Research on Image Super-Resolution Reconstruction Based on BPNN and RBFNN

Abstract:

To overcome inherent resolution limitation of imaging system and provide a high resolution satellite image economically and practically, the paper researched on image super-resolution reconstruction (SRR) based on BPNN and RBFNN. Learning sample images are obtained according to image observation model, input/output mapping pairs are formed by way of vector mappings, and then two NN models of SRR are established respectively. Simulation and generalization tests are carried on the BPNN and RBFNN respectively. Results indicate that reconstruction performance of BPNN is better than that of RBFNN in aspect of reconstruction time-consuming, convergence speed and effect of SRR in generalization tests, although those parameters are better in simulation tests of RBFNN. So BPNN is more suitable for project applications of image super-resolution reconstruction.

• Periodical: Applied Mechanics and Materials (Volumes 20-23)
• Pages: 445-451
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.20-23.445
• Online since: January 2010
• Authors: Fu Zhen Zhu, Jin Zong Li, Bing Zhu, Dong Dong Li, Dong Dong Ma
• Keywords: BP Neural Network (BPNN), Neural Network (NN), Radial Basis Function Neural Network (RBFNN), Super-Resolution Reconstruction, Vector Mapping

References

[1] S.C. Park, M.K. Park and M. G. Kang. Super-resolution Image Reconstruction: A Technical Overview, IEEE Signal Processing Magazine. vol. 20(2003), pp.21-36.

DOI: 10.1109/msp.2003.1203207

[2] Y. Lu, M. Inamura, M. del Carmen Valdes. Super-resolution of the undersampled and subpixel shifted image sequence by a neural network, Imaging systems and technology. vol. 14 (2004), pp.8-15.

DOI: 10.1002/ima.20001

[3] M. Del. Carmen Valdes, M. Inamura. Spatial resolution improvement of remotely sensed images by a fully interconnected neural network approach, Trans. on geoscience and remote Sensing, vol. 38 (2000), pp.2426-2430.

DOI: 10.1109/36.868898

[4] M del Carmen Valdes, M Inamura. Saptial resolution improvement of a low spatial resolution thermal infrared images by backpropagated neural networks, IEICE Trans. on Inform and Sys. vol. E81-D(1998), pp.872-880.

[5] M Forti. Some extensions of a new method to analyze complete stability of neural networks. IEEE Trans. on Neural Networks, vol. 13 (2002), pp.1230-1238.

DOI: 10.1109/tnn.2002.1031956

[6] V. Anastassopoulos. Fusion and super-resolution in multispectral data using neural networks for improved RGB representation, The Imaging Science Journal, vol. 53 (2005), p: 83-94.

DOI: 10.1179/136821905x43945

[7] D. Xu, Z. Wu. In: Systems analysis and design based on Matlab--�eural network, Xidian University Press, Xian(2003), in press.

[8] T. Mitchell. in: Machine Learning, China Machine Press, Beijing(2003).

[9] M. Liu, H. N. Liu, W. D. Liu, et al. Reconstruction of super-resolution image based on neural network, Journal of Harbin Institute of Technology, vol. 35(2003), pp.707-710.

[10] W. Q. Liu, W. H. Li. An algorithm for generalized RBF neural network based on self-organizing selection center, Journal of Xinyang Normal University (Natural Science Edition), vol. 20(2007), pp.515-517.

[11] A Guillen, I Rojas, J Gonzalez, et al. Supervised RBFNN centers and radii initialization for function approximation problems, International Joint Conference on Neural Networks, Canada, 2006, pp.3175-3180.

DOI: 10.1109/ijcnn.2006.247301

[12] S. Chen, C.F.N. Cowan, P.M. Grant. Orthogonal least squares learning algorithm for radial basis function networks, IEEE Trans. on neural networks, vol. 2 (1991), pp.302-309.

DOI: 10.1109/72.80341

[13] T.N. Liu, X.Z. Guan, Z. Y. Liu, et al. RBF neural network with optimal selection cluster algorithm and its application, Proceedings of the 4th World Congress on Intelligent Control and Automation. Shanghai, China.

DOI: 10.1109/wcica.2002.1020813

[14] P. F . Yan. Capacity, learning and computational complexity of artificial neural networks, Chinese Journal of Electronics, vol. 23(1995), pp.63-67.

[15] P. F . Yan. Intelligent Information Processing and Research of Neural Networks, vol. 16( 2001), pp.10-13.