Abstract
Nowadays, the resolution of image sensors in digital cameras is increased by minimizing the size of pixel sensors. As the size is reduced, the pixel sensor receives low light energy and becomes sensitive to thermal noise. The Color Filter Array (CFA) has a significant effect with the presence of noise, and the missing data is required to be reconstructed from the noisy data. This paper proposed a deep convolutional neural network with Honey Badger Algorithm (DCNN-HBA) for Bayer image de-noising. The deep CNN model is easily adopted and flexible for any CFA design with spatially varying color and exposures. After de-noising, attention-based deep residual learning (A-DRL) is applied to de-mosaicking the noise-free Bayer image. The channel attention is involved in which the network considers more relevant information and features. The proposed algorithm improves the quality of the image after reconstruction. The performance of the proposed work is evaluated with the performance metrics such as Peak Signal to Noise Ratio (PSNR), Color Peak Signal to Noise Ratio (CPSNR), Structural Similarity (SSIM), and Mean Structural Similarity (MSSIM) and compared with the traditional de-mosaicking approaches. By using our proposed work, the performance of PSNR, SSIM, CPSNR and MSSIM is improved by 43.23 dB, 0.997, 43.30 dB and 0.9975, respectively.
Similar content being viewed by others
Availability of data and materials
No data Availability.
References
Arbuj A, Jagtap S, Agrawal A, Dalpe R, Sawant P Deep Joint Denoising and Demosaicking Using Convolutional Neural Network
Buades A, Duran J (2018) Joint de-noising and demosaicking of raw video sequences. In 2018 25th IEEE international conference on image processing (ICIP) IEEE 2172-2176
Dong N, Maggioni M, Yang Y, Pérez-Pellitero E, Leonardis A and McDonagh S (n.d.) Residual Contrastive Learning for Image Reconstruction: Learning Transferable Representations from Noisy Images
Elgendy OA, Gnanasambandam A, Chan SH, Ma J (2021) Low-light demosaicking and de-noising for small pixels using learned frequency selection. IEEE Transact Comput Imag 7:137–150
Guo S, Liang Z, Zhang L (2021) Joint de-noising and demosaicking with green channel prior for real-world burst images. IEEE Trans Imag Proc 30:6930–6942
Janjušević N, Khalilian-Gourtani A, Wang Y (2021) CDLNet: noise-adaptive convolutional dictionary learning network for blind Denoising and Demosaicing. arXiv preprint arXiv:2112.00913
Jayachandran S (2017) Digital imaging in dentistry: a review. Contemporary Clin Dentis 8(2):193–194
Jin Q, Facciolo G, Morel JM (2020) A review of an old dilemma: Demosaicking first, or de-noising first? In proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops 514-515.
John Peter K, Prabhu S (2017) Demosaicking on color filter array images using improved posterior decision approach. Int J Print Packag Allied Sci 5(1):729–739
Khadidos AO, Khadidos AO, Khan FQ, Tsaramirsis G, Ahmad A (2021) Bayer image demosaicking and de-noising based on specialized networks using deep learning. Multimed Syst, Springer 27(4):807–819
Kiku D, Monno Y, Tanaka M, Okutomi M (2016) Beyond color difference: residual interpolation for color image demosaicking. IEEE Trans Imag Proc 25(3):1288–1300
Kim Y, Ryu H, Lee S, Lee YJ (2020) Joint Demosaicing and Denoising based on Interchannel nonlocal mean weighted moving least squares method. Sensors 20(17):4697
Kokkinos F, Lefkimmiatis S (2018) Deep image demosaicking using a cascade of convolutional residual de-noising networks. In proceedings of the European conference on computer vision (ECCV) 303-319
Ma K, Gharbi M, Adams A, Kamil S, Li TM, Barnes C, Ragan-Kelley J (2022) Searching for fast Demosaicking algorithms. ACM Transac Graph (TOG) 41:1–18
Mihoubi S, Lapray PJ, Bigué L (2018) Survey of demosaicking methods for polarization filter array images. Sensors 18(11):3688
Morrison AO, Gardner JM (2015) Microscopic image photography techniques of the past, present, and future. Archives Pathol Lab Med 139(12):1558–1564
Ni Z, Ma KK, Zeng H, Zhong B (2020) Color image demosaicing using progressive collaborative representation. IEEE Trans Imag Proc 29:4952–4964
Park SH, Kim HS, Lansel S, Parmar M, Wandell BA (2009) A case for de-noising before demosaicking color filter array data. In 2009 Conference record of the forty-third Asilomar conference on signals, Systems and Computers, IEEE 860–864
Qiao Z, Wen X, Zhou X, Qin F, Liu S, Gao B, Liu W, Chi D, Liu Z (2023) Adaptive iterative guided filtering for suppressing background noise in ptychographical imaging. Optics LasEngin, Elsevier 160:107233
Rafi Nazari M (2017) De-noising and demosaicking of color images. PhD diss., Université d'Ottawa/University of Ottawa
Rostami M, Berahmand K, Nasiri E, Forouzandeh S (2021) Review of swarm intelligence-based feature selection methods. Engineering Applications of Artificial Intelligence, Elsevier, 100: 104210
Syu NS, Chen YS, Chuang YY (2018) Learning deep convolutional networks for demosaicing. arXiv preprint arXiv:1802.03769
Tan DS, Chen WY, Hua KL (2018) DeepDemosaicking: adaptive image demosaicking via multiple deep fully convolutional networks. IEEE Trans Imag Proc 27(5):2408–2419
Tan H, Xiao H, Liu Y, Zhang M (2022) Two-stage CNN model for joint Demosaicing and Denoising of burst Bayer images. Computational intelligence and neuroscience, hindawi.com, 2022
Tian C, Xu Y, Zuo W (2020) Image de-noising using deep CNN with batch renormalization. Neural Netw, Elsevier 121:461–473
Wang Y, Cao R, Guan Y, Liu T, Yu Z (2021) A deep survey in the applications of demosaicking. In 2021 3rd international academic exchange conference on science and technology innovation (IAECST), IEEE 596-602.
Wang S, Zhao M, Dou R, Yu S, Liu L, Wu N (2021) A compact high-quality image Demosaicking neural network for edge-computing devices. Sensors 21(9):3265
Wu F, Huang T, Dong W, Shi G, Zheng Z, Li X (2021) Toward blind joint demosaicing and de-noising of raw color filter array data. Neurocomput Elsevier 453:369–382
Ye W, Ma KK (2015) Color image demosaicing using iterative residual interpolation. IEEE Trans Imag Proc 24(12):5879–5891
Zhang C, Li Y, Wang J, Hao P (2016) Universal demosaicking of color filter arrays. IEEE Trans Imag Proc 25(11):5173–5186
Funding
No funding is provided for the preparation of the manuscript.
Author information
Authors and Affiliations
Contributions
All authors have equal contributions to this work.
Corresponding author
Ethics declarations
Conflict of interest
Authors S.P. Predeep Kumar, K. John Peter, C. Sahaya Kingsly declare that they have no conflict of interest.
Ethical approval
This article does not contain any studies with human participants or animals performed by any authors.
Consent to participate
All the authors involved have agreed to participate in this submitted article.
Consent to publish
All the authors involved in this manuscript give full consent for publication of this submitted article.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kumar, S.P., Peter, K.J. & Kingsly, C.S. De-noising and Demosaicking of Bayer image using deep convolutional attention residual learning. Multimed Tools Appl 82, 20323–20342 (2023). https://doi.org/10.1007/s11042-023-14334-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-14334-z