, vol. 338(1989), p.334–337.

DOI: 10.1038/338334a0

[18] Yuanhua Qiao, Yong Meng, Lijuan Duan, Faming Fang and Jun Miao, Qualitative analysis and application of locally coupled neural oscillator network, Neural Computing and Applications, Jan. (2012).

DOI: 10.1007/s00521-012-0829-1

[19] H.R. Wilson and J.D. Cowan, Excitatory and inhibitory interactions in localized populations of model neurons, Biophys.J. vol. 12(1972), pp.1-24.

DOI: 10.1016/s0006-3495(72)86068-5

[20] Butcher, John C. (2003).

[21] Dr. Ji Zhen, power spectrum analysis, Faculty of Information Engineering, SZU. (2002).