Applications of neural networks with Residial architecture for filtration of impulse noise on images
Abstract
Currently, image processing is one of the fastest growing data processing area. Image data becomes corrupted by noise during the capture and transmission of data over communication channels. Every year data size of the images grows up. Thus, removing noise from images is increasingly relevant task. Recently, neural network approaches of solving this problem become more popular. Additional blocks are used for training deep neural networks, that significantly affects the performance. This paper proposes a neural network architecture with residual blocks that allows filtering with high quality and performance.
About the Authors
D. Salnikov
Национальный технический университет «Харьковский политехнический институт»
Ukraine
Ukraine
O. Vasylchenkov
Национальный технический университет «Харьковский политехнический институт»
Ukraine
Ukraine
References
1. D. Brownrigg. “The weighted median filter,” Commun. Assoc. Comput. Mach., vol. 27, pp. 807-818, Mar. 1984.
2. S. J. Ko and Y. H. Lee, “Center weighted median filters and their applications to image enhancement,” IEEE Trans. Circuits Syst., vol. 38, pp. 984-993, 1991.
3. Bellanger M. G. Adaptive digital filters. 2nd ed. Marcel Dekker, 2001.
4. Haykin S. Adaptive filter theory. 4th ed. Prentice Hall, 2001.
5. Sayed A. H. Fundamentals of adaptive filtering. John Wiley and Sons, 2003.
6. Adaptive signal processing: applications to real-world problems.J. Benesty, Y. Huang, Eds. Springer, 2003.
7. Poularikas A. D., Ramadan Z. M. Adaptive filtering premier with MATLAB. CRC Press, 2006.
8. Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising Kai Zhang, Wangmeng Zuo, Senior Member, IEEE, Yunjin Chen, Deyu Meng, Member, IEEE, and Lei Zhang Senior Member, IEEE.
9. Harold C. Burger, Christian J. Schuler, «Image denoising: Can plain Neural Networks compete with BM3D», IEEE International Conference on Computer Vision and Pattern Recognition 2012.
10. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift Sergey Ioffe, Christian Szegedy (Submitted on 11 Feb 2015 (v1), last revised 2 Mar 2015 (this version, v3)).
11. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 770-778.
Review
For citations:
Salnikov D., Vasylchenkov O. Applications of neural networks with Residial architecture for filtration of impulse noise on images. Bulletin of Shakarim University. Technical Sciences. 2022;(1(5)):13-16. (In Russ.)
Views:
156
Email this article 
