An Image Restoration Algorithm Based on Improved RBF Neural Network

Abstract:

In order to improve large amount of computing and slowly convergence speed, an improved radial basis function (RBF) neural network is raised in this paper. According to feature that the more recent data should be the more important in time-series data, it converts width value from original constant value to step function and accelerates the iterative convergence by using nearest neighbor clustering algorithm only at center, training weight by using gradient descent algorithm to correct network parameters and deleting input neurons adaptively. Network size is streamlined through network optimization training. Simulation shows that the restored image is good in visual and quantitative with faster image restoration processing. The algorithm based on improved RBF neural network has significantly improved the image restoration compared to other methods, but also well keeps image detail.

• Periodical: Advanced Materials Research (Volumes 430-432)
• Pages: 1671-1676
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.430-432.1671
• Online since: January 2012
• Authors: Xue Zhang Zhao, Zhi Yuan Liu, Yun Jiang Xi
• Keywords: Image Degradation, Image Restoration, Neural Network (NN)

References

[1] Castleman K R ED, zhigang Zhu M. Digital Image Processing [M]. beijing: Electronics Industry Press, 1998, 9.

[2] . Koch S, Kaufman H, Biemond J. Restoration of spatially varying blurred images using multiple model-based extended Kalman filters[J]. IEEE Transactions on Image Processing, 1995, 4(4)520-523.

DOI: 10.1109/83.370684

[3] Hongxing Li, Peizhuang Wang. Fuzzy Mathematics. beijing: National Defence Industry Press. 1994: 26-31.

[4] Peizhuang Wang, Hongxing Li. Fuzzy system theory and fuzzy computer. Beijing: Science Press. 1996: 8-13.

[5] Naiyao Zhang, Pingfan Yan, Neural networks and fuzzy control. Beijing: Qinghua University Press. 1998: 25-30.

[6] Shiyong Li. Neural control and intelligent control theory. Haerbin : Haerbin Institute of Technology Press. 1998: 52-66.

[7] LAMPTON M. Damping-undraping strategies for the levenberg-marquardt nonlinear least-squares method [J]. Computers in Physics Joumal, 1997, 11(11): 110-115.

DOI: 10.1063/1.168600

[8] DUBOISE. The sampling and reconstruction of time-varying imagery with application in video systems [J]. Proceeding of IEEE, 1985, 73(4): 502-522.

DOI: 10.1109/proc.1985.13182

[9] Qian W, Clarke L P. Wavelet-based neural network with fuzzy-logic adaptively for nuclear Image restoration . In: proceedings of IEEE, 1996, 84(10): 1458-1473.

DOI: 10.1109/5.537111

[10] B. Jeong and Y. H. Lee. Design of Weighted Order Statistic Filters Using the Perception Algorithm. IEEE Trans. Signal Processing. 2005, 42 (11): 3264~3269.

DOI: 10.1109/78.330393

[11] T. Sun and Y. Neuvo. Detail-preserving Median Based Filters in Image Processing. Pattern Recognit. Letter. 2005, 15(4): 341~347.

DOI: 10.1016/0167-8655(94)90082-5

[12] T. Chen and H. R. Wu. Adaptive Impulse Detection Using Center-weighted Median Filters. IEEE Signal Process. Letter. 2001, 8(1): 1~3.

DOI: 10.1109/97.889633