Abstract:According to the theory of deep learning, the process of image denoising can be regarded as a fitting process of a neural network. In this paper, an image denoising algorithm based on composite convolutional neural network is proposed through constructing a simple and efficient composite convolutional neural network. The first stage includes two convolutional neural networks with two layers. Some initial convolutional kernels of convolutional neural network with three layers in the second stage are trained by these two networks, respectively. The training time in the second stage is decreased and the robustness of the network is enhanced. Finally, the learned convolutional neural network in the second stage is applied to denoise a new image with noises. Experimental results show that the proposed algorithm is comparable to state of the art image denoising algorithms in peak signal to noise ratio(PNSR), structure similarity, and root mean square error(RMSE). Especially, when the noises get heavier, the proposed algorithm performs better with less training time.
[1] AHARON M, ELAD M, BRUCKSTEIN A. The K-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322. [2] MAIRAL J, BACH F, PONCE J, et al. Non-local Sparse Models for Image Restoration // Proc of the 12th IEEE International Confe-rence on Computer Vision. Washington, USA: IEEE, 2009: 2272-2279. [3] DONG W S, ZHANG L, SHI G M, et al. Nonlocally Centralized Sparse Representation for Image Restoration. IEEE Transactions on Image Processing, 2012, 22(4): 1620-1630. [4] BUADES A, COLL B, MOREL J M. A Non-local Algorithm for Image Denoising // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2005, II: 60-65. [5] DAI J J, AU O C, FANG L, et al. Multichannel Nonlocal Means Fusion for Color Image Denoising. IEEE Transactions on Circuits and Systems for Video Technology, 2013, 23(11): 1873-1886. [6] DABOV K, FOI A, KATKOVNIK V, et al. Image Denoising by Sparse 3-D Transform-Domain Collaborative Filtering. IEEE Tran-sactions on Image Processing, 2007, 16(8): 2080-2095. [7] JI H, LIU C Q, SHEN Z W, et al. Robust Video Denoising Using Low Rank Matrix Completion // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2010: 1791-1798. [8] BURGER H C, SCHULER C J, HARMELING S. Image Denoising: Can Plain Neural Networks Compete with BM3D? // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2012: 2392-2399. [9] JAIN V, SEUNG H S. Natural Image Denoising with Convolutional Networks // Proc of the 22nd International Conference on Neural Information Processing Systems. New York, USA: Curran Associates Inc., 2008: 769-776. [10] XIE J Y, XU L L, CHEN E H. Image Denoising and Inpainting with Deep Neural Networks // Proc of the 26th Annual Conference on Neural Information Processing Systems. Berlin, Germany: Springer, 2012: 350-358. [11] 牛和明,杜 茜,张建勋.一种自适应全变分图像去噪算法.模式识别与人工智能, 2011, 24(6): 798-803. (NIU H M, DU Q, ZHANG J X. An Algorithm of Adaptive Total Variation Image Denoising. Pattern Recognition and Artificial Inte-lligence, 2011, 24(6): 798-803.) [12] 纪 建,许双星,李 晓.基于形态成分分析和Contourlet变换的自适应阈值图像去噪方法.模式识别与人工智能, 2014, 27(6): 561-568. (JI J, XU S X, LI X. An Adaptive Thresholding Image Denoising Method Based on Morphological Component Analysis and Contourlet Transform. Pattern Recognition and Artificial Intelligence, 2014, 27(6): 561-568.) [13] 金良海,李德华.基于噪声检测的图像去噪算法.模式识别与人工智能, 2008, 21(3): 298-302. (JIN L H, LI D H. An Image Denoising Algorithm Based on Noise Detection. Pattern Recognition and Artificial Intelligence, 2008, 21(3): 298-302.) [14] 王相海,刘晓倩,张爱迪,等.曲线拟合确定阈值的非抽取小波贝叶斯图像去噪方法.模式识别与人工智能, 2016, 26(4): 322-331. (WANG X H, LIU X Q, ZHANG A D, et al. Undecimated Wavelet Bayesian Image Denoising Method with Its Threshold Determined by Curve Fitting. Pattern Recognition and Artificial Intelligence, 2016, 26(4): 322-331.) [15] NAIR V, HINTON G E. Rectified Linear Units Improve Restricted Boltzmann Machines // Proc of the 27th International Conference on Machine Learning. Madison, USA: Omnipress, 2010: 807-814. [16] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet Cla-ssification with Deep Convolutional Neural Networks // PEREIRA F, BURGES C J C, BOTOU L, et al., eds. Advances in Neural Information Processing Systems 25. Cambridge, USA: The MIT Press, 2012: 1097-1105. [17] DONG C, LOY C C, HE K M, et al. Image Super-Resolution Using Deep Convolutional Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2): 295-307. [18] ZORAN D, WEISS Y. From Learning Models of Natural Image Patches to Whole Image Restoration // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2011: 479-486. [19] MARTIN D, FOWLKES C, TAL D, et al. A Database of Human Segmented Natural Images and Its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics. Report, UCB/CSD-1-1133. Berkeley, USA: University of California, Berkeley, 2001. [20] HE K M, SUN J. Convolutional Neural Networks at Constrained Time Cost // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2015: 5353-5360. [21] JIA Y Q, SHELHAMER E, DONAHUE J, et al. Caffe: Convolutional Architecture for Fast Feature Embedding // Proc of the 22nd ACM International Conference on Multimedia. New York, USA: ACM, 2014: 675-678.