Abstract
Images are unavoidable to be corrupted by impulse noise, causing the degradation of image quality. A directional-weighted-median (DWM) filter is beneficial to restore the degraded edges. However, the DWM filter only utilizes the local properties to define the weights of the neighboring pixels for noisy image denoising. The pixel relationship among similar patches of the image is ignored, causing the restoration performance no further improvement. In this paper, we apply a deep-learning neural network (DLNN) to determine the pixel-variation direction for noisy image denoising. Because the DLNN is trained by using a noisy image and its clean one as the target image, the pixel-variation relationship of the noisy image and its clean one is considered. Thus, the performance of the DWM filter is able to be improved by the proposed DLNN DWM filter significantly. The experimental results reveal that the proposed DLNN DWM filter can effectively remove interference noise in a noisy image with noise density ranging from 10 to 90%.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Lu, C.T., Chou, T.C.: Denoising of salt-and-pepper noise corrupted image using modified directional-weighted-median filter. Pattern Recognit. Lett. 33(10), 1287–1295 (2012)
Ng, P.E., Ma, K.K.: A switching median filter with boundary discriminative noise detection for extremely corrupted images. IEEE Trans. Image Process. 15(6), 1506–1516 (2006)
Dong, Y., Xu, S.: A new directional weighted median filter for removal of random-valued impulse noise. IEEE Signal Process. Lett. 14(3), 193–196 (2007)
Zhang, X., Xiong, Y.: Impulse noise removal using directional difference based noise detector and adaptive weighted mean filter. IEEE Signal Process. Lett. 16(4), 295–298 (2009)
Toh, K.K.V., Mat-Isa, N.A.: Noise adaptive fuzzy switching median filter for salt-and-pepper noise reduction. IEEE Signal Process. Lett. 17(3), 281–284 (2010)
Subrahmanyam, G.R.K.S., Rajagopalan, A.N., Aravind, R.: Importance sampling-based unscented Kalman filter for film-grain noise removal. IEEE Multimedia 15(2), 52–63 (2008)
Wus, J., Tian, Li, J.H., Zhang, D.Q., Jiang, W., Wu, Z., Yu, H.: Toward experiential mobile media processing. IEEE Multimedia 21(2), 80–89 (2014)
Papyan, V., Elad, M.: Muti-scale patch-based image restoration. IEEE Trans. Image Process. 25(1), 249–261 (2016)
Yue, H., Sun, X., Yang, J., Wu, F.: Image denoising by exploring external and internal correlations. IEEE Trans. Image Process. 24(6), 1967–1982 (2015)
Garg, B.: Restoration of highly salt-and-pepper-noise-corrupted images using novel adaptive trimmed median filter. Signal Image Video Process. (2020)
Lu, C.T., Chen, Y.Y., Wang, L.L., Chang, C.F.: Removal of salt-and- pepper noise in corrupted image using three-values-weighted approach with variable-size window. Pattern Recognit. Lett. 80, 188–199 (2016)
Sanaee, P., Moallem, P., Razzazi, F.: An interpolation filter based on natural neighbor Galerkin method for salt and pepper noise restoration with adaptive size local filtering window. Signal Image Video Process. 13, 895–903 (2019)
Lu, C.T., Chen, M.Y., Shen, J.H., Wang, L.L., Hsu, C.C.: Removal of salt-and-pepper noise for X-ray bio-images using pixel-variation gain factors. Comput. Electr. Eng. 71, 862–876 (2018)
Tian, C., Fei, L., Zheng, W., Xu, Y., Zuo, W., Lin, C.W.: Deep learning on image denoising: an overview. Neural Networks 131, 251–275 (2020)
Yang, Q., Yan, P., Zhang, Y., Yu, H., Shi, Y., Mou, X., Kalra, M.K., Zhang, Y., Sun, L., Wang, G.: Low-dose CT image denoising using a generative adversarial network with Wasserstein distance and perceptual loss. IEEE Trans. Med. Imaging 37(6), 1348–1357 (2018)
Zhang, K., Zuo, W., Chen, Y., Meng, D., Zhang, L.: Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising. IEEE Trans. Image Process. 26(7), 3142–3155 (2017)
Liu, Z., Ya, W.Q., Yang, M.L.: Image denoising based on a CNN model, in Proc, pp. 389–3932. Int. Conf. Control, Automation Robotics (2018)
Zhang, K., Zuo, W., Zhang, L.: FFDNet: toward a fast and flexible solution for CNN-based image denoising. IEEE Trans. Image Process. 27(9), 4608–4622 (2018)
Isogawa, K., Shiodera, T., Takeguchi, T.: Deep shrinkage convolutional neural network for adaptive noise reduction. IEEE Signal Process. Lett. 25(2), 224–228 (2017)
Uddin, A.F.M.S., Chung, T., Bae, S.H.: A perceptually inspired new blind image denoising method using L1 and perceptual loss. IEEE Access 7, 90538–90549 (2019)
Lu, Z., Zhang, Y.: Sparse approximation via penalty decomposition methods. SIAM J. Optimization 23(4), 2448–2478 (2013)
Yuan, G., Ghanem, B.: l0 TV: a sparse optimization method for impulse noise image restoration. IEEE Trans. Pattern Anal. Mach. Intell. 41(2), 352–364 (2019)
Chen, R.H., Lu, C.T., Wang, L.L., Lin, C.A., Shen, J.H.: Removal of salt-and-pepper noise using convolutional-neural networks. Proc. Conf. Inform. Technology App. Outlying Islands, 590-597 (2019)
Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process. 13(4), 600–612 (2004)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Lu, CT., Hsu, HJ. & Wang, LL. Image denoising using DLNN to recognize the direction of pixel variation. SIViP 15, 1247–1256 (2021). https://doi.org/10.1007/s11760-021-01855-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-021-01855-z