1,091 Hits in 6.8 sec

A residual U-Net network with image prior for 3D image denoising

J. F. P. J. Abascal, S. Bussod, N. Ducros, S. Si-Mohamed, P. Douek, C. Chappard, F. Peyrin
2021 2020 28th European Signal Processing Conference (EUSIPCO)   unpublished
Denoising algorithms via sparse representation are among the state-of-the art for image restoration. On previous work, we proposed SPADE -a sparse-and prior-based method for 3D-image denoising.  ...  In this work, we extend this idea to learning approaches and propose a novel residual-U-Net prior-based (ResPrU-Net) method that exploits a prior image.  ...  We also remark that U-Net is a 2D denoising method and we did not compare to 3D-U-Net with 3D convolutions.  ... 
doi:10.23919/eusipco47968.2020.9287607 fatcat:kwbmvyk355f35g5fy32p5v46iu

Selective Residual M-Net for Real Image Denoising [article]

Chi-Mao Fan, Tsung-Jung Liu, Kuan-Hsien Liu
2022 arXiv   pre-print
To advance the performanceof denoising algorithms, we propose a blind real image denoising network (SRMNet) by employing a hierarchical architecture improved from U-Net.  ...  Specifically, we use a selective kernel with residual block on the hierarchical structure called M-Net to enrich the multi-scale semantic information.  ...  Traditional model-based denoising methods, such as block-matching and 3D filtering (BM3D) [1] , non-local means (NLM) [2] are all based on the information of image priors.  ... 
arXiv:2203.01645v1 fatcat:lwlyvnmivjhgvkxtl5fjeumgla

Model-Based Deep Learning PET Image Reconstruction Using Forward-Backward Splitting Expectation Maximisation

Abolfazl Mehranian, Andrew J. Reader
2020 IEEE Transactions on Radiation and Plasma Medical Sciences  
For regularization, the use of either a Bowsher prior (using Markov-random fields) or a residual learning unit (using convolutional-neural networks) were considered.  ...  In conclusion, the studied U-Net denoising method achieved a comparable performance to a representative implementation of the FBSEM net.  ...  Furthermore, our previous 2D results showed that post-reconstruction denoising using a U-Net outperforms a residual learning unit.  ... 
doi:10.1109/trpms.2020.3004408 pmid:34056150 pmcid:PMC7610859 fatcat:sddtmzu7zrhaxnnbf4dkdpw4ai

SUD: Supervision by Denoising for Medical Image Segmentation [article]

Sean I. Young, Adrian V. Dalca, Enzo Ferrante, Polina Golland, Bruce Fischl, Juan Eugenio Iglesias
2022 arXiv   pre-print
Training a fully convolutional network for semantic segmentation typically requires a large, labeled dataset with little label noise if good generalization is to be guaranteed.  ...  SUD unifies temporal ensembling and spatial denoising techniques under a spatio-temporal denoising framework and alternates denoising and network weight update in an optimization framework for semi-supervision  ...  Support for this research was provided in part by the BRAIN Initiative Cell Census Network grant U01MH117023, the National Institute for Biomedical Imaging and Bioengineering (P41EB015896, 1R01EB023281  ... 
arXiv:2202.02952v1 fatcat:zuxk2jcnnnat3jrbqxv5wutftu

Adversarial Gaussian Denoiser for Multiple-Level Image Denoising

Aamir Khan, Weidong Jin, Amir Haider, MuhibUr Rahman, Desheng Wang
2021 Sensors  
Subsequently, we proposed an adversarial Gaussian denoiser network, which uses the generative adversarial network-based adversarial learning process for image denoising tasks.  ...  This study proposes and applies a generative adversarial network-based image denoising training architecture to multiple-level Gaussian image denoising tasks.  ...  to train a denoising network for many pairs of images. • Constrained by the prior information used Methodology We proposed an image denoising training scheme by merging adversarial losses with reconstruction  ... 
doi:10.3390/s21092998 pmid:33923320 fatcat:my23mv3xsnahbovlbmgwa5mwp4

3D U-NetR: Low Dose Computed Tomography Reconstruction via Deep Learning and 3 Dimensional Convolutions [article]

Doga Gunduzalp, Batuhan Cengiz, Mehmet Ozan Unal, Isa Yildirim
2022 arXiv   pre-print
In the proposed reconstruction technique, sparse and noisy sinograms are back-projected to the image domain with FBP operation, then the denoising process is applied with a U-Net like 3-dimensional network  ...  More importantly, 3D U-NetR captures medically critical visual details that cannot be visualized by a 2D network on the reconstruction of real CT images with 1/10 of the normal dose.  ...  Network Architecture Based on the success of the 2D FBP-ConvNet [11] architecture and the 3D U-Net used for segmentation [10] , a U-Net like network is built with 3D CNNs.  ... 
arXiv:2105.14130v2 fatcat:egxxgj2ozzhabhtiiytibqjfky

Generative and discriminative model-based approaches to microscopic image restoration and segmentation

2020 Microscopy  
Convolutional neural networks (CNNs), a popular deep learning-based ML architecture, have been developed for image processing applications.  ...  However, the application of ML to microscopic images is limited as microscopic images are often 3D/4D, that is, the image sizes can be very large, and the images may suffer from serious noise generated  ...  Even when a latest GPU computing card is used, the aforementioned models take more than 1 week for training both 3D residual U-net and FFN.  ... 
doi:10.1093/jmicro/dfaa007 pmid:32215571 pmcid:PMC7141893 fatcat:cu6abtej35brpectdh7wgz5jei

Removal of speckle noises from ultrasound images using five different deep learning networks

Onur Karaoğlu, Hasan Şakir Bilge, İhsan Uluer
2021 Engineering Science and Technology, an International Journal  
D-U-Net, BatchRenormalization U-Net/Br-U-Net, Generative Adversarial Denoising Network/DGan-Net, and CNN Residual Network/DeR-Net).  ...  , a residual connection-based denoising network, and a modified generative adversarial network for denoising.  ... 
doi:10.1016/j.jestch.2021.06.010 fatcat:bbo7zhyysfcuxi7z6qvao7r4ai

Residual U-Net Convolutional Neural Network Architecture for Low-Dose CT Denoising

Mattias P. Heinrich, Maik Stille, Thorsten M. Buzug
2018 Current Directions in Biomedical Engineering  
Two different multilayer convolutional neural network (CNN) architectures for the denoising of CT images are investigated.  ...  Both architectures feature a residual connection of the input image to ease learning. Training images are based on realistic simulations by using the XCAT phantom.  ...  The assumption of a certain sparsity in natural and medi-⊕ Fig. 1 : The employed architecture of residual U-Net for CT denoising.  ... 
doi:10.1515/cdbme-2018-0072 fatcat:wopz2vgafrhjro5r5huok6pezi

Artificial Intelligence-Based Image Enhancement in PET Imaging: Noise Reduction and Resolution Enhancement [article]

Juan Liu, Masoud Malekzadeh, Niloufar Mirian, Tzu-An Song, Chi Liu, Joyita Dutta
2021 arXiv   pre-print
We present here a detailed review of recent efforts for AI-based PET image enhancement with a focus on network architectures, data types, loss functions, and evaluation metrics.  ...  AI models for image denoising and deblurring are becoming increasingly popular for post-reconstruction enhancement of PET images.  ...  In the denoising example, a 3D U-Net network was used to produce the denoised image from a 20% low-count image.  ... 
arXiv:2107.13595v1 fatcat:dethv7xhpbhjnikwbunr5e3tgi

A Noise Extraction Method for Cryo-EM Single-Particle Denoising

Huanrong Tang, Sihan Wang, Jianquan Ouyang, Tianming Liu
2022 Journal on Big Data  
In this paper, we propose a construction method of cryo-EM denoising dataset that uses U-Net to extract noise blocks from cryo-EM images, superimpose the noise block with the projected pure particles to  ...  Then we adopt a supervised generative adversarial network (GAN) with perceptual loss to train on our simulated dataset and denoise the real cryo-EM single particle.  ...  for their support in the completion of this article.  ... 
doi:10.32604/jbd.2022.028078 fatcat:3ro7zeswpvc63aypl7k4letmry

Multi-wavelet residual dense convolutional neural network for image denoising [article]

Shuo-Fei Wang, Wen-Kai Yu, Ya-Xin Li
2020 arXiv   pre-print
In this work, we utilize the short-term residual learning method to improve the performance and robustness of networks for image denoising tasks.  ...  Here, we choose a multi-wavelet convolutional neural network (MWCNN), one of the state-of-art networks with large RF, as the backbone, and insert residual dense blocks (RDBs) in its each layer.  ...  Therefore, the CNNs with U-net structure seem to be more suitable to deal with image denoising tasks compared with the traditional networks.  ... 
arXiv:2002.08301v1 fatcat:b3zfzg43vbbm5jnh7uvjbt5z7e

Applications of artificial intelligence in nuclear medicine image generation

Zhibiao Cheng, Junhai Wen, Gang Huang, Jianhua Yan
2021 Quantitative Imaging in Medicine and Surgery  
This work provides an overview of the application of AI in image generation for single-photon emission computed tomography (SPECT) and positron emission tomography (PET) either without or with anatomical  ...  It can also be used for image generation to shorten the time of image acquisition, reduce the dose of injected tracer, and enhance image quality.  ...  (60) combined a U-net structure and a residual block to form a new cycle-GAN generator in which the residual structure is important for learning.  ... 
doi:10.21037/qims-20-1078 pmid:34079744 pmcid:PMC8107336 fatcat:36vdnuatljbmzayjw3azmrtdee

Fast and Automated Hyperreflective Foci Segmentation Based on Image Enhancement and Improved 3D U-Net in SD-OCT Volumes with Diabetic Retinopathy

Sha Xie, Idowu Paul Okuwobi, Mingchao Li, Yuhan Zhang, Songtao Yuan, Qiang Chen
2020 Translational Vision Science & Technology  
Then the enhanced images were cascaded with the denoised images as the two-channel input to the network against the low-contrast HRF.  ...  Finally, we replaced the standard convolution with slice-wise dilated convolution in the last layer of the encoder path of 3D U-Net to obtain long-range information.  ...  However, our network using input 1 performs similar to 3D U-Net using input 1. The possible explanation for this might be that the information of denoised images is limited.  ... 
doi:10.1167/tvst.9.2.21 pmid:32818082 pmcid:PMC7396192 fatcat:yuu7ymwrofekbf65zyr54too7e

A Review of Deep Learning Methods for Compressed Sensing Image Reconstruction and Its Medical Applications

Yutong Xie, Quanzheng Li
2022 Electronics  
In this paper, we review recent works using deep learning method to solve CS problem for images or medical imaging reconstruction including computed tomography (CT), magnetic resonance imaging (MRI) and  ...  It also indicates that the key to solve CS problem and its medical applications is how to depict the image prior.  ...  The backbone of the generative network is a U-Net and the input is a 3d low-dose PET image.  ... 
doi:10.3390/electronics11040586 fatcat:zoz4hlue6vcanehrdvslspjbmi
« Previous Showing results 1 — 15 out of 1,091 results