{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T09:03:22Z","timestamp":1779872602929,"version":"3.53.1"},"reference-count":61,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T00:00:00Z","timestamp":1779840000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Shanxi Province Applied Basic Research Program, China","award":["202203021211116"],"award-info":[{"award-number":["202203021211116"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Deep learning has garnered widespread attention in the computer vision field, particularly for image denoising tasks. While Transformer-based models currently represent the state-of-the-art in denoising performance, this work investigates whether the architecture based on convolutional neural networks (CNN) can achieve comparable results. We propose GatedRedNet, a novel three-layer encoder-decoder network designed for image denoising. To enhance network depth while maintaining efficiency, GatedRedNet incorporates attention mechanisms within its residual blocks. Specifically, we incorporate the Efficient Channel Attention (ECA) module within all residual blocks, enabling effective recovery of edge and texture details with minimal computational overhead. To further improve global feature learning, we introduce a learnable gating factor at the encoder-decoder fusion stage, dynamically adjusting the weight ratio between their feature maps. These designs enable GatedRedNet to better capture local texture features while maintaining robust denoising capabilities. We evaluate GatedRedNet extensively on multiple benchmark datasets, demonstrating denoising performance comparable to state-of-the-art Transformer-based models and even surpassing them in certain aspects. Our results suggest that well-designed CNN architectures remain competitive in image denoising, offering an efficient alternative to Transformer.<\/jats:p>","DOI":"10.7717\/peerj-cs.3900","type":"journal-article","created":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T08:24:21Z","timestamp":1779870261000},"page":"e3900","source":"Crossref","is-referenced-by-count":0,"title":["Gated RedNet for image denoising"],"prefix":"10.7717","volume":"12","author":[{"given":"Huibin","family":"Zhang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liping","family":"Feng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jing","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fang","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guorong","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,5,27]]},"reference":[{"key":"10.7717\/peerj-cs.3900\/ref-1","first-page":"1692","article-title":"A high-quality denoising dataset for smartphone cameras","author":"Abdelhamed","year":"2018"},{"key":"10.7717\/peerj-cs.3900\/ref-2","first-page":"1122-113","article-title":"NTIRE 2017 challenge on single image super-resolution: dataset and study","author":"Agustsson","year":"2017"},{"key":"10.7717\/peerj-cs.3900\/ref-3","first-page":"3155","article-title":"Real image denoising with feature attention","author":"Anwar","year":"2019"},{"key":"10.7717\/peerj-cs.3900\/ref-4","first-page":"3285","article-title":"Attention augmented convolutional networks","author":"Bello","year":"2019"},{"key":"10.7717\/peerj-cs.3900\/ref-5","first-page":"8","article-title":"Toward fast and accurate human pose estimation via soft-gated skip connections","author":"Bulat","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-6","first-page":"213","article-title":"End-to-end object detection with transformers","author":"Carion","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-7","first-page":"4159","article-title":"Fully convolutional pixel adaptive image denoiser","author":"Cha","year":"2019"},{"key":"10.7717\/peerj-cs.3900\/ref-8","first-page":"171","article-title":"Spatialadaptive network for single image denoising","author":"Chang","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-9","doi-asserted-by":"publisher","first-page":"168","DOI":"10.1109\/icip.1994.413553","article-title":"Two deterministic half-quadratic regularization algorithms for computed imaging","volume":"2","author":"Charbonnier","year":"1994","journal-title":"International Conference on Image Processing"},{"issue":"6","key":"10.7717\/peerj-cs.3900\/ref-10","doi-asserted-by":"publisher","first-page":"1256","DOI":"10.1109\/tpami.2016.2596743","article-title":"Trainable nonlinear reaction diffusion: a flexible framework for fast and effective image restoration","volume":"39","author":"Chen","year":"2017","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"10.7717\/peerj-cs.3900\/ref-11","first-page":"12299","article-title":"Pre-trained image processing transformer","author":"Chen","year":"2021"},{"key":"10.7717\/peerj-cs.3900\/ref-12","first-page":"4894","article-title":"NBNet: noise basis learning for image denoising with subspace projection","author":"Cheng","year":"2021"},{"key":"10.7717\/peerj-cs.3900\/ref-13","first-page":"4641","article-title":"Rethinking coarse-to-fine approach in single image deblurring","author":"Cho","year":"2021"},{"issue":"8","key":"10.7717\/peerj-cs.3900\/ref-14","doi-asserted-by":"publisher","first-page":"2080","DOI":"10.1109\/tip.2007.901238","article-title":"Image denoising by sparse 3-D TransformDomain collaborative filtering","volume":"16","author":"Dabov","year":"2007","journal-title":"IEEE Transactions on Image Processing"},{"key":"10.7717\/peerj-cs.3900\/ref-15","article-title":"Kodak lossless true color image suite","author":"Franzen","year":"1999"},{"key":"10.7717\/peerj-cs.3900\/ref-16","first-page":"2862","article-title":"Weighted nuclear norm minimization with application to image denoising","author":"Gu","year":"2014"},{"key":"10.7717\/peerj-cs.3900\/ref-17","first-page":"868","article-title":"Regularize network skip connections by gating mechanisms for electron microscopy image segmentation","author":"Guo","year":"2019a"},{"key":"10.7717\/peerj-cs.3900\/ref-18","first-page":"1712","article-title":"Toward convolutional blind denoising of real photographs","author":"Guo","year":"2019b"},{"key":"10.7717\/peerj-cs.3900\/ref-19","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.7717\/peerj-cs.3900\/ref-20","first-page":"558","article-title":"Bag of tricks for image classification with convolutional neural networks","author":"He","year":"2019"},{"issue":"6","key":"10.7717\/peerj-cs.3900\/ref-21","doi-asserted-by":"publisher","first-page":"448","DOI":"10.14429\/dsj.74.19475","article-title":"Batch normalization: accelerating deep network training by reducing internal covariate shift","volume":"2015","author":"Ioffe","year":"2015","journal-title":"International Conference on Machine Learning ICML"},{"key":"10.7717\/peerj-cs.3900\/ref-22","first-page":"6054","article-title":"FOCNet: a fractional optimal control network for image denoising","author":"Jia","year":"2019"},{"key":"10.7717\/peerj-cs.3900\/ref-23","first-page":"3479","article-title":"Transfer learning from synthetic to real-noise denoising with adaptive instance normalization","author":"Kim","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-24","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2015"},{"key":"10.7717\/peerj-cs.3900\/ref-25","first-page":"1833","article-title":"SwinIR: image restoration using swin transformer","author":"Liang","year":"2021"},{"key":"10.7717\/peerj-cs.3900\/ref-26","first-page":"9992","article-title":"Swin transformer: hierarchical vision transformer using shifted windows","author":"Liu","year":"2021"},{"key":"10.7717\/peerj-cs.3900\/ref-27","first-page":"1673","article-title":"Non-local recurrent network for image restoration","author":"Liu","year":"2018"},{"key":"10.7717\/peerj-cs.3900\/ref-28","first-page":"886","article-title":"Multi-level wavelet-CNN for image restoration","author":"Liu","year":"2018"},{"key":"10.7717\/peerj-cs.3900\/ref-29","first-page":"2802","article-title":"Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections","author":"Mao","year":"2016"},{"key":"10.7717\/peerj-cs.3900\/ref-30","first-page":"416","article-title":"A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics","author":"Martin","year":"2002"},{"key":"10.7717\/peerj-cs.3900\/ref-31","first-page":"191","article-title":"Single image super-resolution via a holistic attention network","author":"Niu","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-32","first-page":"68","article-title":"Stand-alone self-attention in vision models","author":"Parmar","year":"2019"},{"issue":"12","key":"10.7717\/peerj-cs.3900\/ref-33","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1016\/j.neucom.2018.12.075","article-title":"Dilated residual networks with symmetric skip connection for image denoising","volume":"345","author":"Peng","year":"2019","journal-title":"Neurocomputing"},{"key":"10.7717\/peerj-cs.3900\/ref-34","first-page":"234","article-title":"U-Net: convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"10.7717\/peerj-cs.3900\/ref-35","first-page":"860","article-title":"Fields of experts: a framework for learning image priors","author":"Roth","year":"2005"},{"issue":"4","key":"10.7717\/peerj-cs.3900\/ref-36","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1016\/j.media.2019.01.012","article-title":"Attention gated networks: learning to leverage salient regions in medical images","volume":"53","author":"Schlemper","year":"2019","journal-title":"Medical Image Analysis"},{"issue":"1\u20132","key":"10.7717\/peerj-cs.3900\/ref-37","doi-asserted-by":"publisher","first-page":"117","DOI":"10.1016\/j.neunet.2019.12.024","article-title":"Attention-guided CNN for image denoising","volume":"124","author":"Tian","year":"2020","journal-title":"Neural Networks"},{"issue":"11","key":"10.7717\/peerj-cs.3900\/ref-38","doi-asserted-by":"publisher","first-page":"461","DOI":"10.1016\/j.neunet.2019.08.022","article-title":"Image denoising using deep CNN with batch renormalization","volume":"121","author":"Tian","year":"2020","journal-title":"Neural Networks"},{"key":"10.7717\/peerj-cs.3900\/ref-39","doi-asserted-by":"publisher","first-page":"8226","DOI":"10.1109\/tip.2020.3013166","article-title":"Deep graph-convolutional image denoising","volume":"29","author":"Valsesia","year":"2020","journal-title":"IEEE Transactions on Image Processing"},{"key":"10.7717\/peerj-cs.3900\/ref-40","first-page":"5998","article-title":"Attention is all you need","author":"Vaswani","year":"2017"},{"key":"10.7717\/peerj-cs.3900\/ref-41","first-page":"277","article-title":"A deep residual architecture for skin lesion segmentation","volume-title":"Context-Aware Operating Theaters, Computer Assisted Robotic Endoscopy, Clinical Image-Based Procedures, and Skin Image Analysis","author":"Venkatesh","year":"2018"},{"key":"10.7717\/peerj-cs.3900\/ref-42","first-page":"17662","article-title":"Uformer: a general U-shaped transformer for image restoration","author":"Wang","year":"2022"},{"key":"10.7717\/peerj-cs.3900\/ref-43","first-page":"11531","article-title":"ECA-Net: efficient channel attention for deep convolutional neural networks","author":"Wang","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-44","first-page":"2426","article-title":"Identifying recurring patterns with deep neural networks for natural image denoising","author":"Xia","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-45","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tgrs.2024.3450888","article-title":"Pan-denoising: guided hyperspectral image denoising via weighted represent coefficient total variation","volume":"62","author":"Xu","year":"2024","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3900\/ref-46","first-page":"1688","article-title":"Variational denoising network: toward blind noise modeling and removal","author":"Yue","year":"2019"},{"key":"10.7717\/peerj-cs.3900\/ref-47","first-page":"41","article-title":"Dual adversarial network: toward real-world noise removal and noise generation","author":"Yue","year":"2020"},{"key":"10.7717\/peerj-cs.3900\/ref-48","first-page":"2693","article-title":"CycleISP: real image restoration via improved data synthesis","author":"Zamir","year":"2020a"},{"key":"10.7717\/peerj-cs.3900\/ref-49","first-page":"492","article-title":"Learning enriched features for real image restoration and enhancement","author":"Zamir","year":"2020b"},{"key":"10.7717\/peerj-cs.3900\/ref-50","first-page":"14816","article-title":"Multi-stage progressive image restoration","author":"Zamir","year":"2021"},{"issue":"4","key":"10.7717\/peerj-cs.3900\/ref-51","doi-asserted-by":"publisher","first-page":"103397","DOI":"10.1016\/j.dsp.2022.103397","article-title":"Necessary conditions for convergence of CNNs and initialization of convolution kernels","volume":"123","author":"Zhang","year":"2022a","journal-title":"Digital Signal Processing"},{"key":"10.7717\/peerj-cs.3900\/ref-52","article-title":"Residual non-local attention networks for image restoration","author":"Zhang","year":"2019"},{"issue":"10","key":"10.7717\/peerj-cs.3900\/ref-53","doi-asserted-by":"publisher","first-page":"6360","DOI":"10.1109\/tpami.2021.3088914","article-title":"Plug-and-play image restoration with deep denoiser prior","volume":"44","author":"Zhang","year":"2022b","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"7","key":"10.7717\/peerj-cs.3900\/ref-54","doi-asserted-by":"publisher","first-page":"e970","DOI":"10.7717\/peerj-cs.970","article-title":"RatUNet: residual U-Net based on attention mechanism for image denoising","volume":"8","author":"Zhang","year":"2022","journal-title":"Peerj Computer Science"},{"issue":"5","key":"10.7717\/peerj-cs.3900\/ref-55","doi-asserted-by":"publisher","first-page":"749","DOI":"10.1109\/lgrs.2018.2802944","article-title":"Road extraction by deep residual U-Net","volume":"15","author":"Zhang","year":"2018","journal-title":"IEEE Geoscience and Remote Sensing Letters"},{"key":"10.7717\/peerj-cs.3900\/ref-56","first-page":"2472","article-title":"Residual dense network for image superresolution","author":"Zhang","year":"2018"},{"issue":"7","key":"10.7717\/peerj-cs.3900\/ref-57","doi-asserted-by":"publisher","first-page":"2480","DOI":"10.1109\/tpami.2020.2968521","article-title":"Residual dense network for image restoration","volume":"43","author":"Zhang","year":"2021","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"2","key":"10.7717\/peerj-cs.3900\/ref-58","doi-asserted-by":"publisher","first-page":"023016","DOI":"10.1117\/1.3600632","article-title":"Color demosaicking by local directional interpolation and nonlocal adaptive thresholding","volume":"20","author":"Zhang","year":"2011","journal-title":"Journal of Electronic Imaging"},{"issue":"7","key":"10.7717\/peerj-cs.3900\/ref-59","doi-asserted-by":"publisher","first-page":"3142","DOI":"10.1109\/tip.2017.2662206","article-title":"Beyond a gaussian denoiser: residual learning of deep CNN for image denoising","volume":"26","author":"Zhang","year":"2017a","journal-title":"IEEE Transactions on Image Processing"},{"key":"10.7717\/peerj-cs.3900\/ref-60","first-page":"2808","article-title":"Learning deep CNN denoiser prior for image restoration","author":"Zhang","year":"2017b"},{"issue":"9","key":"10.7717\/peerj-cs.3900\/ref-61","doi-asserted-by":"publisher","first-page":"4608","DOI":"10.1109\/tip.2018.2839891","article-title":"FFDNet: toward a fast and flexible solution for CNN-based image denoising","volume":"27","author":"Zhang","year":"2018","journal-title":"IEEE Transactions on Image Processing"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3900.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3900.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3900.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3900.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T08:24:23Z","timestamp":1779870263000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3900"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,27]]},"references-count":61,"alternative-id":["10.7717\/peerj-cs.3900"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3900","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,27]]},"article-number":"e3900"}}