Abstract: Deep learning has proven to be an effective mechanism for computer vision tasks, especially for image denoising and burst image denoising. In this paper, we focus on solving the burst image denoising problem and aim to generate a single clean image from a burst of noisy images. We propose to combine the power of block matching and 3D filtering (BM3D) and a convolutional neural network (CNN) for burst image denoising. In particular, we design a CNN with a divide-and-conquer strategy. First, we employ BM3D to preprocess the noisy burst images. Then, the preprocessed images and noisy images are fed separately into two parallel CNN branches. The two branches produce somewhat different results. Finally, we use a light CNN block to combine the two outputs. In addition, we improve the performance by optimizing the two branches using two different constraints: a signal constraint and a noise constraint. One maps a clean signal, and the other maps the noise distribution. In addition, we adopt block matching in the network to avoid frame misalignment. Experimental results on synthetic and real noisy images show that our algorithm is competitive with other algorithms.

基于双重约束的多帧图像降噪方法

[1]Aharon M, Elad M, Bruckstein A, 2006. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. IEEE Trans Signal Process, 54(11):4311-4322. doi: 10.1109/TSP.2006.881199

[2]Ahn B, Cho NI, 2017. Block-matching convolutional neural network for image denoising. https://arxiv.org/abs/1704.00524

[3]Buades A, Coll B, Morel JM, 2005. A non-local algorithm for image denoising. IEEE Computer Society Conf on Computer Vision and Pattern Recognition, p.60-65. doi: 10.1109/CVPR.2005.38

[4]Burger HC, Schuler CJ, Harmeling S, 2012. Image denoising: can plain neural networks compete with BM3D? IEEE Conf on Computer Vision and Pattern Recognition, p.2392-2399. doi: 10.1109/CVPR.2012.6247952

[5]Chambolle A, 2004. An algorithm for total variation minimization and applications. J Math Imag Vis, 20(1-2):89-97. doi: 10.1023/B:JMIV.0000011325.36760.1e

[6]Dabov K, Foi A, Katkovnik V, et al., 2007. Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans Image Process, 16(8):2080-2095. doi: 10.1109/TIP.2007.901238

[7]Divakar N, Babu RV, 2017. Image denoising via CNNs: an adversarial approach. Proc IEEE Conf on Computer Vision and Pattern Recognition Workshops, p.1076-1083. doi: 10.1109/CVPRW.2017.145

[8]Godard C, Matzen K, Uyttendaele M, 2018. Deep burst denoising. Proc European Conf on Computer Vision, p.560-577. doi: 10.1007/978-3-030-01267-0_33

[9]Krull A, Buchholz TO, Jug F, 2019. Noise2Void—learning denoising from single noisy images. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.2124-2132. doi: 10.1109/CVPR.2019.00223

[10]LeCun Y, Bottou L, Bengio Y, et al., 1998. Gradient-based learning applied to document recognition. Proc IEEE, 86(11):2278-2324. doi: 10.1109/5.726791

[11]Lehtinen J, Munkberg J, Hasselgren J, et al., 2018. Noise2Noise: learning image restoration without clean data. https://arxiv.org/abs/1803.04189

[12]Lempitsky V, Vedaldi A, Ulyanov D, 2018. Deep image prior. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.9446-9454. doi: 10.1109/CVPR.2018.00984

[13]Liu ZW, Yuan L, Tang XO, et al., 2014. Fast burst images denoising. ACM Trans Graph, 33(6):Article 232. doi: 10.1145/2661229.2661277

[14]Mao XJ, Shen CH, Yang YB, 2016. Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections. https://arxiv.org/abs/1603.09056v2

[15]Mildenhall B, Barron JT, Chen JW, et al., 2018. Burst denoising with kernel prediction networks. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.2502-2510. doi: 10.1109/CVPR.2018.00265

[16]Mosseri I, Zontak M, Irani M, 2013. Combining the power of internal and external denoising. IEEE Int Conf on Computational Photography, p.1-9. doi: 10.1109/ICCPhot.2013.6528298

[17]Perona P, Malik J, 1990. Scale-space and edge detection using anisotropic diffusion. IEEE Trans Patt Anal Mach Intell, 12(7):629-639. doi: 10.1109/34.56205

[18]Simonyan K, Zisserman A, 2014. Very deep convolutional networks for large-scale image recognition. https://arxiv.org/abs/1409.1556v4

[19]Tassano M, Delon J, Veit T, 2019. DVDNET: a fast network for deep video denoising. IEEE Int Conf on Image Processing, p.1805-1809. doi: 10.1109/ICIP.2019.8803136

[20]Tomasi C, Manduchi R, 1998. Bilateral filtering for gray and color images. Sixth Int Conf on Computer Vision, p.839-846. doi: 10.1109/ICCV.1998.710815

[21]Vincent P, Larochelle H, Lajoie I, et al., 2010. Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J Mach Learn Res, 11:3371-3408.

[22]Xu J, Zhang L, Zuo WM, et al., 2015. Patch group based nonlocal self-similarity prior learning for image denoising. Proc IEEE Int Conf on Computer Vision, p.244-252. doi: 10.1109/ICCV.2015.36

[23]Yang D, Sun J, 2018. BM3D-Net: a convolutional neural network for transform-domain collaborative filtering. IEEE Signal Process Lett, 25(1):55-59. doi: 10.1109/LSP.2017.2768660

[24]Zhang K, Zuo WM, Chen YJ, et al., 2017. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising. IEEE Trans Image Process, 26(7):3142-3155. doi: 10.1109/TIP.2017.2662206

[25]Zhang K, Zuo WM, Zhang L, 2018. FFDNet: toward a fast and flexible solution for CNN-based image denoising. IEEE Trans Image Process, 27(9):4608-4622. doi: 10.1109/TIP.2018.2839891