Fully Symmetric Convolutional Network for Effective Image Denoising
Source abstract
Abstract
Neural-network-based image denoising is one of the promising approaches to deal with problems in image processing. In this work, a deep fully symmetric convolutional–deconvolutional neural network (FSCN) is proposed for image denoising. The proposed model comprises a novel architecture with a chain of successive symmetric convolutional–deconvolutional layers. This framework learns convolutional–deconvolutional mappings from corrupted images to the clean ones in an end-to-end fashion without using image priors. The convolutional layers act as feature extractor to encode primary components of the image contents while eliminating corruptions, and the deconvolutional layers then decode the image abstractions to recover the image content details. An adaptive moment optimizer is used to minimize the reconstruction loss as it is appropriate for large data and noisy images. Extensive experiments were conducted for image denoising to evaluate the FSCN model against the existing state-of-the-art denoising algorithms. The results show that the proposed model achieves superior denoising, both qualitatively and quantitatively. This work also presents the efficient implementation of the FSCN model by using GPU computing which makes it easy and attractive for practical denoising applications.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Image and Signal Denoising Methods · Advanced Image Fusion Techniques · Advanced Image Processing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Graduate Institute of Applied Science and Engineering, Fu Jen Catholic University, New Taipei City 24205, Taiwan
Department of Electrical Engineering, Fu Jen Catholic University, New Taipei City 24205, Taiwan
References (39)
Show source-provided references
1. Hyperspectral image denoising via noise adjusted iterative low-rank matrix approximation10.1109/JSTARS.2015.2398433 ↗
2. Image denoising by sparse 3d transform domain collaborative filtering10.1109/TIP.2007.901238 ↗
3. Poisson noise reduction with non-local PCA10.1007/s10851-013-0435-6 ↗
4. Xu, J., Zhang, L., Zuo, W., Zhang, D., and Feng, X. (, January 7–December). Patch group based nonlocal self-similarity prior learning for image denoising. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.
5. A survey of sparse representation: Algorithms and applications10.1109/ACCESS.2015.2430359 ↗
6. Nonlocally centralized sparse representation for image restoration10.1109/TIP.2012.2235847 ↗
7. Zha, Z., Liu, X., Huang, X., Hong, X., Shi, H., Xu, Y., Wang, Q., Tang, L., and Zhang, X. (2017, January 10–14). Analyzing the group sparsity based on the rank minimization methods. Proceedings of the IEEE International Conference on Multimedia and Expo (ICME), Hong Kong, China.10.1109/ICME.2017.8019334 ↗
8. Sparse modeling for image and vision processing10.1561/0600000058 ↗
9. A decomposition framework for image denoising algorithms
10. Zhang, R., Bouman, C.A., Thibault, J.B., and Ken, D.S. (2014, January 3–5). Gaussian mixture markov random field for image denoising and reconstruction. Proceedings of the IEEE Global Conference on Signal and Information Processing, Austin, TX, USA.10.1109/GlobalSIP.2013.6737083 ↗
11. Gu, S., Zhang, L., Zuo, W., and Feng, X. (2014, January 25). Weighted nuclear norm minimization with application to image denoising. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, OH, USA.10.1109/CVPR.2014.366 ↗
12. Deep learning10.1038/nature14539 ↗
13. Burger, H.C., Schuler, C.J., and Harmeling, S. Image denoising: Can plain neural networks compete with BM3D? In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, USA, 16–21 June 2012; pp.10.1109/CVPR.2012.6247952 ↗
14. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23–28). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.81 ↗
15. Hong, S., You, T., Kwak, S., and Han, B. (2015, January 6–11). Online tracking by learning discriminative saliency map with convolutional neural network. Proceedings of the Conference on Machine Learning, Lille, France.
16. 3D convolutional neural networks for human action recognition10.1109/TPAMI.2012.59 ↗
17. Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012, January 3–8). Imagenet classificationwith deep convolutional neural networks. Proceedings of the Advances in Neural Information Processing Systems, Lake Tahoe, CA, USA.
18. Ioffe, S., and Szegedy, C. (2015, January 6–11). Batch normalization: Accelerating deep network training by reducing internal covariate shift. Proceedings of the Conference on Machine Learning, Lille, France.
19. A novel approach for denoising and enhancement of extremely low-light video10.1109/TCE.2015.7064113 ↗
20. On fast bilateral filtering using fourier kernels10.1109/LSP.2016.2539982 ↗
21. Image denoising using scale mixtures of gaussians in the wavelet domain10.1109/TIP.2003.818640 ↗
22. Global image denoising10.1109/TIP.2013.2293425 ↗
23. Adaptive multi-column deep neural networks with application to robust image denoising
24. Long, J., Shelhamer, E., and Darrell, T. (2015, January 7–12). Fully convolutional networks for semantic segmentation. Proceedings of the Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.10.1109/CVPR.2015.7298965 ↗
25. Image super-resolution using deep convolutional neural networks10.1109/TPAMI.2015.2439281 ↗
26. Wang, X., Tao, Q., Wang, L., Li, D., and Zhang, M. (2015, January 15–17). Deep convolutional architecture for natural image denoising. Proceedings of the Conference on Wireless Communications and Signal Processing, Nanjing, China.10.1109/WCSP.2015.7341021 ↗
27. Schmidt, U., and Roth, S. (2014, January 23–28). Shrinkage fields for effective image restoration. Proceedings of the Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.349 ↗
28. Beyond a Gaussian denoiser: Residual learning a deep cnn for image denoising10.1109/TIP.2017.2662206 ↗
29. Noh, H., Hong, S., and Han, B. (2015, January 7–13). Learning deconvolution network for semantic segmentation. Proceedings of the Conference on Computer Vision, Santiago, Chile.10.1109/ICCV.2015.178 ↗
30. Pinheiro, P.H., and Collobert, R. (2014, January 21–26). Recurrent convolutional neural networks for scene labeling. Proceedings of the Conference on Machine Learning, Beijing, China.
31. Eigen, D., Krishnan, D., and Fergus, R. (2014, January 3). Restoring an image taken through a window covered with dirt or rain. Proceedings of the Conference on Computer Vision, Sydney, Australia.10.1109/ICCV.2013.84 ↗
32. Tompson, J., Jain, A., LeCun, Y., and Bregler, C. (2014, January 8–13). Joint training of a convolutional network and a graphical model for human pose estimation. Proceedings of the Conference on Neural Information Processing Systems, Montreal, QC, Canada.
33. Trainable nonlinear reaction diffusion: A flexible framework for fast and effective image restoration10.1109/TPAMI.2016.2596743 ↗
34. Kingma, D.P., and Ba, J.L. (2015, January 7–9). ADAM: A method for stochastic optimization. Proceedings of the Conference on learning representations, San Diego, CA, USA.
35. Zeiler, M.D., and Fergus, R. (2014). Visualizing and understanding convolutional networks. European Conference on Computer Vision, Springer.10.1007/978-3-319-10590-1_53 ↗
36. Liu, H., Xiong, R., Zhang, J., and Gao, W. (2015, January 7–12). Image denoising via adaptive soft-thresholding based on non-local samples. Proceedings of the Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.10.1109/CVPR.2015.7298646 ↗
37. Chen, F., Zhang, L., and Yu, H. (2015, January 7–13). External patch prior guided internal clustering for image denoising. Proceedings of the Conference on Computer Vision, Santiago, Chile.10.1109/ICCV.2015.76 ↗
38. Hong, S., Noh, H., and Han, B. (2015, January 7–10). Decoupled deep neural network for semi-supervised semantic segmentation. Proceedings of the Conference on Advances in Neural Information Processing Systems, Montreal, QC, Canada.
39. Burger, H.C., Schuler, C.J., and Harmeling, S. (2013). Learning how to combine internal and external denoising methods. German Conference on Pattern Recognition, Springer.10.1007/978-3-642-40602-7_13 ↗