International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

A Deep Learning and Machine Learning Approach for Image Denoising

Author : Ashly Roy ¹ Anju P ² Linnet Tomy ³ Dr.M.Rajeswari ⁴

Date of Publication :20th May 2021

Abstract: Image denoising is the way toward elimi- nating any distortions in an image and numerous pro- cedures exists for this reason. The traditional image denoising methods comprises of spatial domain filter- ing and transform domain filtering. These techniques eliminated commotion to a more prominent stretch out from an image yet neglected to preserve image textures and informationâ€™s. To counter this issue deep learning came right into it. Itâ€™s strong learning capacity made convolutional neural network the best and exact answer for image denoising. This paper discusses on one such CNN approach that utilize LeNet architecture for image denoising.We additionally presented the machine learning approach in image denoising to get an obvious thought on which learning is more effective in eliminating the distortions. K-Nearest Neighbor, Na ÌˆÄ±ve Bayes, Support Vector Machine, Random Forest and Decision Tree are the main five algorithms utilized in the machine learning approach. At long last, a correlation is made between machine learning algorithms and deep learning approach based on accuracy and an effective algorithm for image denoising is found.

Reference :

1. Tian, C., Xu, Y., Zuo, ”Image denoising using deep CNN with batch renormalization,” Neural Networks,121, 461–473,2020.
2. Chunwei Tian, Yong Xu, Zuoyong Li, Wangmeng Zuo, Lunke Fei, Hong Liu, ”Attention-guided CNN for image de-noising,” Neural Networks,124, 117– 129,2020.
3. Peng, Y., Zhang, L., Liu, S., Wu, X., Zhang, Y., Wang, X, “Dilated residual networks with symmetric skip connection for image denoising,” Neurocomputing, 345, 67–76,2019.
4. Chuanhui Shana, Xirong Guob, and Jun Oua, “Residual learning of deep convolutional neural networks for image denoising,” Journal of Intelligent Fuzzy Systems, 37, 2809–2818,2019.
5. Rini Smita Thakur , Ram Narayan Yadav, Lalita Gupta, “State-of-art analysis of image denoising methods using convolutional neuralnetworks,” IET Image Processing, vol. 13 Iss. 13, pp. 2367- 2380,2019.
6. Sridhar, “Digital image processing” Oxford Publications,New Delhi, India, 2016, 2nd edn., pp. 1– 7.
7. Zha, Z., Liu, X., Huang, X., Shi, H., Xu, Y., Wang, Q., et al, “Analyzing the group sparsity based on the rank minimization methods,” In 2017 IEEE international conference on multimedia and expo (pp. 883–888). IEEE,2017.
8. Mairal, J., Bach, F. R., Ponce, J., Sapiro, G., Zisserman,A, “Non-local sparse models for image restoration,” In ICCV, Vol. 29 (pp. 54–62). Citeseer,2009.
9. Malfait, M., Roose, “Wavelet-based image denoising using a Markov random field a priori model,” IEEE Transactions on image processing, 6(4), 549– 565,1997.
10. Chambolle, A. “An algorithm for total variation minimization and applications,” Journal of Mathematical imaging and vision, 20(1–2), 89–97,2004.
11. Danielyan, A., Katkovnik, V., Egiazarian, K.: “BM3D frames and variational image deblurring”, IEEE Trans. Image Pro- cess., 2012, 21, (4), pp. 1715–1728.
12. S. Gu , L. Zhang , W. Zuo , X. Feng, “Weighted nuclear norm minimization with application to image denoising,” in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014, pp. 2862–2869.
13. Beck, A., Teboulle, M., “Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems,” IEEE