Huang Ruji. REALIZATION OF DIRECTED FUNDAMENTAL CUTSET MATRIX BY HYPERGRAPH THEORY[J]. Journal of Electronics & Information Technology, 1992, 14(1): 50-60.
Citation:
Shaoping XU, Zhenyu LIN, Yan CUI, Ruirui LIU, Xiaohui YANG. A Dual-Channel Deep Convolutional Neural Network Model for Random-Valued Impulse Noise Removal[J]. Journal of Electronics & Information Technology, 2020, 42(10): 2541-2548. doi: 10.11999/JEIT190796
Huang Ruji. REALIZATION OF DIRECTED FUNDAMENTAL CUTSET MATRIX BY HYPERGRAPH THEORY[J]. Journal of Electronics & Information Technology, 1992, 14(1): 50-60.
Citation:
Shaoping XU, Zhenyu LIN, Yan CUI, Ruirui LIU, Xiaohui YANG. A Dual-Channel Deep Convolutional Neural Network Model for Random-Valued Impulse Noise Removal[J]. Journal of Electronics & Information Technology, 2020, 42(10): 2541-2548. doi: 10.11999/JEIT190796
A Dual-channel Denoising Convolutional Neural Network (D-DnCNN) model for the removal of Random-Valued Impulse Noise (RVIN) is proposed. To obtain the reference image quickly, several Rank-Ordered Logarithmic absolute Difference (ROLD) statistics and one edge feature statistic are first extracted from a local window to construct a RVIN-aware feature vector which can describe the central pixel of the patch is RVIN or not. Next, a noise detector based on Deep Belief Network (DBN) is trained to map the extracted feature vectors to their corresponding noise labels to detect all noise-like pixels in the observed image. Then, under the guidance of noise labels, the Delaunay triangulation-based interpolation algorithm is exploited to restore all detected noise-like pixels quickly and generate a preliminary restored image used as reference image. Finally, the reference image and the noisy image are simultaneously fed into the D-DnCNN model to output its corresponding residual image, and the final restored image can be obtained by subtracting the residual image from the noisy image. Extensive experimental results show that, the denoising effect of the proposed D-DnCNN denoising model outperforms the existing state-of-art switching ones across a range of noise ratios, and it also works better than the ordinary single-channel DnCNN model.
如图1所示,对于给定一幅噪声图像,首先以光栅扫描方式顺序提取重叠图块(overlapped patches),然后利用预训练的DBN噪声检测器判定各图块对应中心像素点的噪声标签,最后将噪声标签按照提取的逆顺序重组后即可构成二值噪声标签矩阵(Mask),0表示像素点未受噪声干扰,1表示像素点受噪声干扰。在快速获得给定噪声图像的二值噪声标签矩阵后,即可利用Delaunay三角剖分插值算法依据噪声标签矩阵对噪声图像进行快速修复。如图2示,通过Delaunay插值简单复原受40%RVIN噪声干扰的Lena图像,复原图像(即图2(d))中所含有能引起人眼不适的噪声像素点已经明显减少了很多(尽管仔细观察还有一些)。图2(d)其峰值信噪比(Peak Signal to Noise Ratio, PSNR)值可以达到29.98 dB,这表明其与原始无失真图像图2(a)基本相似,故能提供大量有效的图像细节信息,以它作为参考图像可以为下一阶段的双通道D-DnCNN降噪打下坚实的基础。
MA Jitong, QIU Tianshuang, LI Rong, et al. Dual-mode blind equalization algorithm based on Renyi entropy and fractional lower order statistics under impulsive noise[J]. Journal of Electronics &Information Technology, 2018, 40(2): 378–385. doi: 10.11999/JEIT170366
XU Shaoping, ZHANG Guizhen, LI Chongxi, et al. A fast random-valued impulse noise detection algorithm based on deep belief network[J]. Journal of Electronics &Information Technology, 2019, 41(5): 1130–1136. doi: 10.11999/JEIT180558
GARNETT R, HUEGERICH T, CHUI C, et al. A universal noise removal algorithm with an impulse detector[J]. IEEE Transactions on Image Processing, 2005, 14(11): 1747–1754. doi: 10.1109/TIP.2005.857261
DONG Yiqiu, CHAN R H, and XU Shufang. A detection statistic for random-valued impulse noise[J]. IEEE Transactions on Image Processing, 2007, 16(4): 1112–1120. doi: 10.1109/TIP.2006.891348
ROY A, SINGHA J, DEVI S S, et al. Impulse noise removal using SVM classification based fuzzy filter from gray scale images[J]. Signal Processing, 2016, 128: 262–273. doi: 10.1016/j.sigpro.2016.04.007
SOLEIMANY S and HAMGHALAM M. A novel random-valued impulse noise detector based on MLP neural network classifier[C]. 2017 Artificial Intelligence and Robotics, Qazvin, Iran, 2017: 165–169. doi: 10.1109/RIOS.2017.7956461.
ZHANG Kai, ZUO Wangmeng, CHEN Yunjin, et al. Beyond a Gaussian denoiser: Residual learning of deep CNN for image denoising[J]. IEEE Transactions on Image Processing, 2017, 26(7): 3142–3155. doi: 10.1109/TIP.2017.2662206
ZHANG Kai, ZUO Wangmeng, and ZHANG Lei. FFDNet: Toward a fast and flexible solution for CNN-based image denoising[J]. IEEE Transactions on Image Processing, 2018, 27(9): 4608–4622. doi: 10.1109/TIP.2018.2839891
GUO Shi, YAN Zifei, ZHANG Kai, et al. Toward convolutional blind denoising of real photographs[C]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: 1712–1722. doi: 10.1109/CVPR.2019.00181.
YU Hancheng, ZHAO Li, and WANG Haixian. An efficient procedure for removing random-valued impulse noise in images[J]. IEEE Signal Processing Letters, 2008, 15: 922–925. doi: 10.1109/LSP.2008.2005051
SUN Jiande, LIU Xiaocui, WAN Wenbo, et al. Video hashing based on appearance and attention features fusion via DBN[J]. Neurocomputing, 2016, 213: 84–94. doi: 10.1016/j.neucom.2016.05.098
AKKOUL S, LEDEE R, LECONGE R, et al. A new adaptive switching median filter[J]. IEEE Signal Processing Letters, 2010, 17(6): 587–590. doi: 10.1109/LSP.2010.2048646
XIONG Bo and YIN Zhouping. A universal denoising framework with a new impulse detector and nonlocal means[J]. IEEE Transactions on Image Processing, 2012, 21(4): 1663–1675. doi: 10.1109/TIP.2011.2172804
CHEN C L P, LIU Licheng, CHEN Long, et al. Weighted couple sparse representation with classified regularization for impulse noise removal[J]. IEEE Transactions on Image Processing, 2015, 24(11): 4014–4026. doi: 10.1109/TIP.2015.2456432
JIN K H and YE J C. Sparse and low-rank decomposition of a Hankel structured matrix for impulse noise removal[J]. IEEE Transactions on Image Processing, 2018, 27(3): 1448–1461. doi: 10.1109/TIP.2017.2771471
LIU Peng and FANG Ruogu. Learning pixel-distribution prior with wider convolution for image denoising[EB/OL]. https://arxiv.org/abs/1707.09135, 2017.
MAO Xiaojiao, SHEN Chunhua, and YANG Yubin. Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections[C]. The 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, 2016: 2810–2818.
HUANG Tao, DONG Weisheng, XIE Xuemei, et al. Mixed noise removal via Laplacian scale mixture modeling and nonlocal low-rank approximation[J]. IEEE Transactions on Image Processing, 2017, 26(7): 3171–3186. doi: 10.1109/TIP.2017.2676466
ARBELÁEZ P, MAIRE M, FOWLKES C, et al. Contour detection and hierarchical image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(5): 898–916. doi: 10.1109/TPAMI.2010.161
YUE Zongsheng, YONG Hongwei, ZHAO Qian, et al. Variational denoising network: Toward blind noise modeling and removal[EB/OL]. https://arxiv.org/abs/1908.11314v1, 2019.
Huang Ruji. REALIZATION OF DIRECTED FUNDAMENTAL CUTSET MATRIX BY HYPERGRAPH THEORY[J]. Journal of Electronics & Information Technology, 1992, 14(1): 50-60.
Huang Ruji. REALIZATION OF DIRECTED FUNDAMENTAL CUTSET MATRIX BY HYPERGRAPH THEORY[J]. Journal of Electronics & Information Technology, 1992, 14(1): 50-60.