A Rapid Method for Image Compression Based on Wavelet Transform and SOFM Neural Network

Abstract:

The current self-organizing feature map (SOFM) neural network algorithm used for image compression, of which a large amount of network training time and the blocking effect in the reconstructed image existed in codebook design vector calculation. Based on the above issue, this paper proposed an improved SOFM. The new SOFM introduced normalized distance between the sum of input vectors and the sum of the codeword vectors as a constraint in the process of searching for the winning neuron, which can remove redundant Euclidean distance calculation in the competitive process. Furthermore, this paper has done image compression by combining wavelet transform with the improved SOFM (WT & improved SOFM). The method firstly conducted wavelet decomposition for the image, retained low-frequency sub-band, then put the high-frequency sub-band into improved SOFM network, and achieved the purpose of compression. Experimental results showed that this algorithm can greatly reduce the network training time and enhance the learning efficiency of neural network, while effectively improve the PSNR (increased 0.6dB) of reconstructed.

• Periodical: Applied Mechanics and Materials (Volumes 135-136)
• Pages: 126-131
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.135-136.126
• Online since: October 2011
• Authors: Hong Ke Xu, Wei Song Yang, Jian Wu Fang, Chang Bao Wen, Wei Sun
• Keywords: Image Compression, Self-Organizing Feature Map Neural Network, Vector Quantitative, Wavelet Transform (WT)

References

[1] L. Wang, G. X. Shao, Y. He, Optimization algorithm of adaptive image compression based on singular value decomposition, Journal of Lanzhou University of Technology, vol. 35, No. 5, 2009, pp.95-98.

[2] J. H. Tan, J. Zhang, Application of SOFM neural network in compression coding design, Computing Technology and Automation, vol. 26, No. 1, 2007, pp.22-25.

[3] N. V. Boul Gouris, D. Tzovaras, M. G. Strinntzis, Lossless image compression based on optimal prediction, adaptive lifting, and conditional arithmetic coding, IEEE Transactions on Image Processing, vol. 10, No. 1, 2001, pp.1-14.

DOI: 10.1109/83.892438

[4] S. K. Meher, A. K. Pradhan, G. Panda, An integrated data compression scheme for power quality events, Electric Power Systems Research, vol. 69, 2004, pp.213-220.

DOI: 10.1016/j.epsr.2003.10.001

[5] M. J. Gong, R. L. Wang, G. X. Li, Image vector quantum based on CL multi-wavelet and SOFM, Application Research of Computers, vol. 8, 2005, pp.238-240.

[6] Z. L. Luo, Image compression method based on wavelet and neural network, Journal of Shaoguan University·Natural Science, vol. 31, No. 3, 2010, pp.25-28.

[7] N. M. Nasrabadi, Y. Feng, Vector quantization of images based upon the Kohonen self-organizing feature maps, Proc of IEEE on Neural Networks, 1988, pp.101-108.

DOI: 10.1109/icnn.1988.23837

[8] M. M. Fu, R. Q. Lin, Self-organizing feature maps based image compression technology, Computer Engineering and Applications, vol. 43, No. 32, 2007, pp.38-39.

[9] F. Z. Zhu, B. Wu, SOFM Network and the realization of SOFM in Matlab, Micro-computer information (Embedded with the SOC), vol. 21, No. 12, 2005, pp.163-165.

[10] X. Zhu, Y. D. Wu, Image compression based on self-organizing feature map neural network, Computer Engineering, vol. 29, No. 20, 2003, pp.121-123.

[11] N. R. Pal et al, Generalized clustering networks and Kohonen's self-organizing scheme, IEEE Trans on neural network, vol. 4, No. 4, 1993, pp.549-557.

DOI: 10.1109/72.238310

[12] X. Cheng, Y. H. Yang, W. L. Chen, Application and analysis of self-organizing feature map, Journal of Chang Chun Teachers College (Natural Science), vol. 24, No. 4, 2005, pp.55-59.

[13] Y. B. Liu, The research of image compression based on the wavelet transform, Journal of Jilin Agricultural Science and Technology College, vol. 19, No. 3, 2010, pp.83-85.

[14] X. Chen, The discussion of wavelet transform and image compression, Education Study, vol. 4, 2006, pp.67-68.